Differential regulation of Fas-mediated apoptosis in both thyrocyte and lymphocyte cellular compartments correlates with opposite phenotypic manifestations of autoimmune thyroid disease.
ABSTRACT Several mechanisms are probably involved in determining the evolution of autoimmune thyroid disease (AITD) towards either hypothyroidism and the clinical syndrome known as Hashimoto's thyroiditis (HT) or toward hyperthyroidism and the symptoms of Graves' disease (GD). To gain further insight into such mechanisms we performed an exhaustive comparative analysis of the expression of key molecules regulating cell death (Fas, Fas ligand [FasL], Bcl-2) and apoptosis in both thyrocytes and thyroid infiltrating lymphocytes (TILs) from patients with either GD or HT. GD thyrocytes expressed less Fas/FasL than HT thyrocytes, whereas GD TILs had higher levels of Fas/FasL than HT TILs. GD thyrocytes expressed increased levels of the antiapoptotic molecule Bcl-2 compared to the low levels detected in HT thyrocytes. The opposite pattern was observed in GD (low Bcl-2) and HT (high Bcl-2) TILs. The patterns of apoptosis observed were consistent with the regulation of Fas, FasL, and Bcl-2 described above. Our findings suggest that in GD thyroid the regulation of Fas/FasL/Bcl2 favors apoptosis of infiltrating lymphocytes, possibly limiting their autoreactive potential and impairing their ability to mediate tissue damage. Moreover, the reduced levels of Fas/FasL and increased levels of Bcl-2 should favor thyrocyte survival and favor the thyrocyte hypertrophy associated with immunoglobulins stimulating the thyrotropin (TSH) receptor. In contrast, the regulation of Fas/FasL/Bcl2 expression in HT promotes thyrocyte apoptosis, tissue damage, and a gradual reduction in thyrocyte numbers leading to hypothyroidism. These findings help define key molecular mechanisms contributing to the clinical outcome of thyroid autoimmunity.
- Endocrine Reviews 12/1989; 10(4):537-62. · 14.87 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The revolution in molecular techniques has allowed dissection of the autoimmune response in a way impossible to imagine 10 yr ago. There have been spectacular advances in our understanding of self-tolerance mechanisms and how these may fail, combined with a detailed comprehension of antigen presentation, functional T cell subsets, and TCR utilization in autoimmunity, albeit usually in animal models that resemble, but do not exactly duplicate, human diseases. More gradually, these findings are being translated to thyroid autoimmunity, where the major achievement of the last decade has been the molecular characterization of the three main thyroid autoantigens. This in turn has allowed epitope identification, although again the only clear data so far have come from animal models of EAT. Another advance has been the recognition that the thyrocyte is not a helpless target of autoaggression, being capable of expressing a wide array of immunologically active molecules, which may exacerbate or diminish the autoimmune response. In 1983, there was considerable excitement at the discovery of the first of these phenomena, namely MHC class II expression, but its possible role in autoantigen presentation remains to be defined. By analogy with pancreatic beta-cells, and based on our own data, we believe that class II-expressing thyrocytes have little, if any, such role and suspect that instead this may be a mechanism for inducing peripheral tolerance. Defining the contribution of thyrocytes to the intrathyroidal autoimmune response, whether from released cytokines or surface-bound molecules, will be crucial to our future understanding, as well as holding the promise that these thyroid-derived products might be therapeutic targets. Despite molecular developments in HLA analysis, there have been no really major improvements in our understanding of the immunogenetics of thyroid autoimmunity, equivalent to those made in type 1 diabetes mellitus. The available data suggest strongly that non-MHC genes play an important role in susceptibility, and novel approaches will be required to identify these. On the other hand, we know more about the importance of environmental and endogenous (most probably hormonal) factors in thyroid autoimmunity. Understanding the basic immunological changes in the postpartum period is still poor, however, as most studies to date have concentrated on epidemiology and clinical delineation. As PPTD undergoes spontaneous remission, elucidation of these mechanisms has clear implications for treatment.(ABSTRACT TRUNCATED AT 400 WORDS)Endocrine Reviews 01/1995; 15(6):788-830. · 14.87 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Bcl-2 and related cytoplasmic proteins are key regulators of apoptosis, the cell suicide program critical for development, tissue homeostasis, and protection against pathogens. Those most similar to Bcl-2 promote cell survival by inhibiting adapters needed for activation of the proteases (caspases) that dismantle the cell. More distant relatives instead promote apoptosis, apparently through mechanisms that include displacing the adapters from the pro-survival proteins. Thus, for many but not all apoptotic signals, the balance between these competing activities determines cell fate. Bcl-2 family members are essential for maintenance of major organ systems, and mutations affecting them are implicated in cancer.Science 09/1998; 281(5381):1322-6. · 31.03 Impact Factor
Volume 11, Number 3, 2001
Mary Ann Liebert, Inc.
Differential Regulation of Fas-Mediated Apoptosis in Both
Thyrocyte and Lymphocyte Cellular Compartments
Correlates with Opposite Phenotypic Manifestations of
Autoimmune Thyroid Disease
Carla Giordano,1Pierina Richiusa,1Marcello Bagnasco,2Giuseppe Pizzolanti,1Francesco Di Blasi,3
Maria S. Sbriglia,1Antonina Mattina,1Gianpaola Pesce,2Paola Montagna,2Francesca Capone,1
Gabriella Misiano,1Alessandro Scorsone,1Alberto Pugliese,4and Aldo Galluzzo1
Several mechanisms are probably involved in determining the evolution of autoimmune thyroid disease (AITD)
towards either hypothyroidism and the clinical syndrome known as Hashimoto’s thyroiditis (HT) or toward
hyperthyroidism and the symptoms of Graves’ disease (GD). To gain further insight into such mechanisms we
performed an exhaustive comparative analysis of the expression of key molecules regulating cell death (Fas,
Fas ligand [FasL], Bcl-2) and apoptosis in both thyrocytes and thyroid infiltrating lymphocytes (TILs) from pa-
tients with either GD or HT. GD thyrocytes expressed less Fas/FasL than HT thyrocytes, whereas GD TILs had
higher levels of Fas/FasL than HT TILs. GD thyrocytes expressed increased levels of the antiapoptotic mole-
cule Bcl-2 compared to the low levels detected in HT thyrocytes. The opposite pattern was observed in GD
(low Bcl-2) and HT (high Bcl-2) TILs. The patterns of apoptosis observed were consistent with the regulation
of Fas, FasL, and Bcl-2 described above. Our findings suggest that in GD thyroid the regulation of Fas/FasL/Bcl2
favors apoptosis of infiltrating lymphocytes, possibly limiting their autoreactive potential and impairing their
ability to mediate tissue damage. Moreover, the reduced levels of Fas/FasL and increased levels of Bcl-2 should
favor thyrocyte survival and favor the thyrocyte hypertrophy associated with immunoglobulins stimulating
the thyrotropin (TSH) receptor. In contrast, the regulation of Fas/FasL/Bcl2 expression in HT promotes thy-
rocyte apoptosis, tissue damage, and a gradual reduction in thyrocyte numbers leading to hypothyroidism.
These findings help define key molecular mechanisms contributing to the clinical outcome of thyroid autoim-
two forms of autoimmune thyroid disease (AITD):
Hashimoto’s thyroiditis (HT) and Graves’ disease (GD) (1).
HT is characterized by lymphocytic infiltration of the thy-
roid, severe tissue damage, and hypothyroidism. In contrast,
the main features of GD are thyroid hypertrophy and hy-
perthyroidism resulting from the effects of stimulating
immunoglobulins (TSIs) against the thyrotropin (TSH) re-
ceptor and autoimmune tissue damage (2). Although au-
toimmune responses against specific antigens appear as ma-
OSS OF IMMUNOLOGICAL TOLERANCE to thyroid autoantigens
(thyroid peroxidase, thyroglobulin) is associated with
jor determinants in AITD pathogenesis, other molecular
mechanisms may play a role in determining the opposite
phenotypic outcomes of AITD. Great attention has been fo-
cused on the expression of Fas, Fas ligand (FasL), and Bcl-2,
molecules that are involved in the regulation of apoptotic
cell death (3–5). Growing evidence suggests that the abnor-
mal expression of Fas/FasL and impaired regulation of Fas-
mediated apoptosis may be critical in a number of autoim-
mune diseases, including systemic lupus erithematosus
(SLE), multiple sclerosis, type 1 diabetes, and AITD (3,6–8).
We and others have recently characterized Fas/FasL ex-
pression in thyrocytes from normal thyroids and nontoxic
goiters (NTG). Both normal and NTG thyrocytes have been
1Laboratory of Immunology, Endocrinology, Institute of Clinica Medica, University of Palermo, Italy.
2Allergology and Clinical Immunology, D.I.M.I., University of Genoa, Italy.
3Institute of Biology, CNR, Palermo, Italy.
4Immunogenetics, Diabetes Research Institute, University of Miami School of Medicine, Miami, Florida.
shown to express FasL, an observation we originally made
using an antibody that was associated with cross-reactivity
artifacts but that was later confirmed by different groups us-
ing a variety of well-characterized monoclonal antibodies
(9–15). Moreover, we reported that thyrocytes from HT
glands, in contrast to NTG and normal thyroids, hyperex-
press Fas and FasL on their surface (9). Interleukin-1b (IL-
1b), a cytokine abundantly produced in HT glands, induces
Fas expression in normal thyrocytes, and cross-linking of Fas
results in massive thyrocyte apoptosis that can be prevented
by antibodies that block Fas. Thus, the interaction of
Fas/FasL on the surface of neighboring thyrocytes may trig-
ger their “fratricidal” apoptosis and represent an important
mechanism of thyrocyte destruction in HT.
Similarly, autoimmune responses against several thyroid-
specific antigens and lymphocytic infiltration of the thyroid,
although to a lesser extent than in HT, are also observed in
GD. Although TSIs may explain thyroid hypertrophy and
the dramatic symptoms of hyperthyroidism, the presence of
autoreactive lymphocytes in GD thyroid leads to less signif-
icant tissue damage (2). The enlargement of the thyroid gland
typical of active GD can be related to the increased rates of
cell proliferation and decreased rates of cell death, processes
that are regulated by the expression of different molecules
including Fas and FasL. Thus, a different pattern of Fas/FasL
expression may characterize GD thyroids compared to HT
glands, and such differences may be responsible for the re-
duced thyroid cell apoptosis observed in GD (16). To test this
hypothesis, we purified and independently examined the
two main cellular components of the disease, thyroid infil-
trating lymphocytes (TILs) and follicular thyrocytes, from
surgical thyroid specimens obtained from patients with GD,
HT, or NTG. We evaluated the expression of Fas, FasL, Bcl-
2, and apoptosis in both thyrocytes and lymphocytes and
found evidence that differential regulation of apoptosis in
these two cellular compartments correlates with the oppo-
site phenotypic outcomes (hypothyroidism and hyperthy-
roidism) of thyroid autoimmunity.
Materials and Methods
Patients and tissue specimens
Thyroid specimens were obtained from 25 patients with
AITD (16 with GD, 9 with HT) who underwent total or par-
tial thyroidectomy at our clinic. We also obtained specimens
from 25 patients with NTG. Most patients underwent
surgery because of the presence of large goiters with tracheal
dislocation (Table 1). Three patients with HT had their thy-
roids removed because cytology examination of fine-needle
aspiration biopsy specimens suggested the presence of a thy-
roid lymphoma, but this possibility was excluded in all cases
by subsequent histological examination. All patients were
Caucasian, 7 were males, 43 were females, the age range was
21–73 years. All GD patients (n 5 16) were treated with an
antithyroid agent (methimazole) and b-adrenergic antago-
nists. Methimazole has been associated with upregulation of
FasL expression in thyrocytes both in vitroand in vivo(17,18).
The diagnosis was established on the basis of commonly ac-
cepted clinical and laboratory parameters and the typical his-
tological features associated with each condition (1). All pa-
tients were euthyroid at the time of subtotal or total
thyroidectomy. None of the specimens were obtained from
patients treated with radioactive iodine (131I), Lugol, or SSKI
because these agents may affect apoptosis in thyrocytes
(Giordano C, personal observation).
Preparation and flow-cytometry analysis of thyrocytes and
Thyrocyte isolation was performed using the collagenase
method as previously described (9). The digest was filtered
through a 200-mm mesh, and cells were plated in 75-cm2
flasks in complete medium (RPMI 1640 supplemented with
10% fetal calf serum, antibiotics, and glutamine). After
overnight adhesion, unattached cells (intrathyroidal lym-
phocytes) were collected, filtered (Falcon strainer: 100 mm,
70 mm, 40 mm) and purified by Ficoll Hypaque density gra-
dient separation to eliminate debris, red blood cells, and
dead cells. Cell viability was tested by orange acridine ex-
clusion. For lymphocyte staining, cells were incubated for 30
minutes at 4°C with anti-Fas (UB2, immunoglobulin G1
(IgG1); Kamiya Biomedical Company, Thousand Oaks, CA)
or anti-FasL (NOK-2, IgG2a; PharMingen, San Diego, CA;
H11, IgG2a, Alexis, San Diego, CA) monoclonal antibodies
(mAbs), or with isotype control IgG1 and IgG2a mAbs. Cells
were washed and treated with fluorescein isothiocyanate
conjugated (FITC) goat anti-mouse (Dako, Copenhagen, Den-
mark) or rabbit anti-rat IgG (Southern Biotechnology Asso-
ciates, Inc., Birmingham, AL). After another washing, cells
were incubated for 10 minutes with 6% species-specific nor-
mal serum and treated with saturating concentrations of
phycoerythrin (PE)-conjugated anti-CD3 (IgG1) or control
PE-IgG1 (Becton Dickinson, San Jose, CA). Alternatively,
cells were incubated for 30 minutes at 4°C with a combina-
tion of FITC anti-CD3, and the following PE mAbs: anti-CD4
(IgG1), anti-CD8 (IgG1), anti-CD19 (IgG1), anti-CD25 (IgG1),
anti-CD69 (IgG1), anti-CD14 (IgG2b), anti-CD45 (IgG1) (all
from Becton Dickinson). Intracellular Bcl-2 was detected in
permeabilized cells, as previously described (19). The anti-
Bcl-2 mAb (C2, IgG1; Santa Cruz Biotechnology, CA) was
used at the concentration of 1 mg per 106cells. Adherent cells
(thyrocytes) were removed from the flasks with trypsin eth-
ylenediaminetetraacetic acid (EDTA) (0.05% and 0.02%, re-
spectively). For immunodepletion of contaminating hema-
topoietic cells, single-cell thyrocyte suspensions were treated
with anti-CD45 (anti-Leukocyte; Becton Dickinson) for 30
minutes at 4°C. CD451cells were then removed after 30 min-
utes binding to sheep anti-mouse IgG-coupled beads (Dynal,
Wirral Merseyside, UK) and magnetic depletion (9). For thy-
rocyte staining, cells were incubated for 30 minutes at 4°C
with human serum containing anti-TPO antibodies or anti-
TPO mAb (IgG1; Biocytex, Marseille, France), or isotype-
matched control Ig. Cells were washed and labeled with PE-
conjugated goat anti-human Ig (Dako), washed again,
saturated, and labeled as already described with anti-Fas or
anti-FasL mAbs. All samples were analyzed with a FACScan
flow cytometer (Becton Dickinson).
Immunohistochemistry and double staining
Immediately after surgical removal thyroid specimens
were reduced to fragments, quick-frozen in 2-methylbutane
and stored at 280°C. Frozen sections (4 mm) were cut in a
Reichert-Jung cryotome and kept at 270°C until use. Im-
munohistochemical staining was performed according to the
GIORDANO ET AL.
streptavidin alkaline phosphatase (AP) method, as previ-
ously described, with minor modifications (20). Briefly, sec-
tions were pretreated with the avidin/biotin blocking kit
(BioGenex, San Ramon, CA) and then incubated with the ap-
propriate dilution of each mAb (30 minutes). Tissue sections
were washed in Tris-buffered saline (TBS, pH 7.6) and incu-
bated for 20 minutes at room temperature with biotinylated
anti-mouse immunoglobulins, followed by streptavidin-AP
complex (all from BioGenex). The substrate solution con-
tained basic new fucsin, naphthol As-Bi phosphate, and lev-
amisole (all from Sigma Chemical Co., St Louis, MO). Con-
trol sections were incubated with irrelevant isotype-matched
mAbs. The preparations were counterstained with Carazzi’s
hematoxylin. For double staining, streptavidin-peroxidase
and streptavidin-AP were used in combination, as previ-
ously described (20). The preparations were mounted in
APOPTOSIS PATTERN IN AITD
TABLE 1. CLINICAL CHARACTERISTICS OF THE PATIENTS STUDIED
Case #GenderAgeDiseaseTPO-Ab (U/mL) TRAb (U/L)TSH (mU/L)
Normal values: TPO-Ab , 18 U/mL; TRAb , 9 U/L; TSH 0.23–4 mU/L.
TPO-Ab, thyroid peroxidase antibodies; TRAb, Thyrotropin receptor antibodies; TSH, thyrotropin; HT, Hashimoto’s thy-
roiditis; GD, Graves’ disease; NTG, nontoxic goiter.
buffered glycerol and examined by means of a Leitz Labor-
lux microscope (Wild Microscopes, Rockleigh, NJ) equipped
Cell death or apoptosis was assessed with several meth-
ods. The TUNEL method was used to detect apoptotic cells
on frozen tissue sections (Boehringer Mannheim). Briefly,
sections were fixed in 4% paraformaldehyde for 10 minutes
and, after washing, blocking of endogenous peroxidase
(0.3% H2O2in methanol) and cell permeabilization (0.1% Tri-
ton-X100 in 0.1% sodium citrate) were performed. Sections
were covered with coverslips and incubated in a humidified
chamber for 60 minutes at 37°C with 50 mL TUNEL reaction
mixture (terminal deoxynucleotidyl-transferase from calf
thymus and nucleotide mixture buffer). After three wash-
ings, sections were incubated in a humidified chamber for
30 minutes at 37°C with 50 mL convert-POD (antifluorescein
antibody, Fab fragment from sheep, conjugated with horse-
radish peroxidase). DAB was used as substrate. Sections
were counterstained with Carazzi’s hematoxylin, mounted
in buffered glycerol, and examined with a Leitz Laborlux mi-
croscope (Wild Microscopes, Rockleigh, NJ). For cultured
cells, the percentage of hypodiploid nuclei was determined
by hypotonic fluorochrome staining by flow-cytometry ac-
cording to Nicoletti’s method (21).
RNA extraction and cDNA synthesis
Total RNA was extracted from purified thyrocytes using
TRIzol reagent (Life Technologies, Paisley, UK) according to
the manufacturer’s instructions. RNA was treated with
RNAse-free DNAse I and further purified using the S.N.A.P.
kit (Invitrogen, Carlsbad, CA). Two micrograms of total RNA
and 1 mL of oligo(dT)15 (final concentration 1 mM; Perkin
Elmer, Foster City, CA) were heated to 65°C for 5 minutes fol-
lowed by cooling on ice. cDNA synthesis was performed by
adding 100 units of Moloney murine leukemia virus (MMLV)
reverse transcriptase (Amersham, Buckinghamshire, UK), 40U
RNAsin (Amersham), 1 mM dNTPs (Amersham) in a final
volume of 20 mL, and the reaction was allowed to proceed at
25°C for 10 minutes and then at 37°C for 60 minutes. Heating
at 95°C for 5 minutes terminated the reaction.
Semiquantitative multiplex RT-PCR for Fas, FasL and
We chose a multiplex approach to analyze the relative ex-
pression levels of Fas gene compared to b-actin, as already
described by Otsuki et al. (22). Primer sequences were pre-
viously described (22). Each reverse transcription-poly-
merase chain reaction (RT-PCR) contained 1 mL of cDNA
(equivalent to a cDNA amount from 100 ng of initial total
RNA), 200 mM of each deoxynucleotide triphosphates, 1.5
mM of MgCl2, 13 TaqGold buffer (Perkin-Elmer), 200 nM of
Fas primers and 20 nM of actin primers, 1.25 units of Taq-
Gold in a final volume of 50 mL. After an initial period of
10 minutes at 95°C, 29 cycles of denaturation (95°C for 1
minute), annealing (57°C for 1 minute) and extension (72°C
for 1 minute) were performed on a MasterCycler (Eppen-
dorf, Milan, Italy) followed by a final extension of 7 minutes
at 72°C. Our choice of 29 PCR cycles allowed analyzing re-
sults in the exponential phase of the amplification curve.
Products were electrophoresed on 2% agarose gel and the
bands corresponding to Fas (366 bp), FasL (294 bp), and b-
actin (661 bp) visualized by ethidium bromide staining. Gels
were photographed using 667 Polaroid film, and pictures
scanned using a Epson GT5000 scanner.
Purified thyrocytes or lymphocytes were washed twice
with Ca21- and Mg21-free PBS buffer. Cell extracts were pre-
pared by lysing 107cells in 100 mL of modified RIPA buffer
containing 50 mM Tris-HCl pH 7.4, 1% NP-40, 0.25% sodium
deoxycholate, 150 mM sodium chloride, 1 mM EDTA, 1 mM
sodium ortovanadate, 1 mM sodium fluoride, 1 mM phenyl-
methylsulfonyl fluoride and aprotinin, leupeptin, pepstatin
A, E-64 1 mg/mL of each (all from Sigma), at 4°C. Insoluble
materials were removed by centrifugation for 20 minutes at
14,000 rpm at 4°C. Protein content was determined using the
Bradford method (Bio-Rad protein assay kit, Hercules, CA).
Protein samples (10 or 20 mg of protein) were boiled for 5
minutes in a sodium dodecyl sulfate (SDS) buffer and sepa-
rated in 12% SDS-polyacrylamide denaturing gels (SDS-
PAGE) according to Laemmli (23). Proteins were elec-
troblotted to nitrocellulose membranes (Bio-Rad). The
loading and transfer of equal amounts of protein was con-
firmed by staining the membrane with Ponceau S. The blots
were blocked by incubation in PBS with 0.05% Tween 20
(PBST) and 5% nonfat dry milk at 4°C overnight. After a brief
rinse, blots were incubated for 1 hour at room temperature
or overnight at 4°C in PBST-5% milk with one of the fol-
lowing primary antibodies: rabbit anti-Fas polyclonal anti-
body (Fas-Ab1, Calbiochem, La Jolla, CA) at 0.5 mg/mL; rab-
bit anti-FasL polyclonal antibody (FasL C-178, Santa Cruz)
at 1 mg/mL; mAb for Bcl-2 (Bcl2 C-2, Santa Cruz) at 1 mg/mL.
Blots were washed four times with PBST, incubated with an-
tirabbit or antimouse IgG-peroxidase conjugate (Santa Cruz)
(1:4000) in PBST-5% milk for 1 hour at room temperature,
washed again, and developed with a enhanced chemilumi-
nescence reagent (Pierce, Rockford, IL).
Quantitative analysis of mRNA
Fas and FasL mRNA was measured by S1 quantitative
analysis as described (24). In brief, two oligonucleotides of 77
bases with the first 72 bases complementary to the Fas anti-
gen mRNA and 65 bases with the first 60 bases complemen-
tary to FasL mRNA (Fas oligo 59-CCC CAA GTT AGA TCT
GGA TCC TTC CTC TTT GCA CTT GGT GTT GCT GGT GAG
TGT GCA TTC CTT GAT GAT TCC GCA CC-39; Fas ligand
oligo 59-ACC CCG GAA GTA TAC TTT GGA ATA TAC AAA
GTA CAG CCC AGT TTC ATT GAT CAC AAG GCC CAA
AG-39) were radiolabeled with T4 polynucleotide kinase
(Amersham). The Fas probe was hybridized with 15 mg of to-
tal RNA extracted from thyrocytes, intrathyroidal lympho-
cytes, or 5 mg from HeLa (positive control). The FasL probe
was hybridized with 40 mg of total RNA extracted from thy-
rocytes or 10 mg from phorbol hyristate acetate (PMA)-acti-
vated Jurkat cells (positive control). Both Fas and FasL probes
were then digested with 100 units of S1 nuclease (Roche Mo-
lecular System, Mannheim, Germany). Samples were resolved
on 6% denaturing polyacrylamide gels and subject to autora-
diography. The loading of equal amounts of RNA was con-
firmed by Northern Blot analysis using a radiolabeled probe
specific for human b-actin (data not shown).
GIORDANO ET AL.
Autoradiograms from S1 RNA analysis and Western blots
were scanned and the intensity of the bands quantified in
densitometric units (DU) using a Image Quant Software
(Molecular Dynamics). Results are expressed as ratio of the
intensity of the band of interest over the intensity of the ap-
propriate controls. Data are expressed as mean values 6 SD.
Statistical significance was evaluated using the Student’s
t test for independent samples. Multiple comparisons be-
tween groups were done by analysis of variance (ANOVA).
Differences between groups with p , 0.05 were considered
Fas expression in thyrocytes and TILs
Fas mRNA expression was evaluated by quantitative S1
analysis in both thyrocytes and TILs isolated from GD and
HT thyroids (Fig. 1A). The levels of Fas transcript in GD thy-
rocytes were approximately half than in HT thyrocytes
(Table 2) (p , 0.001). Similar findings were obtained by semi-
quantitative RT-PCR (Fig. 2A), with Fas mRNA being clearly
more abundantly expressed in HT (mean Fas/actin ratio
0.99 6 0.2) compared to GD (mean Fas/actin ratio 0.55 6 0.3,
p , 0.001) and NTG thyrocytes (mean Fas/actin ratio 0.42 6
0.22). Western blot analysis (Fig. 2B) confirmed that the dif-
ferences in Fas mRNA expression in purified thyrocytes
translated into similar differences at the protein level. In-
deed, Fas protein was highly expressed in all HT thyrocyte
samples (p , 0.001 vs. GD and NTG) while considerably re-
duced Fas protein levels characterize GD thyrocytes. Only 6
of 18 NTG thyroids exhibited trace amounts of Fas protein
We also analyzed and compared Fas expression in TILs
from HT and GD glands (Table 2). As shown in Figure 2B,
GD TILs expressed Fas protein at much higher level than HT
lymphocytes (p , 0.001). Similar findings were obtained at
the RNA level (Fig. 1A). By immunohistochemistry, Fas ex-
pression was mostly visible in the areas of disrupted folli-
cles in HT glands (Fig. 3B). Fas protein was not detected in
cytokeratin-positive follicular cells in GD thyroids (Fig. 3D)
but it was seen near largely infiltrated areas (Fig. 7A). All
specimens were also analyzed by flow cytometry as shown
in Fig. 4A. Thyrocytes purified from NTG thyroids showed
negligible Fas surface expression, similar to thyrocytes puri-
APOPTOSIS PATTERN IN AITD
rocytes were immunodepleted from CD451cells as described in Methods. TILs were isolated and purified through deple-
tion of adherent cells and examined by flow cytometry (. 99% as CD31cells). RNA (15 mg) was analyzed from Graves’
disease (GD) thyrocytes (n 5 5, lanes 2–4 and 6–7; from #14, #18, #19, #23, #25 human glands; see Table 1), GD TILs (n 5
1, lane 5, from #23 human gland), Hashimoto’s thyroiditis (HT) thyrocytes (n 5 4, lanes 8–11, from #2, #4, #6, #7 human
glands) and TILs (n 5 1, lane 12, from #2 human gland). No S1 nuclease (undigested probe) was added to the radioactive
probe and a 50-fold lower amount of probe was loaded (lane 1). Positive (5 mg HeLa RNA) and negative (100 mg of Es-
cherichia coli tRNA) controls are shown in lanes 13 and 14, respectively. B: FasL mRNA expression in thyrocytes (THY). A
quantitative S1 analysis of FasL mRNA expression in hematopoietic cell-depleted thyrocytes is shown. Data from GD thy-
rocytes (n 5 5, lanes 2–6, from #12, #14, #20, #23, and #24 human glands; see Table 1), HT thyrocytes (n 5 1, lane 7, from
#2 human gland) and nontoxic goiter (NTG) thyrocytes (n 5 1, lane 8, from #46 human gland) are shown as representative
of three independent experiments. RNA extracted from PMA-activated Jurkat cells was included as positive control (lane
9), while a negative control is shown in lane 10. No nuclease S1 (undigested probe) was added to the radioactive probe and
a 50-fold lower amount of probe was loaded (lane 1).
A: Fas mRNA expression in thyrocytes (THY) and infiltrating lymphocytes (TILs) by quantitative S1 analysis. Thy-
GIORDANO ET AL.
TABLE 2. MEAN VALUES 6 SD OF FAS AND FASL EXPRESSION IN THE DISEASE SPECIMENS
AVAILABLE FOR EACH ANALYTICAL METHOD
S1 analysis Western blot
(Fas/actin ratio) (D.U.) (Fas/actin ratio) D.U.)
CasesFas1THYFas1TIL Fas1THY Fas1TIL Fas1THYFas1TILFas1THYFas1TIL
HT 60 6 8.4*
(n 5 9)
8.5 6 2.1*
(n 5 12)
7.0 6 2.8*
(n 5 22)
95 6 12*
(n 5 6)
47 6 7.6 *
(n 5 8)
0.66 6 0.05*
(n 5 5)
0.29 6 0.03*
(n 5 7)
0.14 6 0.02*
(n 5 10)
0.60 6 0.02*
(n 5 2)
0.80 6 0.02*
(n 5 3)
8550 6 176*
(n 5 8)
1220 6 714*
(n 5 13)
800 6 523*
(n 5 18)
3630 6 612*
(n 5 8)
9350 6 412*
(n 5 8)
0.99 6 0.2*
(n 5 8)
0.55 6 0.3*
(n 5 12)
0.42 6 0.2*
(n 5 16)
HT78 6 18*
(n 5 8)
16 6 5.7*
(n 5 14)
14 6 4.8*
(n 5 20)
20.8 6 8****
(n 5 5)
65 6 15*
(n 5 8)
0.60 6 0.05*
(n 5 7)
0.41 6 0.06*
(n 5 8)
0.45 6 0.02*
(n 5 18)
7759 6 235*
(n 5 8)
3053 6 465*
(n 5 12)
2875 6 343*
(n 5 18)
3132 6 445*
(n 5 7)
5777 6 428*
(n 5 8)
0.89 6 7.2*
(n 5 8)
0.47 6 6.8*
(n 5 12)
0.32 6 9.2*
(n 5 16)
*p , 0.001 vs. HT.
Numbers in parentheses show the number of AITD glands examined.
TIL, thyroid infiltrating lymphocytes; HT, Hashimoto’s thyroiditis; GD, Graves’ disease; NTG, nontoxic goiter.
For the data expression, see Methods Section.
mRNA expression in hematopoietic cell-depleted thyrocytes from nontoxic goiter (NTG) (n 5 2, lanes 1–2 from #28 and #42;
see Table 1), Graves’ disease (GD) thyroids (n 5 4, lanes 3–6, from #15, #17, #20, #24 human glands) and Hashimoto’s thy-
roiditis (HT) glands (n 5 4, lanes 7–10, from #1, #4, #8, #9, human glands). Lanes 11 and 12 show positive controls (HeLa
and spleen). B: Fas and FasL protein expression by Western blot. Fas and FasL proteins were compared in hematopoietic
cell-depleted thyrocytes from NTG (lane 1, from #31 human gland; see Table 1), GD (lane 3, from #18 human gland) and
HT (lane 5, from #5 human gland) specimens, as well as in infiltrating lymphocytes from GD (lane 2, from #18 human
gland) and HT (lane 4, from #5 human gland) glands.
A: Fas mRNA expression in thyrocytes. Reverse transcription-polymerase chain reaction (RT-PCR) analysis of Fas
fied from GD glands (Table 2, p 5 NS). In contrast, increased
proportions of thyrocytes purified from HT glands expressed
Fas (p , 0.001). Flow cytometry analysis of TILs (Fig. 4B)
showed that most CD31lymphocytes from GD glands
(90%–99%) expressed Fas compared to approximately 50%
of the lymphocytes from HT thyroids (40%–53.5%). Of note,
TILs (CD31) were equally represented in both GD and HT
samples when they were examined by gating in flow cy-
tometry (.90%), confirming that with our method of purifi-
cation cell populations were similarly enriched. Fas was sim-
ilarly expressed in CD41and CD81lymphocytes purified
from both GD and HT thyroids (data not shown). Only mod-
est proportions of B cells (CD191) expressed Fas (mean 5
15% 6 7%) in both GD and HT glands (data not shown). Col-
lectively, the data presented here indicate that Fas expres-
sion is reduced in purified thyrocytes from GD glands com-
pared to HT glands. In contrast, Fas expression is increased
in TILs isolated from GD glands compared to those isolated
from HT thyroids.
FasL expression in thyrocytes and infiltrating lymphocytes
As shown in Figure 1B, FasL mRNA expression was higher
in HT compared to NTG (p , 0.001 vs. HT) and GD thyro-
cytes (p , 0.001 vs. HT; p 5 NS vs. NTG). Similar results
were obtained by Western blot (Fig. 2B). FasL protein was
highly expressed in all HT thyrocyte samples (p , 0.001 vs.
GD and NTG) (Table 2). Although to a lesser extent, all GD
and NTG thyroids examined exhibited FasL protein and
mRNA (Table 2). As shown in Figure 4A, flow cytometry
analysis of TPO-positive GD thyrocytes demonstrated that
10%–20% expressed FasL, similarly to NTG thyrocytes
(range, 13%–25%). A higher proportion of thyrocytes ex-
pressed FasL in HT glands (63%–92%), as previously re-
ported (9). The highest expression of FasL was found in TILs
(CD31) from GD glands (p , 0.001) whereas FasL expression
was much lower in HT lymphocytes (Fig. 4B). The highest
FasL protein in GD TILs was confirmed by Western blot (Fig.
2B, p , 0.001).
Bcl-2 expression in thyrocytes and lymphocytes
We also evaluated the expression of Bcl-2, a known anti-
apoptotic molecule. As shown in Figure 5A, Western blot
analysis revealed that Bcl-2 was highly expressed in GD thy-
rocytes (mean densitometric units (DU) 6 SD 5 9356 6 725)
at levels that were much higher than in NTG specimens
(6225 6 482). In contrast, Bcl-2 was almost absent in HT thy-
rocytes (16296 212, p , 0.001 vs. GD and NTG). Remarkable
differences in Bcl-2 expression were also seen in TILs (Fig. 5A).
The Bcl-2 levels found in HT lymphocytes (7220 6 678) were
higher than the reduced levels observed in GD lymphocytes
(3276 175, p , 0.001 vs. HT). By flow cytometry (Fig. 5B), HT
lymphocytes expressed high Bcl-2 levels compared to GD lym-
phocytes (mean values 70% 6 16% vs. 6.2% 6 7.8%, respec-
tively, p , 0.001). Similar findings were obtained for Bcl-2
mRNA levels by semiquantitative RT-PCR (data not shown).
Apoptosis in thyrocytes and infiltrating lymphocytes
Low levels of apoptosis were detected in PI-stained GD
thyrocytes by flow cytometry (mean 8% 6 2.6%) (Fig. 6A).
In contrast, high levels of thyrocyte apoptosis were found in
HT glands (mean 56% 6 10%), as previously reported (Fig.
APOPTOSIS PATTERN IN AITD
(A and B) and Graves’ disease (GD) (C and D) with negative control (A and C) and anti-Fas and anticytokeratin mono-
clonal antibodies (B and D). B: Fas staining (arrow) in HT tissue [(from #1 thyroid gland: positivity (red)] appears to be ho-
mogeneous in follicular structures; D: Fas staining (arrow) in GD tissue (from #24 human gland): follicular structures are
positive only for cytokeratin (brown). Magnification: 503.
Fas expression in tissue sections. Immunostaining of thyroid cryostat sections from Hashimoto’s thyroiditis (HT)
6B) (9). However, the opposite was observed for TILs, with
higher proportions of lymphocytes being apoptotic in GD
(mean 60.3% 6 18.4%) compared to HT glands (mean
15.6%6 6.2%, p , 0.001) by flow cytometry analysis per-
formed immediately after purification (Fig. 6C and 6D).
TUNEL staining of thyroid sections confirmed that TILs but
not thyrocytes were undergoing apoptosis in GD glands (Fig.
7). CD31lymphocytes expressed Fas in markedly infiltrated
areas (Fig. 7A) and were intensely apoptotic (Fig. 7B). Dou-
ble-staining for apoptosis (TUNEL) and cytokeratin showed
that apoptosis was predominantly localized in cytokeratin-
negative cells and in nonfollicular areas (Fig. 7C and 7D),
confirming that apoptosis in GD mainly affects TILs. Thus,
apoptosis in GD is markedly different than in HT, where it
dramatically affects thyrocytes and to some degree also in-
volves TILs (9).
Abundant literature shows that variation in Fas/FasL ex-
pression levels may critically alter apoptosis, a mechanism
of cell death that appears to play a critical role in several au-
GIORDANO ET AL.
toxic goiter (NTG), Graves’ disease (GD), and Hashimoto’s thyroiditis (HT) glands were purified and depleted of contam-
inating hematopoietic cells as described in Methods. Cells were labeled with control serum or control isotype-matched mon-
oclonal antibody (mAb) (immunoglobulin G1 [IgG1]) (dotted lines), anti-thyroid peroxidase (TPO) mAb (left, solid lines),
anti-Fas mAb (middle, solid lines), and FasL mAb (right, solid lines). Cells were gated on thyrocyte physical parameters
and TPO positivity (upper panels). Numbers shown indicate the percentage of positive cells. One representative experi-
ment is shown (from #33, #12, and #1 human glands, see Table 1). B: Flow cytometry analysis of Fas and FasL expression
in thyroid infiltrating lymphocytes (TILs). TILs were isolated from GD and HT specimens. Fas and FasL expression in elec-
tronically gated CD31cells is shown (left). Cells were labeled with isotype-matched antibodies (dotted lines) or anti-Fas
mAb and anti-FasL mAb (solid lines). Numbers shown indicate the percentage of positive cells. One representative exper-
iment is shown (from #7 and #18 human glands; see Table 1). Numbers shown indicate the percentage of positive cells.
A: Flow cytometry analysis of Fas and FasL expression in purified thyrocytes. Thyroid follicular cells from non-
APOPTOSIS PATTERN IN AITD
from #48 human gland; see Table 1), Hashimoto’s thyroiditis (HT) (lane 3, from #4 human gland), and Graves’ disease (lane
5, from #23 human gland) specimens. Lanes 2 and 4 show Bcl-2 protein expression in thyroid infiltrating lymphocytes (TILs)
isolated from HT (from #4) and GD (from #23) glands, respectively. Data from one of five experiments are shown. B: Flow
cytometry analysis of infiltrating lymphocytes from GD and HT glands. Bcl-2 expression was measured in permeabilized
and electronically gated CD31cells. Cells were labelled with isotype-matched antibody (dotted lines), or with anti-Bcl-2
monoclonal antibody (mAb) (solid lines). Numbers shown indicate the percentage of positive cells. One representative ex-
periment is shown from #23 and #4 human glands.
Bcl-2 expression. A: Western blot in hematopoietic cell-depleted thyrocytes from nontoxic goiter (NTG) (lane 1,
see Table 1) and Hashimoto’s thyroiditis (HT) glands (B, from #8 human gland) were PI-stained and analyzed by flow cy-
tometry for their DNA content in nuclei from purified thyrocytes cultured for 48 hours. Detection of apoptosis in infiltrat-
ing lymphocytes: PI-stained, purified lymphocytes from GD (C, from #15 human gland) and HT glands (D, from #8 hu-
man gland) were analyzed by flow cytometry for their DNA content of nuclei. Numbers shown indicate the proportion of
apoptotic cells. One representative experiment is shown.
Detection of apoptosis in thyrocytes: Purified thyrocytes from Graves’ disease (GD) (A, from #15 human gland;
toimmune diseases, including AITD (3,16). Thus, it is con-
ceivable that differences in the expression of Fas and FasL
may contribute to the heterogeneous phenotypes observed
in AITD and may play a role in determining the opposite
clinical syndromes (hypothyroidism and hyperthyroidism)
observed in AITD. Although several studies have evaluated
Fas/FasL expression and apoptosis in AITD (9,11,25–28), the
pattern of expression in the two relevant cell populations,
thyrocytes and TILs, has not been fully characterized. In this
study we report an extensive evaluation of Fas/FasL/Bcl-2
expression and apoptosis in GD and HT thyroid specimens.
The purification techniques utilized in this study allowed us
to separately examine both thyrocytes and TILs. To our
knowledge, there has not yet been such an analysis of both
cellular compartments from both HT and GD glands in the
same study, and this is the first time that Fas/FasL/Bcl-2 ex-
pression and apoptosis are simultaneously evaluated in TILs
and thyrocytes purified from GD glands.
The comparison between GD and HT specimens revealed
significant differences in the expression of Fas, FasL, Bcl-2,
and in the occurrence of apoptosis. Thyrocytes from a rela-
tively large series of GD patients were found to express sig-
nificantly less Fas than HT thyrocytes, suggesting that an an-
tiapoptotic milieu predominates in GD thyroids. This
interpretation is also supported by the higher levels of the
antiapoptotic molecule Bcl-2 that we found expressed by GD
thyrocytes compared to HT thyrocytes. Consistent with our
findings, increased Bcl-2 expression has also been reported
in GD thyrocytes by Hiromatsu et al. (15). The patterns of
apoptosis observed in GD and HT thyrocytes also fit well
with this interpretation, because significant levels of thyro-
cyte apoptosis were detected in HT but not GD glands (Figs.
6 and 7). Thus, the constitutive expression of Bcl-2 and the
low expression of Fas observed in both GD and NTG thy-
roids may help explain why GD thyrocytes are resistant to
apoptosis and is consistent with a report suggesting that Fas-
mediated apoptosis is normally blocked in thyrocytes (15).
In all the cases examined, GD thyrocytes expressed FasL
at levels that were similar to those detected in normal thy-
roids and NTG specimens. Similarly, comparable levels of
FasL were recently reported in GD thyrocytes by Hiromatsu
et al. (15) and a recent report by Mitsiades et al. (17) suggests
that FasL expression is upregulated in methimazole-treated
GD thyrocytes. The expression of FasL by thyrocytes may
help explaining the lack of tissue damage observed in GD
thyroids, since FasL expression could enable thyrocytes to
protect themselves from autoreactive lymphocytes. This hy-
pothesis is consistent with recent in vitro studies demon-
strating the ability of FasL-positive GD thyrocytes to induce
Fas-mediated apoptosis of lymphocytes (17). Unlike HT lym-
phocytes, GD lymphocytes expressed high levels of Fas and
low levels of Bcl-2, and predictably should be more prone to
apoptosis. Indeed, significant proportions of GD lympho-
cytes were found to be apoptotic in this study (Figs. 6 and
7). Moreover, we have previously shown that FasL expressed
on human thyrocytes is functional and capable of mediating
apoptosis of Fas-positive lymphoid cells in vitro (9), and thy-
rocytes expressing FasL can reportedly induce apoptosis of
infiltrating lymphocytes in experimental autoimmune thy-
roiditis (29,30). Thus, our findings are consistent with the hy-
pothesis that GD thyrocytes expressing FasL may induce
apoptosis of Fas-positive TILs and lend support to the con-
GIORDANO ET AL.
man gland; see Table 1) were stained for Fas and CD3 (A) or CD3 and apoptosis (TUNEL staining, B). A: CD3 and Fas
staining shows numerous double-stained lymphocytes (red and brown, and fusion amaranth color, see arrow). B: Almost
all CD3-positive cells (see arrow) are also TUNEL-positive, i.e., apoptotic. C: Lack of colocalization of apoptosis (TUNEL
staining, brown) with cytokeratin (red) in GD sections: cytokeratin-positive follicular cells are TUNEL-negative (see black
arrow) while cytokeratin-negative infiltrating cells are apoptotic (TUNEL-positive, see white arrow). D: Magnification 503.
Arrow shows the TUNEL negativity of thyrocytes.
Fas expression and Apoptosis in Graves’ disease (GD) thyroids. Tissue sections from GD specimens (from #18 hu-
cept that FasL expression may confer immune privilege to
thyrocytes, similar to the eye and testis (9,31–33). Moreover,
this mechanism may help explain the less destructive form
of autoimmunity observed in GD thyroids.
However, an imbalance in the expression of Fas and FasL
can turn a protective mechanism into a potentially damag-
ing one, and we have recently produced evidence that such
a mechanism may play a key role in the development of HT
(9). The simultaneous expression of both Fas and FasL on
the surface of neighboring thyrocytes appears to trigger frat-
ricidal apoptosis of thyrocytes, and this may represent an
important mechanism of thyrocyte destruction in HT (33).
Thus, the severe tissue damage associated with HT can be
seen as the consequence of abnormal Fas/FasL expression
and reduced Bcl-2 levels in thyrocytes, both of which may
be induced by proapoptotic cytokines secreted by infiltrat-
ing mononuclear cells. The observation that lymphocytes ex-
press low FasL levels suggests that they may be unable to
mediate significant tissue damage and that further supports
the concept that thyrocytes may be the main mediators of
their own demise.
An important implication of this and other related stud-
ies is that we may have to modify our perception of the tar-
get organ in organ-specific autoimmunity. The data dis-
cussed here suggest that the phenotypic outcomes of AITD
may be significantly affected by the thyrocyte response to
the stimuli provided by the infiltrating lymphocytes (34).
Similar observations have recently been made in type 1 di-
abetes, another organ-specific autoimmune disorder in
which pancreatic b-cells are destroyed by autoreactive lym-
phocytes infiltrating the endocrine pancreas (6). Both a be-
nign, nondestructive, and a malignant form of pancreatic au-
toimmunity have been described (35), and a recent study has
suggested that the switch of the disease process from the be-
nign to the malignant state does indeed require a response
from the target cells. In animal models of diabetes, such a
response is mediated through the stimulation of the tumor
necrosis factor receptor and, similar to what we describe here
for the thyroid, appears to involve changes in apoptosis of
pancreatic b-cells (6,36). In conclusion, this study illustrates
critical differences in Fas/FasL expression and apoptosis that
may influence key molecular mechanisms modulating the
phenotypic outcome of AITD in man and contribute to the
pathogenesis of thyroid autoimmunity.
We thank Drs. G. Modica, A. Lo Monte, and M. Romano,
Department of Surgery, University of Palermo, Italy, for pro-
viding the surgical specimens used in this study. This work
was supported by grants from MURST 98 (A.G.) and JDFI#
1. De Groot LJ, Quintans J 1989 The causes of autoimmune thy-
roid disease. Endocr Rev 10:537–562.
2. Weetman AP, McGregor AM 1994 Autoimmune thyroid dis-
ease: Further developments in our understanding. Endocr
3. Thompson CB 1995 Apoptosis in the pathogenesis and treat-
ment of disease. Science 267:1456–1462.
4. Nagata S, Golstein P 1995 The Fas death factor. Science
5. Adams JM, Cory S 1998 The Bcl-2 protein family: Arbiters
of cell survival. Science 281:1322–1325.
6. Stassi G, De Maria R, Trucco G, Rudert W, Testi R, Galluzzo
A, Giordano C, Trucco M 1997 Nitric oxide primes pancre-
atic b-cells for Fas-mediated destruction in insulin-depen-
dent diabetes mellitus. J Exp Med 186:1193–1200.
7. French LE, Tschopp J 1997 Thyroiditis and hepatitis: Fas on
the road to disease. Nat Med 3:387–388.
8. Ohsako S, Elkon KB 1999 Apoptosis in the effector phase of
autoimmune diabetes, multiple sclerosis and thyroiditis.
Cell Death Differ 6:13–21.
9. Giordano C, Stassi G, De Maria R, Todaro M, Richiusa P,
Papoff G, Ruberti G, Bagnasco M, Testi R, Galluzzo A 1997
Potential involvement of Fas and its ligand in the patho-
genesis of Hashimoto’s thyroiditis. Science 275:960–963.
10. Papoff G, Stassi G, De Maria R, Giordano C, Galluzzo A,
Bagnasco M, Ruberti G, Testi R 1998 Constitutive expression
of FasL in thyrocytes (technical comments). Science
11. Smith D, Sieg S, Kaplan D 1998 Technical note: Aberrant de-
tection of cell surface Fas ligand with anti-peptide antibod-
ies. J Immunol 160:4159–4160.
12. Herr I, Posovsky C, Böhler T, Debatin KM 2000 mAb33
from transduction laboratories specifically binds to human
CD95-L. Cell Death Differ 7:129–130.
13. Baker JR, Bretz JD 2000 Specificity questions concerning the
clone 33 anti-Fas ligand antibody. Cell Death Differ 7:8–9.
14. Fiedler P, Eibel H 2000 Antibody mAb33 from Transduction
Laboratories detects human CD95L in ELISA but not in im-
munoblots. Cell Death Differ 7:126–128.
15. Hiromatsu Y, Hoshino T, Yagita H, Koga M, Sakisaka S,
Honda J, Yang D, Kayagaki N, Okumura K, Nonaka K 1999
Functional Fas ligand expression in thyrocytes from patients
with Graves’ disease. J Clin Endocrinol Metab 84:2896–2902.
16. Baker JR 1999 Dying (apoptosing?) for a consensus on the
Fas death pathway in the thyroid [editorial]. J Clin En-
docrinol Metab 84:2593–2595.
17. Mitsiades N, Poulaki V, Tseleni-Balafouta S, Chrousos GP,
Koutras DA 2000 Fas ligand expression in thyroid follicular
cells from patients with thionamide-treated Graves’ disease.
18. Davies TF 2000 A new role for methimazole in autoimmune
thyroid disease: Inducing T cell apoptosis. Thyroid
19. Giordano C, Stassi G, Todaro M, De Maria R, Richiusa P,
Scorsone A, Giordano M, Galluzzo A 1995 Low Bcl-2 ex-
pression and increased spontaneous apoptosis in T-lym-
phocytes from newly diagnosed patients with insulin-de-
pendent diabetes. Diabetologia 38:953–958.
20. Bagnasco M, Pesce GP, Caretto A, Paolieri F, Pronzato
C, Villaggio BS, Giordano C, Betterle C, Canonica GW
1995 Follicular thyroid cells of autoimmune thyroiditis
may coexpress ICAM-1 (CD54) and its natural ligand
LFA-1 (CD11a/CD18). J Allergy Clin Immunol 95:
21. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F, Riccardi
C 1991 A rapid and simple method for measuring thymo-
cyte apoptosis by propidium iodide staining and flow cy-
tometry. J Immunol Methods 139:271–279.
22. Otsuki T, Sakaguki H, Tomokuni A, Aikoh T, Matsuki T,
Kawakami Y, Kusaka M, Ueki H, Kita S, Ueki A 1998 Solu-
ble Fas mRNA is dominantly expressed in cases with sili-
cosis. Immunology 94:258–262.
APOPTOSIS PATTERN IN AITD
23. Laemmli UK 1970 Cleavage of structural protein during the
assembly of the head of bacteriophage T4. Nature
24. Ausubel FM, Kingston RE, Moore DD, Seidman JG, Smith
JA, Struhl KC 1987 In: Current Protocols in Molecular Biol-
ogy. John Wiley & Sons, New York, pp. 16–17.
25. Arscott PL, Knapp J, Rymaszewski M, Bartron JL, Bretz JD,
Thompson NW, Baker JR 1997 Fas (APO-1, CD95)-mediated
apoptosis in thyroid cells is regulated by a labile protein in-
hibitor. Endocrinology 138:5019–5027.
26. Hammond LJ, Lowdell MW, Cerrano PG, Goode AW, Bot-
tazzo GF, Mirakian R 1997 Analysis of apoptosis in relation
to tissue destruction associated with Hashimoto’s autoim-
mune thyroiditis. J Pathol 182:138–144.
27. Kawakami A, Eguchi K, Matsuoka N, Tsuboi M, Kawabe Y,
Ishikawa N, Ito K, Nagataki S 1996 Thyroid-stimulating hor-
mone inhibits Fas-antigen-mediated apoptosis of human
thyrocytes in vitro. Endocrinology 137:3163–3169.
28. Mitsiades N, Poulaki V, Kotoula V, Mastorakos G, Tseleni-
Balafouta S, Koutras DA, Tsokos M 1998 Fas/Fas ligand up-
regulation and BCL-2 down-regulation may be significant
in the pathogenesis of Hashimoto’s thyroiditis. J Clin En-
docrinol Metab 83:2199–2203.
29. Batteux F, Tourneur L, Trebeden H, Charreire J, Chioccia G.
1999 Gene therapy of experimental autoimmune thyroiditis
by in vivo administration of plasmid DNA coding for Fas
ligand. J Immunol 162:603–608.
30. Batteux F, Lores P, Bucchini D, Chioccia G 2000 Transgenic
expression of Fas ligand on thyroid follicular cells prevents
autoimmune thyroiditis. J Immunol 164:1681–1688.
31. Griffith TS, Brunner T, Fleitcher SM, Green DR, Ferguson
TA 1995 Fas ligand-induced apoptosis as a mechanism of
immune privilege. Science 270:1189–1192.
32. Bellgrau D, Gold D, Selawry H, Moore J, Franzusoff A, Duke
RC 1995 A role for CD95 ligand in preventing graft rejec-
tion. Nature 377:630–632.
33. Williams N 1997 Thyroid disease: A case of cell suicide? [edito-
rial] Science 275:926.
34. Prabhakar BS, Fan JL, Seethramaiah GS 1997 Thyrotropin-
receptor-mediated disease: A paradigm for receptor au-
toimmunity. Immunol Today 18:437–442.
35. Dlits SM, Lafferty KJ 1999 Autoimmune diabetes: The in-
volvement of benign and malignant autoimmunity. J Au-
36. Pakala SV, Chivetta M, Kelly CB, Katz JD 1999 In autoim-
mune diabetes the transition from benign to pernicious in-
sulitis requires an islet cell response to tumor necrosis fac-
tor alpha. J Exp Med 189:1053–1062.
Address reprint requests to:
Carla Giordano, M.D.
Laboratory of Immunology
Institute of Clinica Medica
University of Palermo
Piazza delle Cliniche 2
GIORDANO ET AL.