Several years ago, we saw a number of neonates who
developed systemic exanthema, fever, low-positive
serum C-reactive protein (CRP) values, and thrombo-
cytopenia within the first week of life and described
this disorder as a new disease entity (1, 2). Subse-
quently, several research groups, including our own,
found that virtually all of the neonatal patients with
this disease had been colonized by methicillin-resist-
ant Staphylococcus aureus (MRSA) that produced selec-
tively toxic shock syndrome (TSS) toxin-1 (TSST-1)
(3–6), which has superantigenic activity (7, 8). This
finding suggested a close relationship between this
disease and TSS, which is caused by overactivation of
TSST-1–reactive T cells (8–11). We found recently that
Vβ2+T cells, which are the major TSST-1–reactive
human T cells (12), were polyclonally expanded in the
neonatal patients, and we named this disease neonatal
TSS-like exanthematous disease (NTED) (13). NTED
regressed spontaneously without antibiotic therapy in
full-term neonates, but most preterm neonates devel-
oped severe symptoms (13). According to the respons-
es to our questionnaires, neonates fulfilling the clini-
cal criteria of NTED were observed in 25.7% (19/74) of
major neonatal care units in Japan in 1995 and in
70.8% (63/89) in 1998, indicating that the incidence of
NTED in Japan has been increasing. This finding is
thought to be related to the conversion of MRSA into
the TSST-1–producing type in Japan (14). There is
apprehension that NTED will become a widespread
disease throughout the world, because MRSA has been
spreading on a global level (15, 16).
NTED, like TSS, is caused by the overactivation of
TSST-1–reactive T cells (13). There are several immuno-
logical points that must be resolved to obtain clues that
will allow us a comprehensive understanding of NTED
and will enable the development of methods of preven-
tion. First, a biphasic response consisting of a transient
massive expansion and a subsequent protracted anergic
and deleted state has been found to be induced in super-
antigen-reactive T-cell populations in experiments in
mice (7, 8, 17–19). It remains to be elucidated whether
the same response pattern is induced in NTED patients.
Second, around 20% of the neonates in the neonatal
The Journal of Clinical Investigation|December 2000|Volume 106|Number 11
of an emerging neonatal infectious disease
induced by a bacterial superantigen
Naoto Takahashi,1Hidehito Kato,2Ken’ichi Imanishi,2Keishi Miwa,3Sadao Yamanami,4
Hiroshi Nishida,1and Takehiko Uchiyama2
1Maternal and Perinatal Center, and
2Department of Microbiology and Immunology, Tokyo Women’s Medical University School of Medicine, Tokyo, Japan
3Medical Devices and Diagnostics Research Laboratories, Toray Industries Inc., Ohtsu, Japan
4Neonatal Center, Kawaguchi Municipal Medical Center, Kawaguchi, Japan
Address correspondence to: Naoto Takahashi, Maternal and Perinatal Center, Tokyo Women’s Medical University
School of Medicine, 8-1 Kawadacho, Shinjuku-ku, Tokyo, Japan 162-8666.
Phone: 81-3-3355-3010; Fax: 81-3-5269-7345; E-mail: firstname.lastname@example.org.
Received for publication May 22, 2000, and accepted in revised form October 23, 2000.
We recently discovered an emerging neonatal infectious disease, neonatal toxic shock syn-
drome–like (TSS-like) exanthematous disease (NTED), which is induced by a superantigen, TSS
toxin-1 (TSST-1), produced by methicillin-resistant Staphylococcus aureus (MRSA). Here, we ana-
lyzed the activation and the response of TSST-1–reactive Vβ2+T cells in NTED patients during the
acute and recovery phases and in asymptomatic infants exposed to MRSA. In the acute phase, Vβ2+
T cells were anergic to stimulation with TSST-1 and underwent marked expansion, but by 2 months
after disease onset, their numbers had declined to about 10% of the control level. Although the per-
centage of Vβ2+T cells in the ten asymptomatic neonatal MRSA carriers was within the control
range, these individuals could be divided into two groups on the basis of Vβ2+T-cell activation.
Vβ2+CD4+T cells from three of these infants (Group 1) highly expressed CD45RO and were aner-
gic to TSST-1, whereas in the other seven asymptomatic neonatal MRSA carriers (Group 2), these
cells expressed CD45RO at the control level and were highly responsive to stimulation with
TSST-1. The serum anti–TSST-1 IgG Ab titer was negligible in the four NTED patients in the acute
phase and the three asymptomatic neonatal MRSA carriers in Group 1, but it was high in the seven
asymptomatic carriers in Group 2. We suggest that maternally derived anti–TSST-1 IgGs helps to
suppress T-cell activation by TSST-1 and protects infants from developing NTED.
J. Clin. Invest. 106:1409–1415 (2000).
care unit of Tokyo Women’s Medical University Hospi-
tal are MRSA carriers, and 10% of the 20% have mani-
fested symptoms of NTED. We speculated that T-cell
activation by TSST-1 was induced in a certain propor-
tion of asymptomatic neonatal MRSA carriers and that
a large proportion of them was protected from the
development of NTED by the transplacental transfer of
anti–TSST-1 Ab of maternal origin.
In the present study we obtained evidence indicating
that long-lasting immunological tolerance was
induced in the Vβ2+T cells of the NTED patients, that
activation of Vβ2+T cells occurred in a certain pro-
portion of asymptomatic neonatal MRSA carriers, and
that anti–TSST-1 IgG Ab of maternal origin plays a
protective role in preventing the development of
NTED. We discuss the implications of our findings in
relation to the pathogenetic mechanism underlying
NTED and its prevention.
Neonates examined in the present study. Four NTED
patients, ten asymptomatic neonatal MRSA carriers
and eight MRSA-free neonates were registered as the
subjects of this study. The NTED patients had been
admitted to the neonatal care unit of Tokyo Women’s
Medical University Hospital or Kawaguchi Municipal
Medical Center and had been diagnosed on the basis
of the clinical criteria established by us for NTED, ery-
thema plus at least one of the following three mani-
festations: thrombocytopenia, a low-positive serum
CRP value, and fever (1, 2, 13). Nasal, oropharyngeal
and umbilical swabs, and stools, collected from
neonates on postnatal day 3, were examined for the
presence of MRSA to select asymptomatic MRSA-car-
riers and MRSA-free neonates. After obtaining the
informed consent of their parents, peripheral blood
samples were collected from the NTED patients in the
acute phase and in the recovery phase and from the
asymptomatic MRSA carriers and MRSA-free neonates
on postnatal day 5.
Reagents and mAb’s. FITC-conjugated MPB2D5 (anti-
Vβ2), CH92 (anti-Vβ3), SFCI12T4D11 (anti-CD4), phy-
coerythrin-cyanin 5.1–conjugated (PC5-conjugated)
UCHT1 (anti-CD3), 13B8.2 (anti-CD4), and B9.11 (anti-
CD8) were purchased from Coulter Corp. (Hialeah,
Florida, USA). Phycoerythrin-conjugated (PE-conjugat-
ed) UCHL1 (anti-CD45RO) and SK1 (anti-CD8) were
purchased from Becton Dickinson Immunocytometry
Systems (Mountain View, California, USA). TSST-1 and
staphylococcal enterotoxin A (SEA) were purchased
from Toxin Technology (Sarasota, Florida, USA). The
RPMI-1640 culture medium used contained 10% FCS
and 5 ×10–5M 2-ME. The recombinant IL-2 used in the
IL-2 assay was kindly provided by Takeda Chemical
Industries (Kyoto, Japan).
Characterization of the immunological phenotypes of T cells
by flow cytometry. PBMCs were isolated by Ficoll-Conray
density-gradient centrifugation, as described previously
(20, 21). To examine the percentage of Vβ2+T cells (reac-
tive with TSST-1) or Vβ3+T cells (reactive with staphy-
lococcal enterotoxin B; SEB) (7, 8), PBMCs were stained
with several combinations of adequate PC5-, FITC-, or
PE-conjugated mAb’s and examined by three-color flow-
cytometry analysis using a EPICS XL flow cytometer
(Coulter Corp.), as described previously (20, 21).
Assay of superantigen-induced IL-2 production by mononu-
clear cells. Isolated PBMCs (2 × 105) were stimulated
with 10 ng of TSST-1 or SEA per milliliter in 200-µl vol-
The Journal of Clinical Investigation| December 2000| Volume 106|Number 11
Clinical profiles of the NTED patients, asymptomatic neonatal MRSA carriers, and MRSA-free neonates
Asymptomatic MRSA carriers
MRSA-free neonates (mean ± SD)B
AThe white blood cell counts and platelet counts were maximal and minimal, respectively, on day 2 or 3 after the onset of exanthema in the four NTED patients
and were examined on postnatal day 5 in the asymptomatic MRSA carriers and MRSA-free neonates. BMean ± SD in the eight MRSA-free neonates on post-
natal day 5. GA, gestational age; BW, birth weight; Lym, lymphocyte count; +, MRSA isolated; –, MRSA not isolated; nd, not done.
umes in round-bottom 96-well culture plates (Becton
Dickinson, Franklin Lakes, New Jersey, USA) for vari-
ous durations. IL-2 activity in the culture supernatants
was determined by using IL-2–dependent CTLL-2 cells,
as reported previously (20, 21). Data are shown as units
of IL-2 per milliliter.
Measurement of anti–TSST-1 Ab’s by ELISA. Titers of
anti–TSST-1 Ab’s (IgG and IgM) were measured by
ELISA, as described previously (22). Briefly, serum
diluted to 1:1000 and 1:100 for titration of IgG and
IgM anti–TSST-1 Ab’s, respectively, was applied to
either TSST-1–precoated or noncoated plates, in dupli-
cate. Peroxidase-conjugated rabbit anti-human IgG or
IgM Ab’s (Organo Teknika Corp., West Chester, Penn-
sylvania, USA) and tetramethyl benzidine (Sigma
Chemical Co., St. Louis, Missouri, USA) were added to
the plates, and the Ab titers were determined based on
the OD at 450 nm. Data are shown as OD.
Statistical analysis. Statistical analysis was performed
using the Mann-Whitney’s U-test. P values less than
0.05 were considered significant.
Clinical profiles and laboratory data of NTED patients,
asymptomatic MRSA carriers, and MRSA-free neonates. The
clinical profiles of the four NTED patients, ten asymp-
tomatic MRSA carriers, and eight MRSA-free neonates
examined are shown in Table 1. NTED patient P2 was
admitted to the intensive care unit soon after birth due
to being a preterm infant. The other NTED patients P1,
P3, and P4, were transferred from the newborn nursery
to the intensive care unit after the onset of NTED. Sys-
temic exanthema, fever, low-positive serum CRP values,
and thrombocytopenia were consistently observed in
the full-term neonates, P1, P3, and P4. No fever was
noted in the preterm neonate, P2, who exhibited apnea
attacks and food intolerance and had symptomatic
patent ductus arteriosus. While all three full-term
neonates with NTED recovered spontaneously without
any antibiotic therapy, the preterm neonate, P2, recov-
ered after treatment with vancomycin. On day 2–3 after
the onset of the exanthema, the white blood cell (WBC)
and lymphocyte counts of all four NTED patients were
high, and their platelet counts were low. With the
exception of a slight increase in WBC count in one
MRSA carrier, Ca1, the laboratory data in the ten
asymptomatic neonatal MRSA carriers were almost the
same as in the MRSA-free neonates. It is noteworthy
that the umbilicus was a major site of colonization by
TSST-1–producing MRSA. All asymptomatic MRSA
carriers and MRSA-free neonates remained in the new-
born nursery and were discharged from the hospital
without any clinical manifestations. None of the ten
asymptomatic MRSA carriers showed any symptoms
during the first month of life.
Selective expansion and specific anergy induction in Vβ2+
T cells in acute-phase NTED patients. The immunological
state of the four NTED patients was investigated. First,
the peripheral blood mononuclear (PBM) T cells of the
NTED patients and MRSA-free neonates were exam-
ined to determine the percentage of Vβ2+T cells. The
results are summarized in Table 2. The number of
CD3+T cells was increased in three of the four NTED
patients, and the percentages of Vβ2+CD4+and Vβ2+
CD8+T cells were significantly higher in the four
NTED patients than in the eight MRSA-free neonates.
A high percentage of these expanded Vβ2+T cells in the
NTED patients expressed CD45RO, whereas only a low
or negligible percentage of the Vβ2+T cells of the eight
MRSA-free neonates expressed CD45RO. The percent-
ages of Vβ3+CD4+and Vβ3+CD8+T cells of the NTED
patients that are reactive with staphylococcal entero-
toxin B (SEB) (7, 8) were low (Table 2), as reported pre-
viously (13). These results indicate that Vβ2+T cells
The Journal of Clinical Investigation|December 2000| Volume 106|Number 11
Expansion and anergy induction to TSST-1 in TCR Vβ2+T cells from NTED patients in the acute phase
Percentage of IL-2 production (U/ml)C
Vβ2+ T cells
Vβ3+ T cells
Period of stimulation
8 h CD8+
Toxin24 h 48 h
P45 9.86.0nd nd
MRSA-free neonates (controls)D
54.4 ± 1.74.0 ± 2.0 11.5 ± 1.5
(5.1 ± 2.5) (1.0 ± 1.1) (6.2 ± 1.8)(0.8 ± 0.8)
6.1 ± 0.9 5.3 ± 2.7 4.9 ± 2.7TSST-1
2.7 ± 0.9 33.6 ± 5.2 76.3 ± 13.4
2.8 ± 0.6 40.1 ± 8.8 86.0 ± 55.4
APercentage of Vβ2+CD4+and Vβ2+CD8+T cells or Vβ3+CD4+and Vβ3+CD8+T cells among PBM T cells. BPercentage of the CD45RO+fraction among the Vβ2+CD4+
and Vβ2+CD8+T cells or Vβ3+CD4+and Vβ3+CD8+T cells. CPBMCs (2 × 105/culture) were stimulated in vitro with 10 ng of TSST-1 or SEA per milliliter for indi-
cated periods, and the culture supernatants were assayed for IL-2 activity. DMean ± SD in the eight MRSA-free neonates on postnatal day 5. nd, not done.
exhibited selective activation and expansion induced by
TSST-1 in the NTED patients, irrespective of the CD4+
or CD8+T-cell subsets. The flow-cytometric findings
for the expansion of Vβ2+T cells expressing CD45RO
in patient P2 are shown in Figure 1a.
Second, to investigate whether the expanded TSST-
1–reactive T cells retained their ability to respond to
stimulation with TSST-1 in the NTED patients, we
examined IL-2 production by PBM T cells from the
NTED patients and MRSA-free neonates in response to
in vitro stimulation with TSST-1 or an unrelated super-
antigen, SEA. As shown in Table 2, PBM T cells from
MRSA-free neonates exhibited marked IL-2 production
in response to stimulation with either TSST-1 or SEA.
By contrast, the PBM T cells from the NTED patients
exhibited no or only a minimum level of IL-2 produc-
tion in response to stimulation with TSST-1, but exhib-
ited IL-2 production that was substantial, although
slightly lower than in the MRSA-free neonates, in
response to stimulation with SEA. The results indicate
that the expanded TSST-1–reactive Vβ2+T cells of the
NTED patients were specifically anergic to TSST-1, as
seen in the superantigen-reactive T cells of mice inject-
ed with these superantigens (17–19).
Protracted deletion of Vβ2+T cells in NTED patients. We
monitored the Vβ2+T cells in the peripheral blood of
the four NTED patients for certain periods after the
onset of the disease to determine their fate. The
results are shown in Figure 2. The percentages of
Vβ2+CD4+T cells and Vβ2+CD8+T cells in the four
patients were quite high in the acute phase, as shown
above, but decreased to around the levels of the
MRSA-free neonates on postnatal day 5 (the control)
within 10 days and to very low levels, around 10% of
the control level, by 1 or 2 months after the onset of
disease (Figure 2, a and b). The 5-month follow-up
examination of patient P1 revealed 50% recovery of
the deleted Vβ2+T cells to the control level. The per-
centages of Vβ2+CD4+and Vβ2+CD8+T cells were
almost the same in the eight MRSA-free neonates on
postnatal day 5, the one MRSA-free neonate on post-
natal day 39, and the seven healthy adults, respective-
ly (Figure 2, a and b), indicating that the levels of
Vβ2+CD4+and Vβ2+CD8+T cells do not change much
with age in healthy individuals. These findings indi-
cate that a biphasic response consisting of transient
expansion and subsequent specific deletion was
induced in the Vβ2+T cells of the NTED patients.
Immunological state of TSST-1–reactive T cells in asympto-
matic neonatal MRSA carriers. The immunological state
of the asymptomatic neonatal MRSA carriers was
investigated. First, PBM T cells from the ten asympto-
matic neonatal MRSA carriers were examined for the
expression of Vβ2 and CD45RO on postnatal day 5.
Although the percentage of their Vβ2+T cells was with-
in the normal range for both CD4+and CD8+T cells, we
found increased expression of CD45RO by Vβ2+CD4+
T cells in three (Ca1–Ca3) of the ten neonatal carriers
(Table 3), suggesting that Vβ2+CD4+T cells were acti-
The Journal of Clinical Investigation|December 2000| Volume 106|Number 11
Expression of T-cell receptor Vβ2 versus CD45RO on CD4+and CD8+
T cells obtained from an NTED patient in the acute and recovery
phases, an asymptomatic MRSA carrier, and an MRSA-free neonate.
The percentage of Vβ2+T cells and the expression levels of CD45RO
by Vβ2+CD4+and Vβ2+CD8+T cells were examined in the preterm
NTED patient P2 on postnatal days 4 and 32 (a), in the asympto-
matic neonatal MRSA carrier Ca1 on postnatal day 5 (b), and in an
MRSA-free neonate on postnatal day 5 (c). The numbers are the per-
centages of stained cells in each area.
Fate of TCR Vβ2+T cells in NTED patients. The four NTED patients
in the acute and recovery phases, nine MRSA-free neonates on post-
natal day 5 or 39, and seven healthy adults were examined for the
percentage of Vβ2+CD4+T cells (a) and Vβ2+CD8+T cells (b) among
PBMCs. NTED patients P1 (open circles), P2 (open triangles),
P3 (open squares), and P4 (open diamonds) refer to the same cases
as in Tables 1 and 2. Filled squares, MRSA-free neonates on postna-
tal day 5; filled diamonds, MRSA-free neonates on postnatal day 39;
filled circles, healthy adults.
vated by TSST-1 in these neonates and intact in the
other seven neonates. The flow-cytometric findings in
Ca1 are shown in Figure 1b.
We then examined IL-2 production by PBM T cells in
response to in vitro stimulation with TSST-1 and SEA
in the above three (Ca1–Ca3) of the ten asymptomatic
neonatal MRSA carriers and in three (Ca4–Ca6) of the
other seven carriers. The PBM T cells of neonates
Ca1–Ca3 did not produce IL-2 in response to stimula-
tion with TSST-1 but responded normally to stimula-
tion with SEA. By contrast, the PBM T cells from
neonates Ca4-Ca6 responded normally to both TSST-
1 and SEA (Table 3). The IL-2 response of neonates
Ca7–Ca10 was not examined. The results indicated
that the Vβ2+CD4+T cells of neonates Ca1–Ca3 were
anergic to TSST-1, but that the Vβ2+CD4+T cells of
neonates Ca4–Ca6 were intact. The PBM T cells of
neonates Ca7–Ca10 presumably responded normally
to both TSST-1, and SEA because the level of expres-
sion of CD45RO by their Vβ2+CD4+T cells was within
the control range (Table 3).
The results indicated that the Vβ2+CD4+T cells of
three (Ca1–Ca3) of the ten asymptomatic neonatal
MRSA carriers had been activated by TSST-1 and sug-
gested that they were not activated by TSST-1 in the
other seven, but were intact. We therefore arbitrarily
divided the ten asymptomatic neonatal MRSA carriers
into two groups on the basis of Vβ2+T-cell activation:
Group 1, neonatal carriers (Ca1–Ca3) with Vβ2+CD4+T
cells activated by TSST-1; and Group 2, carriers
(Ca4–Ca10) with intact Vβ2+CD4+T cells.
Protective role of anti–TSST-1 Ab of maternal origin against
the influence of TSST-1. Neonates are generally protected
against various infectious agents by specific IgG Ab’s
transferred transplacentally from their mothers (23). As
anti–TSST-1 Ab’s are known to play a protective role
against the development of TSS in adults (24), it seems
likely that anti–TSST-1 Ab’s transferred placentally from
their mothers play a role in protecting neonates from
developing NTED. We determined the titers of
anti–TSST-1 IgG and IgM Ab’s in the serum of the
NTED patients in the acute and recovery phases, the
asymptomatic neonatal MRSA carriers, and the MRSA-
free neonates on postnatal day 5 (Figure 3). In the four
neonates with NTED, the serum anti–TSST-1 IgG Ab
titer was negligible in the acute phase of the disease, but
The Journal of Clinical Investigation| December 2000| Volume 106| Number 11
Immunologic state of the asymptomatic neonatal MRSA carriers
Percentage of Vβ2+T cellsA
IL-2 production (U/ml)C
Period of stimulation
CD4 to CD8
8 h24 h
4.4 ± 1.74.0 ± 2.0 11.5 ± 1.5 6.1 ± 0.9TSST-12.7 ± 0.933.6 ±
(5.1 ± 2.5)(1.0 ± 1.1)SEA2.8 ± 0.6 40.1 ±
Ten asymptomatic neonatal MRSA carriers were examined on postnatal day 5 for expression of CD45RO by Vβ2+PBM T cells and TSST-1–induced IL-2 production by
PBMCs. They were divided arbitrarily into two groups on the basis of activation level of Vβ2+CD4+T cells in three cases (Ca1–Ca3) with high CD45RO expression by
Vβ2+CD4+T cells and low TSST-1–induced IL-2 production by PBMCs (Group 1) and seven cases (Ca4–Ca10) with low CD45RO expression by Vβ2+CD4+T cells and
high IL-2 production (Group 2). Although TSST-1–induced IL-2 production was not examined in four neonates (Ca7–Ca10), we classified them into Group 2 on the
basis of their low expression of CD45RO by Vβ2+CD4+T cells. APercentage of Vβ2+T cells among PBM CD4+or CD8+T cells. BPercentage of CD45RO+cells among
Vβ2+CD4+and Vβ2+CD8+T cells (P =0.017 between Groups 1 and 2). CPBMCs (2 ×105/culture) were stimulated in vitro with 10 ng of TSST-1 or SEA per milliliter for
8 hours or 24 hours, and the culture supernatants were assayed for IL-2 activity. DMean ±SD in eight MRSA-free neonates on postnatal day 5. nd, not done.
increased within 1 month in the patients P1, P3, and P4,
but not in the preterm patient P2. In the asymptomatic
neonatal MRSA carriers, the anti–TSST-1 IgG Ab titer
was almost negligible in the three Group 1 neonates
with Vβ2+CD4+T cells activated by TSST-1 (Table 3), but
was high (0.2 or more) in the Group 2 neonates without
Vβ2+CD4+T cells activated by TSST-1. The anti–TSST-1
Ab titer was low (less than 0.2) in three of the eight
MRSA-free neonates examined and high (more than 0.2)
in the other five. The anti–TSST-1 IgM Ab titer was neg-
ligible soon after their birth (4–7 days) in all four NTED
patients, the asymptomatic neonatal MRSA carriers,
and the MRSA-free neonates (data not shown). At 1
month of life, the anti–TSST-1 IgM Ab titer had
increased only slightly, from less than 0.01 to 0.02–0.03,
in three NTED patients.
In the present study, we examined several immunolog-
ical aspects of a newly discovered neonatal disease,
NTED, induced by superantigen TSST-1 in order to
obtain clues that could lead to a comprehensive under-
standing of the pathogenetic mechanism underlying
NTED and the development of methods to prevent the
disease. The results clarified the immunological events
that occurred in TSST-1–reactive human T cells and
the protective role of anti–TSST-1 IgG Ab’s transferred
transplacentally against the development of NTED.
The TSST-1–reactive Vβ2+T cells that exhibited
expansion in the acute phase of NTED were found to
be anergic to TSST-1 (Table 2), and the Vβ2+T cells in
the expanded state were subsequently eliminated to
around 10% of the control level by 1–2 months after the
onset of disease (Figure 2), indicating that a response
pattern similar to that in mice injected with bacterial
superantigens (7, 8, 17–19) was also induced in human
neonates exposed to TSST-1. Recovery of the deleted
Vβ2+T cells to 50% of the control level occurred at
around 5 months (Figure 2), suggesting that 6 months
or more are required for the recovery of deleted clones
to normal levels in neonates. As in the NTED patients,
expansion of Vβ2+T cells was found previously in adult
TSS patients (ref. 11 and our unpublished data). The
deletion of the expanded Vβ2+T cells, however, was not
as profound or rapid in the adult TSS patients; the
Vβ/Cαratio of the Vβ2+examined by the PCR method
in two blood samples drawn 25 and 50 days after the
onset of symptoms was still higher than that in the
control (11). We think that the implications of the
results of these studies in TSS in adults and NTED are
important in terms of T-cell maturity in the neonatal
period and the outcome of NTED.
According to Burnet’s clonal selection theory (25),
elimination of self-reactive lymphocytes occurs early
in life, and the results of many experiments, including
those of Billingham et al. (26), mainly using mice, have
supported his concept and given rise to the notion
that T cells in the neonatal period are intrinsically sus-
ceptible to anergy induction. This view has been refut-
ed by reports suggesting that the T cells of neonatal
and mature individuals are not qualitatively different,
as reviewed by Stockinger (27). Recently, however, we
observed that human CD1a–CD4+T cells in the final
stage of maturation in the thymus and cord-blood
CD4+T cells that supposedly had migrated recently
from the thymus were susceptible to anergy induction
by in vitro stimulation with TSST-1, whereas adult
PBM CD4+T cells were resistant (20, 21), indicating
the immaturity of human T cells during the neonatal
period. On the basis of the results of the studies
described above, we think that the rapid elimination
of TSST-1–reactive Vβ2+T cells after their transient
expansion in NTED patients is a reflection of the
intrinsically immature state of the T cells resident in
neonates. Because NTED is caused by overactivation
of TSST-1–reactive T cells, mainly Vβ2+T cells (13), the
rapid recovery from the illness without any complica-
tions in most full-term NTED patients (1, 2, 13) seems
to be largely attributable to this immaturity of the
T cells, the high susceptibility to anergy induction, and
rapid deletion of Vβ2+T cells in the early neonatal
period. The multiorgan failure seen in mature TSS
patients could be caused by the persistent activated
state of the Vβ2+T cells.
The percentages of Vβ2+CD4+T cells and Vβ2+CD8+
T cells in the ten asymptomatic neonatal MRSA car-
riers examined were within the normal range (Table
3), however, they could be divided into two groups,
Group 1 and Group 2, on the basis of the TSST-
1–induced activation of their Vβ2+T cells (Table 3).
The Vβ2+T cells in Group 1 neonates were activated
by TSST-1, but they were intact and not activated in
Group 2 neonates. These results suggest that the
TSST-1–reactive T cells are activated by TSST-1 in
The Journal of Clinical Investigation|December 2000| Volume 106|Number 11
Titers of anti–TSST-1 IgG Ab in NTED patients, asymptomatic MRSA
carriers, and MRSA-free neonates. Anti–TSST-1 IgG Ab titers were
examined by ELISA in the serum of the four NTED patients P1 (open
circles), P2 (open triangles), P3 (open squares), P4 (open diamonds)
in the acute and recovery phases, ten asymptomatic neonatal MRSA
carriers on postnatal day 5 (filled circles), and seven MRSA-free
neonates on postnatal day 5 (filled squares). They were determined
based on the OD at 450 nm.
about 30% of neonatal MRSA-carriers, although the Download full-text
number of neonates examined was too low to evalu-
ate the these findings statistically.
A question arises as to what factors divided the neona-
tal MRSA carriers into NTED patients and the Group 1
and Group 2 asymptomatic neonatal MRSA carriers.
The view that NTED is caused by overactivation of
TSST-1–reactive T cells (13) suggests that the level of
activation of Vβ2+T cells governs the development of
NTED. We think that the level of activation of Vβ2+T
cells can be determined mainly by the amount of TSST-
1 absorbed and anti–TSST-1 IgG Ab that can neutralize
the superantigenic activity of TSST-1 in neonates. The
level of Vβ2+T-cell activation in the four NTED patients
was clearly higher than in the three Group 1 asympto-
matic neonatal MRSA carriers, as shown in Table 2,
Table 3, and Figure 1. Another clue as to the answer to
the question was obtained from the analysis of the
anti–TSST-1 IgG Ab titer in the NTED patients and the
asymptomatic neonatal MRSA carriers. The results
showed that the serum anti–TSST-1 IgG Ab titer was
negligible in all four NTED patients and Group 1
neonates, but high in the Group 2 neonates in the early
days after birth (Figure 3). We presume that higher
amounts of TSST-1 were absorbed in the NTED patients
than in the Group 1 neonates.
Anti–TSST-1 IgG Ab was found to be effective in
blocking the superantigenic activity of TSST-1 and pro-
tecting against the development of NTED, as shown in
Figure 3 and discussed above. This finding indicates that
transfer of high amounts of anti–TSST-1 IgG Ab from
mothers to their children through the placenta are effec-
tive in protecting against the development of NTED.
The effectiveness of vaccination in preventing abnormal
superantigen-induced reactions by a superantigen that
has lost its superantigenic activity has been studied in
animal experiments (28). Vaccination of pregnant
women with attenuated TSST-1 may be one means of
preventing the development of NTED in their children.
There are several questions as to whether TSST-
1–reactive T cells of NTED patients retain long-term
memory against TSST-1 that can be seen in the anergy
induction to stimulation with TSST-1 and whether the
TSST-1–reactive T cells in an activated state in the
Group 1 neonates normalize with time. Does complete
deletion of Vβ2+T cells occur in the recovery phase in
NTED patients, when the patients are exposed to a
high amounts of TSST-1? These questions are left to be
clarified in future studies.
This work was supported by grants from the Ministry
of Education, Science, Sports and Culture of Japan, the
Ministry of Health and Welfare of Japan, and The Moth-
er and Child Health Foundation. The authors thank M.
Maruyama for identification of the MRSA carriers and
I. Sakuma for collecting the blood specimens.
1.Takahashi, N., et al. 1995. A new exanthematous disease in newborn
infants. Acta Neonatologica Japonica. 31:371–377.
2.Takahashi, N., and Nishida, H. 1997. A new exanthematous disease with
thrombocytopenia in newborn infants. Arch. Dis. Child. 77:F79.
3.Makimoto, A., Kubo, M., and Kawakami, K. 1996. Neonatal thrombo-
cytopenia with skin eruption, probably caused by exotoxin-producing
methicillin-resistant Staphylococcus aureus(MRSA): clinical evaluation in
14 cases. Journal of the Japan Pediatric Society. 100:609–615.
4.Matsuo, Y., et al. 1997. Clinical studies of early neonatal exanthematous
disease with thrombocytopenia: in relation to methicillin resistant
Staphylococcus aureus (MRSA). J. Jpn. Soc. for Premature Newborn Med.
5.Okada, T., Furukawa, S., Satomura, S., Hayabuchi, Y., and Ohta, A. 1997.
Bacterial examinations of a new exanthematous disease in newborn
infants. Journal of the Japan Pediatric Society. 101:631–637.
6.Sakata, Y., et al. 1997. Exanthema of the newborn related to toxic shock
syndrome toxin-1 (TSST-1). Japanese Journal of Obstetrical, Gynecological
and Neonatal Hematology. 7:146–150.
7.Kotzin, B.L., Leung, D.Y., Kappler, J., and Marrack, P. 1993. Superanti-
gens and their potential role in human disease. Adv. Immunol.54:99–166.
8.Uchiyama, T., Yan, X.-J., Imanishi, K., and Yagi, J. 1994. Bacterial super-
antigens: mechanism of T cell activation by the superantigens and their
role in the pathogenesis of infectious diseases. Microbiol. Immunol.
9.Uchiyama, T., et al. 1987. Study of the biological activities of toxic shock
syndrome toxin-1: induction of the proliferative response and inter-
leukin 2 production by T cells from human peripheral mononuclear
cells stimulated with the toxin. Clin. Exp. Immunol. 68:638–647.
10.Uchiyama, T., et al. 1989. Activation of human T cells by toxic shock syn-
drome toxin-1: the toxin binding structures expressed on human lym-
phoid cells acting as accessory cells are HLA class II molecules. Eur. J.
11.Choi, Y., et al. 1990. Selective expansion of T cells expressing Vβ2 in toxic
shock syndrome. J. Exp. Med. 172:981–984.
12.Choi, Y., et al. 1989. Interaction of Staphylococcus aureus toxin “super-
antigens” with human T cells. Proc. Natl. Acad. Sci. USA. 86:8941–8945.
13.Takahashi, N., et al. 1998. Exanthematous disease induced by toxic shock
syndrome toxin 1 in the early neonatal period. Lancet. 351:1614–1619.
14.Cui, L., Hanaki, H., and Hiramatsu, K. 1998. The epidemiology of MRSA
producing TSST-1 and the influence of serum on the production of
TSST-1 [in Japanese]. Japanese Journal of Bacteriology. 53:155. (Abstr.)
15.Ayliffe, G.A. 1997. The progressive intercontinental spread of methicillin-
resistant Staphylococcus aureus. Clin. Infect. Dis. 24(Suppl. 1):S74–S79.
16.Archer, G.L. 1998. Staphylococcus aureus: a well-armed pathogen. Clin.
Infect. Dis. 26:1179–1181.
17.Kawabe, Y., and Ochi, A. 1990. Selective anergy of Vβ8+, CD4+T cells in
Staphylococcus enterotoxin B-primed mice. J. Exp. Med.172:1065–1070.
18.Rellahan, B.L., Jones, L.A., Kruisbeek, A.M., Fry, A.M., and Matis, L.A.
1990. In vivo induction of anergy in peripheral Vβ8+ T cells by staphy-
lococcal enterotoxin B. J. Exp. Med. 172:1091–1100.
19.Kuroda, K., et al. 1996. Implantation of IL-2-containing osmotic pump
prolongs the survival of superantigen-reactive T cells expanded in mice
injected with bacterial superantigen. J. Immunol. 157:1422–1431.
20.Takahashi, N., Imanishi, K., Nishida, H., and Uchiyama, T. 1995. Evi-
dence for immunologic immaturity of cord blood T cells. Cord blood T
cells are susceptible to tolerance induction to in vitro stimulation with
a superantigen. J. Immunol. 155:5213–5219.
21.Imanishi, K., et al. 1998. Post-thymic maturation of migrating thymic
single positive T cells: thymic CD1a-CD4+T cells are more susceptible
to anergy induction by toxic shock syndrome toxin-1 than cord blood
CD4+T cells. J. Immunol. 160:112–119.
22.Terai, M., et al. 1995. The absence of evidence of staphylococcal toxin involve-
ment in the pathogenesis of Kawasaki disease. J. Infect. Dis. 172:558–561.
23.Wilson, C.B. 1990. Developmental immunology and role of host defens-
es in neonatal susceptibility. In Infectious diseases of the fetus and newborn
infant.3rd edition. J.S. Remington and J.O. Klein, editors. W.B. Saunders
Co. Philadelphia, Pennsylvania, USA. 17–67.
24.Vergerout, J., et al. 1983. Prevalence of serum antibody to staphylococ-
cal enterotoxin F among Wisconsin residents: implications for toxic
shock syndrome. J. Infect. Dis. 148:692–698.
25.Burnet, F.M. 1959. The clonal selection theory of acquired immunity. Cam-
bridge University Press. Cambridge, United Kingdom. 208 pp.
26.Billingham, R.E., Brent, L., and Medawar, P.B. 1953. ‘Actively acquired
tolerance’ of foreign cells. Nature. 172:603–606.
27.Stockinger, B. 1996. Neonatal tolerance mysteries solved. Immunol.
28.Bavari, S., Dyas, B., and Ulrich, R. 1996. Superantigen vaccines: a com-
parative study of genetically attenuated receptor-binding mutants of
staphylococcal enterotoxin A. J. Infect. Dis. 174:338–345.
The Journal of Clinical Investigation|December 2000| Volume 106|Number 11