| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Laboratory Animal Science, Tokyo Metropolitan Organization for Medical Science, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan;
Department of Molecular Immunology, Graduate School of Medicine, Chiba University, Chiba, Japan;
Japan Society for the Promotion of Science, Saitama, Japan; and
Department of Immune Regulation, Tokyo Medical and Dental University, Graduate School, Tokyo, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI expressed on their
surface. Previous studies demonstrated that IgE binding induced the
stabilization and accumulation of FcRI on the cell surface and
resulted in up-regulation of FcRI. In this study we have further
analyzed the maintenance of IgE-mediated memory in mast cells and
basophils in vivo by comparing kinetics of serum IgE levels, FcRI
expression, and ability to induce systemic anaphylaxis. A single i.v.
injection of trinitrophenyl-specific IgE induced 8-fold up-regulation
of FcRI expression on peritoneal mast cells in B cell-deficient
(µm-/-) mice. Serum IgE levels became undetectable by
day 6, but the treatment of mice with anti-IgE mAb induced a
significant drop in body temperature on days 14, 28, and 42. The
administration of trinitrophenyl -BSA, but not BSA, in place of
anti-IgE mAb gave similar results, indicating the Ag specificity of
the allergic response. This long term maintenance of Ag-specific
reactivity in the allergic response was also observed in normal mice
passively sensitized with IgE even though the duration was shorter than
that in B cell-deficient mice. The appearance of IgE with a different
specificity did not interfere with the maintenance of IgE-mediated
memory of mast cells and basophils. These results suggest that
IgE-mediated stabilization and up-regulation of FcRI enables mast
cells and basophils not only to acquire Ag specificity, but also to
maintain memory in vivo for lengthy periods of
time. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI) expressed on these cells is a
critical component in allergic responses, since this receptor allows
the cells to bind IgE. Ligation of IgE-bound FcRI by multivalent Ags
results in the activation of multiple signaling pathways leading to
diverse effector responses, including the release of mediators
responsible for allergic inflammatory reactions. Thus, the binding of
Ag-specific IgE confers specific reactivity to that Ag on mast cells
and basophils (4, 5, 6, 7, 8).
IgE binding to FcRI induces the up-regulation of FcRI expression
on mast cells and basophils in both humans and mice
(9, 10, 11, 12, 13, 14, 15, 16). In IgE-deficient mice, levels of FcRI
expression on mast cells are only 
20% of normal levels
(12). However, FcRI expression could be up-regulated by
as much as 32-fold after in vitro incubation of mast cells with IgE or
injection of IgE in vivo (12, 17). Furthermore,
IgE-transgenic mice had highly elevated levels of FcRI on mast
cells, as has also been observed in patients with allergies (9, 15, 18). The IgE-mediated FcRI up-regulation was shown to
enhance the ability of mast cells and basophils to release chemical
mediators such as serotonin and cytokines (12, 19).
Indeed, the i.v. administration of nonanaphylactogenic anti-IgE Ab
to atopic patients resulted in down-regulation of FcRI on basophils
in parallel with the reduction of mediator release from activated
basophils (20, 21). These results emphasize the functional
importance of FcRI regulation by IgE.
Two different mechanisms have been proposed to explain the IgE-mediated
FcRI up-regulation (7, 8, 13, 15). One mechanism
involves the suppression of loss of preformed cell surface FcRI by
protection against degradation of FcRI. The other mechanism involves
an enhancement of the synthesis and/or transport of the FcRI complex
through FcRI-mediated signaling. We and others have recently
demonstrated that the stabilization and accumulation of cell surface
FcRI through IgE binding are the major mechanism of IgE-mediated
FcRI up-regulation (17, 22). Cell surface FcRI is
unstable and is quickly removed from cell surface unless it binds IgE.
FcRI with bound IgE is stabilized and stays on the surface longer
than FcRI with no bound IgE. In the presence of excess amounts of
IgE, every new FcRI transported to the cell surface is loaded with
IgE, stabilized, and accumulates on the cell surface, leading to the
up-regulation of surface FcRI expression.
Up to 80% of the FcRI on mast cells is occupied with IgE even
in normal BALB/c mice with basal levels of serum IgE, and IgE-bound
FcRI molecules are stable on the cell surface (17).
Therefore, only limited space is available for newly produced IgE in
the absence of IgE-mediated FcRI up-regulation. In the case of T and
B cells, each cell only has specificity to a single Ag. Therefore, the
population reactive to a particular Ag is extremely small, and clonal
expansion is necessary to protect the host against the foreign Ag. In
contrast, large numbers of mast cells can simultaneously acquire new Ag
specificity through FcRI up-regulation induced by newly produced IgE
without expansion of cells. In the present study we asked how long mast
cells and basophils that had acquired Ag specificity through IgE
binding to FcRI could maintain Ag-specific reactivity sufficient to
induce allergic responses in vivo. Our previous study demonstrated that
levels of IgE-bound FcRI on cultured mast cells did not
significantly decline in vitro at least for 24 h
(17). Therefore, one may assume that mast cells and
basophils can maintain Ag-specific reactivity through FcRI-bound IgE
for lengthy periods, not only in vitro but also in vivo. On the other
hand, the turnover of mast cells and IgE could be different in vivo and
in vitro, thus shortening the duration of in vivo memory. There are no
reported comprehensive studies on mast cell memory in vivo. Here we
show that mice passively sensitized with Ag-specific IgE keep
Ag-specific reactivity that can induce allergic responses such as
systemic anaphylaxis long after the serum Ag-specific IgE has become
undetectable. Here we will discuss the pathophysiological roles of
IgE-mediated FcRI up-regulation as they relate to memory formation
of mast cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice were purchased from Japan SLC (Hamamatsu, Japan). C57BL/6-µm-/- mice (23) were provided by Dr. K. Rajewsky (University of Cologne, Germany) and Dr. D. Kitamura (Tokyo Science University, Tokyo, Japan) and were maintained in our own animal facility. All the experiments in this study were performed according to the Guidelines for Animal Use and Experimentation as specified by our institutions.
Antibodies
Anti-mouse IgEa mAb (UH297, rat IgG1),
FITC-conjugated rat IgG mAb (G28-5), FITC-conjugated anti-mouse IgE
mAb (R35-72), biotinylated anti-mouse c-Kit mAb (2B8), and
allophycocyanin-conjugated streptavidin were purchased from BD
PharMingen (San Diego, CA). Trinitrophenyl
(TNP)3-specific mouse
IgEb mAb (IGELb4), TNP-specific mouse
IgEa mAb (IGELa2), anti-mouse IgE mAb (6DH5)
and anti-Fc
II/III mAb (2.4G2) were described
previously (24, 25, 26). A hybridoma producing
phenylarsonate-specific mouse IgEa mAb (SE1.3)
was purchased from American Type Culture Collection (Manassas,
CA).
Cell preparation and flow cytometry
Peritoneal cells isolated from mice were depleted of RBCs by
lysis with hypotonic buffer and preincubated with 2.4G2 mAb at room
temperature for 5 min to prevent nonspecific binding of other Abs. To
detect IgE-bound FcRI on the cell surface, cells were stained with
FITC-anti-IgE mAb R35-72. To detect total (IgE-bound plus IgE-free)
FcRI, cells were stained with R35-72 after incubation at 4°C with
excess amounts of IgE (IGELb4) to saturate FcRI with IgE. Cells were
also stained with biotinylated anti-c-Kit mAb followed by
allophycocyanin-conjugated streptavidin and then analyzed by
FACSCalibur (BD Biosciences, Mountain View, CA). Autofluorescent cells
(primarily macrophages) were rejected to clearly identify
c-Kit+ mast cells (12, 17). The geo
mean value of fluorescence intensity was converted to the linear scale
number by the number of molecules of equivalent soluble fluorochrome
units (MESF) using Quantum 25 microbeads (Flow Cytometry Standards Co.,
San Juan, Puerto Rica). 
MESF was calculated by subtracting MESF of
control staining from MESF of sample (17). The half-life
of IgE-bound FcRI in vivo was calculated from kinetics of FcRI
levels on peritoneal mast cells in mice treated with Ag-specific IgE.
Namely, after administered IgE became undetectable in sera from the
treated mice, levels of injected IgE-bound FcRI on peritoneal mast
cells were determined at various time points by flow cytometry using
anti-IgE mAb R35-72 in the case of µm-/-
mice treated with IGELb4 or anti-IgEa mAb
UH297 in the case of C57BL/6 mice treated with IGELa2
(IgEa).
Sensitization of mice with IgE and induction of systemic analysis
Mice were treated with an i.v. injection of 300 µg of TNP-specific IgE. Total IgE levels in serum were measured by ELISA as described previously (18). At various time points after IgE injection, mice were challenged with i.v. administration of 10 µg of anti-IgE mAb (6DH5) or 30 µg of BSA in PBS, either with or without conjugation to TNP. The body temperatures of the mice were monitored with a rectal probe (Shibaura Electronics, Tokyo, Japan).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI expression on mast cells
in B cell-deficient mice passively sensitized with IgE
C57BL/6-µm-/- mice, which possess no
mature B cells and hence no endogenous IgE, were treated with i.v.
injection of 300 µg of monoclonal IgE specific for hapten TNP. One
day after the injection, serum IgE levels were 20 µg/ml (Fig. 1, bar graph), comparable to those in
mice infected with parasites such as Nippostrongylus
brasiliensis (16, 27). Serum IgE levels declined
quickly thereafter and became undetectable by day 6. Before IgE
injection, the expression of FcRI at low levels was detected on
peritoneal mast cells even in the absence of IgE (Fig. 1, line graph).
IgE injection induced marked up-regulation of FcRI on peritoneal
mast cells. While serum IgE levels quickly dropped from day 2, levels
of FcRI expression continuously increased up to 8-fold by day 4 and
then decreased gradually (Fig. 1). On day 7 even though serum IgE
levels became undetectable, FcRI expression was still 2–3 times
higher than that before injection. After 28 days FcRI expression had
returned to basal levels.
|
We next questioned how long
C57BL/6-µm-/- mice passively sensitized with
IgE maintained the ability to induce IgE-mediated allergic reactions.
The mice passively sensitized with IgE were treated at various times
with i.v. injection of 10 µg of rat anti-mouse IgE mAb to
cross-link IgE-bound FcRI on mast cells, and change in body
temperature was monitored (Fig. 2A). The treatment with
anti-IgE mAb caused no significant change in body temperature in
unsensitized mice (Fig. 2A, 
). In contrast, severe
temperature drops of -6°C were observed within 35 min in sensitized
mice, followed by their death, even as late as day 14 after
sensitization when serum IgE levels were undetectable and levels of
FcRI were 7 times lower than on day 4 (Fig. 2A, 
).
Treatment with control rat IgG mAb produced no significant change in
body temperature (data not shown). The significant drop in body
temperature was still induced by anti-IgE treatment on days 28 and
42, even though FcRI expression had returned nearly to the basal
level (Fig. 2A, 
, 
). On day 56, three of four
sensitized mice examined in one experiment still showed drops in body
temperature (-1, -1.5, and -6°C respectively) when treated with
anti-IgE mAb (data not shown).
|
IgE-mediated memory formation of mast cells in normal mice
In contrast to the C57BL/6-µm-/- mice
used in the above experiments, normal C57BL/6 mice have low basal
levels of serum IgE, and 50% of the FcRI on peritoneal mast
cells were occupied with IgE. Therefore, we tested whether IgE-mediated
memory can be formed in the presence of such low basal levels of IgE.
Normal C57BL/6 mice were passively sensitized with 300 µg of
TNP-specific IgE, followed by challenge with either TNP-BSA or control
BSA at various time points. Serum IgE levels of <0.4 µg/ml before
sensitization increased to 25 µg/ml on day 1, but returned to the
basal level (<0.4 µg/ml) by day 5 (Fig. 3A, bar graph). Levels of
FcRI expression on peritoneal mast cells increased 5-fold on day 1,
then decreased gradually (Fig. 3A, line graph). The i.v.
injection of TNP-BSA, but not control BSA, induced body temperature
drops even on day 14 when FcRI expression on peritoneal mast cells
had returned to the basal level (Fig. 3, A and
B). One of four sensitized mice examined in one experiment
showed significant drops of body temperature (-4°C) as late as day
21 (data not shown).
|
We next examined whether the appearance of IgE with a different Ag
specificity influenced the maintenance of IgE-mediated memory.
C57BL/6-µm-/- mice were passively sensitized
with 300 µg of TNP-specific IgE (day 0). The expression of FcRI on
peritoneal mast cells increased until day 4 after treatment with
anti-TNP IgE and then declined, as observed in Fig. 1. On day 11
when TNP-specific IgE had become undetectable in their sera, the mice
were treated with i.v. injection of 300 µg of phenylarsonate
(PA)-specific IgE or control PBS. In control mice treated with PBS,
FcRI expression continued to decline, and the relative number of
FcRI on day 14 was 0.44 ± 0.01 x
105 MESF (n = 4), close to the
basal level (0.35 ± 0.02 x 105
MESF; n = 4). In contrast, in mice treated with
anti-PA IgE on day 11, FcRI expression was again up-regulated
after treatment. The relative number of FcRI on day 14 was 2.2
± 0.54 x 105 MESF (n =
4), 5 times higher than that in control mice. In separate experiments
mice treated with anti-TNP IgE, followed by anti-PA IgE or PBS
in the same protocol, were challenged on day 14 with i.v. injection of
30 µg of TNP-BSA or control BSA, and body temperature was monitored
(Fig. 4). A drop in body temperature was
observed in a TNP-specific manner in the mice treated with PA-specific
IgE as much as or even more intensely than in PBS-treated control mice.
Thus, the maintenance of allergic reactivity to the Ag recognized by
the first IgE was not disturbed by subsequent appearance of the second
IgE with a different Ag specificity.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FcRI expressed on mast cells and basophils is most likely involved
in the memory formation, since mice deficient for the 
-chain of
FcRI have been shown to be resistant to IgE-dependent systemic
anaphylaxis (28). In B cell-deficient mice, levels of
serum IgE quickly declined soon after passive IgE sensitization and
became undetectable by day 6. On the other hand, levels of FcRI
expression on peritoneal mast cells continuously increased up to 8-fold
until day 4. The high affinity binding of IgE by FcRI and the
stabilization of FcRI by IgE binding appeared to be sufficient to
up-regulate FcRI on mast cells even as serum IgE levels decreased.
FcRI expression subsided gradually after day 4 and returned to basal
levels by day 28. However, typical systemic anaphylaxis could still be
elicited on days 28 and 42 upon ligation of IgE on mast cells and
basophils by either multivalent Ag or anti-IgE Ab. Therefore, the
elicitation of Ag/IgE-mediated systemic anaphylaxis does not seem to
require increased expression of FcRI. However, we cannot
exclude the possibility that some mast cells localized outside
peritoneal cavity maintained high levels of FcRI expression.
The IgE-mediated FcRI up-regulation and stabilization as well as the
slow kinetics of IgE dissociation from FcRI (8) appear
to be essential for establishing long-lasting memory of mast cells and
basophiles. Approximately 80% of FcRI on mast cells are occupied
with IgE even in normal BALB/c mice with basal levels of serum IgE, and
IgE-bound FcRI molecules are stable on the cell surface
(17). Therefore, only limited space is available for newly
produced IgE unless IgE-mediated FcRI up-regulation is induced. When
substantial amounts of IgE are produced in response to an Ag, every new
FcRI transported to the cell surface binds the Ag-specific IgE and
becomes stabilized on the cell surface. This results in accumulation of
the Ag-specific IgE/FcRI complex on mast cells (Fig. 5). The large number of complexes and
their stability on mast cells enable the IgE-mediated memory to persist
even after the Ag-specific IgE disappears from the circulation.
|
RI on
cultured bone marrow-derived mast cells persisted in vitro at least for
24 h. Levels of IgE-bound FcRI on freshly prepared
peritoneal mast cells also showed no significant change during 16-h
culture ex vivo. In contrast, the half-life of injected IgE-bound
FcRI on mast cells in vivo calculated as described in
Materials and Methods was 24 h in both
µm-/- and normal C57BL/6 mice,
which is much shorter than in vitro. This could be attributed to the
quicker turnover of mast cells in vivo due to the elimination of old
mast cells, a new supply of mast cells from bone marrow, and the
proliferation of mast cells in response to growth factors such as stem
cell factor, IL-3, and IL-9 (29, 30). Nevertheless, the
24-h half-life of IgE-bound FcRI in vivo is still much longer than
the 14-h half-life of free IgE in serum in the present study. Thus,
IgE-mediated stabilization and up-regulation FcRI enable mast cells
and basophils not only to acquire Ag specificity but also to maintain
memory in vivo for long periods of time. The IgE-mediated memory
formation of mast cells and basophils was observed in both B
cell-deficient and normal mice. However, the duration of the memory was
shorter in normal than in B cell-deficient mice. This difference could
be explained in part by the presence of the basal level of IgE in
normal mice. Due to the presence of endogenous IgE, the basal level of
FcRI on mast cells is 2- to 3-fold higher in normal C57BL/6 than in
B cell-deficient mice. Approximately 50 and 80% of FcRI molecules
on mast cells in normal C57BL/6 and BALB/c mice, respectively, are
already occupied with endogenous IgE, whereas all FcRI molecules in
B cell-deficient mice are free of IgE. Consequently, the occupancy of
FcRI by exogenous IgE cannot reach 100% in normal mice even after
IgE-mediated FcRI up-regulation. This appears to explain at least in
part why the duration of the memory was shorter in normal than in B
cell-deficient mice. Infection with parasites sometimes results in
elevation of serum IgE to levels much higher than those observed in our
experiments (27). In such cases, the IgE-mediated mast
cell memory could be maintained for a much longer time in normal
mice.
Analysis of B cell-deficient mice sensitized sequentially with two
different IgE revealed that once the memory had been established, the
maintenance of IgE-mediated memory was not disturbed by the appearance
of IgE carrying the other specificity. The treatment of mice with the
second IgE induced the second wave of FcRI up-regulation. On day 3
after the treatment, FcRI expression became 5 times higher than that
in control mice treated with PBS, and the vast majority of IgE bound to
FcRI was the second IgE. Nevertheless, mast cells kept the
reactivity to the Ag recognized by the first IgE to induce systemic
anaphylaxis. This appears to be attributed to IgE-mediated FcRI
stabilization and slow kinetics of IgE dissociation from FcRI.
Interestingly, the Ag-specific systemic anaphylaxis was more severe in
mice treated with the second IgE than in control mice treated with PBS.
It has been shown that IgE-mediated up-regulation of FcRI expression
significantly enhances the ability of mast cells to release chemical
mediators such as serotonin and cytokines (12, 19). In
mice treated with the second IgE, FcRI expression at the time of Ag
challenge was 5 times higher than that in control mice treated with
PBS. This could be the reason why systemic anaphylaxis was more severe
in mice treated with the second IgE.
In the present study we demonstrated that IgE-mediated memory of mast
cells and basophils could be established and maintained even after
Ag-specific IgE disappears from the circulation. This mechanism of
memory formation and maintenance is totally different from the
immunological memory of B and T cells. In the case of T and B cells, a
small proportion of cells that expanded clonally in response to an Ag
remained as memory cells. In contrast, large numbers of mast cells can
simultaneously acquire new Ag specificity and establish memory. This
mechanism might be beneficial in the protection against repeated
infection with pathogens such as Shistosoma hematobium
(31), but deleterious for allergic patients. IgE-mediated
stabilization and up-regulation of FcRI are the driving forces for
both acquisition of Ag specificity and memory formation in mast cells
and basophils. Therefore, it is important to understand the mechanism
of IgE-mediated FcRI stabilization to develop new types of therapies
for allergic disorders. It would be intriguing to explore the
possibility of memory formation through other Ig isotypes, such as IgG
and IgA.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Shuichi Kubo, Department of Laboratory Animal Science, Tokyo Metropolitan Institute of Medical Science, Tokyo Metropolitan Organization for Medical Science, 3-18-22 Honkomagome, Bunkyo-ku, Tokyo 113-8613, Japan. E-mail address: skubo{at}rinshoken.or.jp
3 Abbreviations used in this paper: TNP, trinitrophenyl; MESF, molecules of fluorescence intensity; PA, phenylarsonate.
Received for publication August 9, 2002. Accepted for publication November 13, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI): from physiology to pathology. Annu. Rev. Immunol. 17:931.[Medline]
RI. Nature 402:B24.
RI expression in vitro and in vivo: evidence for a novel amplification mechanism in IgE-dependent reactions. J. Exp. Med. 185:663. by human mast cells: increased anti-IgE-dependent secretion after IgE-dependent enhancement of mast cell IgE-binding ability. Lab. Invest. 77:185.[Medline]
RI expression on human basophils by IgE antibody. Blood 91:1633.RI expression in vivo. J. Immunol. 158:2517.[Abstract]
RI on mast cells is induced by IgE binding through stabilization and accumulation of FcRI on the cell surface. J. Immunol. 167:3427. receptor I expression. J. Exp. Med. 188:1135.RI expression on human basophils during in vivo treatment of atopic patients with anti-IgE antibody. J. Immunol. 158:1438.[Abstract]
RI surface densities and mediator release by anti-IgE-infusion is reversible in vitro and in vivo. J. Immunol. 162:5624.RI. J. Immunol. 167:1290.This article has been cited by other articles:
|
|
 |
|
  A. S. Iyer and A. August The Tec Family Kinase, IL-2-Inducible T Cell Kinase, Differentially Controls Mast Cell Responses J. Immunol., June 15, 2008; 180(12): 7869 - 7877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Turley and S. D. Miller Peripheral Tolerance Induction Using Ethylenecarbodiimide-Fixed APCs Uses both Direct and Indirect Mechanisms of Antigen Presentation for Prevention of Experimental Autoimmune Encephalomyelitis J. Immunol., February 15, 2007; 178(4): 2212 - 2220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Demarest, J. Hopp, J. Chung, K. Hathaway, E. Mertsching, X. Cao, J. George, K. Miatkowski, M. J. LaBarre, M. Shields, et al. An Intermediate pH Unfolding Transition Abrogates the Ability of IgE to Interact with Its High Affinity Receptor Fc{epsilon}RI{alpha} J. Biol. Chem., October 13, 2006; 281(41): 30755 - 30767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Smith, T. N. Eagar, J. L. Strominger, and S. D. Miller Differential induction of IgE-mediated anaphylaxis after soluble vs. cell-bound tolerogenic peptide therapy of autoimmune encephalomyelitis PNAS, July 5, 2005; 102(27): 9595 - 9600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Vangelista, E. Soprana, M. Cesco-Gaspere, P. Mandiola, G. Di Lullo, R. N. Fucci, F. Codazzi, A. Palini, G. Paganelli, O. R. Burrone, et al. Membrane IgE Binds and Activates Fc{epsilon}RI in an Antigen-Independent Manner J. Immunol., May 1, 2005; 174(9): 5602 - 5611. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Mitre, R. T. Taylor, J. Kubofcik, and T. B. Nutman Parasite Antigen-Driven Basophils Are a Major Source of IL-4 in Human Filarial Infections J. Immunol., February 15, 2004; 172(4): 2439 - 2445. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
