See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/12105992

An unexpected version of horror autotoxicus: Anaphylactic shock to a self-

peptide

Article in Nature Immunology · April 2001

DOI: 10.1038/85266 · Source: PubMed

CITATIONS READS
185 215

9 authors, including:

Rosetta Pedotti Jochen Wedemeyer

Fondazione I.R.C.C.S. Istituto Neurologico Carlo Besta KRH, Klinikum Robert Koch Gehrden
58 PUBLICATIONS 5,406 CITATIONS 148 PUBLICATIONS 4,551 CITATIONS

SEE PROFILE SEE PROFILE

Marcela Karpuj Eyas M Hattab

Bar Ilan University University of Louisville
23 PUBLICATIONS 2,633 CITATIONS 168 PUBLICATIONS 5,802 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Eyas M Hattab on 16 October 2015.

The user has requested enhancement of the downloaded file.

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

An unexpected version of horror

autotoxicus: anaphylactic shock
to a self-peptide
© 2001 Nature Publishing Group http://immunol.nature.com

Rosetta Pedotti1,2, Dennis Mitchell1, Jochen Wedemeyer3, Marcela Karpuj1, Dorothée

Chabas1, Eyas M. Hattab3, Mindy Tsai3, Stephen J. Galli3 and Lawrence Steinman1
EAE can refer either to experimental autoimmune encephalomyelitis or experimental allergic
encephalomyelitis.Although EAE is classically a prototypic T helper 1 (TH1) cell–mediated autoimmune
disease, it can also be induced by TH2 cells. Characteristically, the most severe manifestation of allergy,
anaphylaxis, is associated with exposure to a foreign antigen that is often derived from medication,
insect venom or food.We report here that, after self-tolerance to myelin is destroyed, anaphylaxis may
be triggered by a self-antigen, in this case a myelin peptide. “Horror autotoxicus”, which was initially
described by Ehrlich, may not only include autoimmunity to self, it may also encompass immediate
hypersensitivity to self, which leads to shock and rapid death.

Experimental autoimmune encephalomyelitis (EAE) is the prototypi- aspects of the pathology of EAE or MS, including demyelinization19,20.
cal animal model for T cell autoimmunity and shares many clinical In agreement with a potentially important role for mast cells in the
and pathological features with the human disease multiple sclerosis expression of EAE, mice that are genetically deficient in mast cells
(MS). CD4+ T cells that are reactive to myelin antigens are the prima- exhibit significantly reduced incidence and severity of EAE21.
ry mediators of the disease1–3. Naturally circulating in the periphery, In the course of developing peptide-based antigen-specific therapy in
these cells must undergo activation and clonal expansion to cause EAE, we encountered animals that died very rapidly after the adminis-
EAE1. Production of T helper 1 (TH1)-associated pro-inflammatory tration of the self-antigen myelin proteolipid protein (PLP) amino acids
cytokines such as interferon γ (IFN-γ), interleukin 2 (IL-2) and tumor 139 through 151, or PLPp(139–151). We report here our investigations
necrosis factor α (TNF-α) play a critical role in the development of into the mechanisms that underlie this unexpected phenomenon.
EAE and in subsequent brain inflammation2. Antibodies are the criti-
cal mediators of allergic reactions, which, in their most extreme man- Results
ifestation, result in anaphylactic shock4,5. Immediate hypersensitivity Induction of anaphylaxis with a self-peptide
and allergy are associated with the action of a set of cytokines that In three consecutive experiments, EAE was induced with PLPp(139–
include IL-4, IL-5 and IL-13 (TH2 cytokines), which induce both a 151) in SJL/J mice. EAE occurred in 79% of mice (23 of 29) with dis-
shift in antibody isotype and eosinophilia3,5. Although in humans ease onset at day 11.7±0.14 (mean±s.e.m.) and a disease score at day 14
immunoglobulin E (IgE) is the antibody isotype that is involved in of 2.4±0.28. Mice were challenged with PLPp(139–151) 15, 21 or 28
immediate-type hypersensitivity4–6, both IgE and IgG1 can elicit aller- days after induction of EAE. In 71% of mice challenged after 4 weeks
gic reactions in mice7,8. They do this by binding to FcεRI and FcγRIII, (10 of 14) and 71% of mice challenged after 3 weeks (5 of 7), but in
respectively, on mast cells and basophils8. none of the mice challenged after 2 weeks (0 of 8), reactions that had the
TH1 responses are associated with autoimmune disease such as EAE characteristics of immediate hypersensitivity occurred within a few min-
and MS. Conversely, TH2 responses protect animals from EAE and may utes of injection (Fig. 1). Affected mice developed piloerection; pros-
ameliorate MS9. Nevertheless, adoptive transfer of myelin-reactive TH2 tration; erythema of the tail, ears and footpads; were unable to move
cells to immunodeficient mice induces EAE with an eosinophilic infil- after stimulation; and presented with dyspnea and shallow breathing.
trate in the brain10. This suggests that a TH2-type immune response could Death occurred within 30 min in 43% of the mice challenged after 28
play a role in the development of this disease. Mast cells represent major days (6 of 14) and in 14% of the mice challenged after 21 days (1 of 7).
effector cells of TH2-associated immune responses3–5, yet several lines of The nature and time-course of these symptoms were highly suggestive
evidence indicate that mast cells also may contribute to the development of anaphylactic shock4,5,8,23. In addition, as is characteristic of mice that
of EAE and participate in the pathology of MS11–21. Localized release of exhibit IgE- or IgG1-dependent anaphylaxis8,24, the mice also developed
vasoactive amines, cytokines, chemokines and other mediators from a significant drop in body temperature (Fig. 2a) and evidence of signifi-
mast cells that are found at the sites of inflammatory demyelination in cantly increased airway resistance (Fig. 2b). In addition, histopathologi-
EAE10,14 and MS15–17 may promote the entry of autoreactive T cells into cal examination of tissues obtained from autopsies of these mice showed
the central nervous system (CNS)22. They may also contribute to other extensive vascular congestion in lungs, liver and heart (Fig. 3).

Departments of 1Neurology and Neurological Science and 3Pathology, Stanford University Medical Center, Stanford, CA 94305, USA. 2Institute of Clinical Neurology,
University of Milan, IRCCS Ospedale Maggiore Policlinico, Milan, Italy. Correspondence should be addressed to L. S. ([email protected]).

216 nature immunology • volume 2 no 3 • march 2001 • http://immunol.nature.com

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

Figure 1. Challenge with soluble upon challenge at day 21 and 91% at day 28. (Reactions
PLPp(139–151) after the acute phase developed in 10 of 11 mice with elevated IgG1 versus 0
of EAE causes signs of anaphylactic of 3 mice without elevated IgG1; P=0.011 by Fisher’s
shock. In three consecutive experiments,
EAE was induced in 29 SJL/J female mice
exact test.) These findings suggest a correlation between
with subcutaneously injected PLPp(139– high titers of anti-PLPp(139–151) IgG1 and the expres-
151) in CFA. Mice were then challenged sion of immediate hypersensitivity reactions to peptide
with the same peptide (100 µg in PBS) challenge, particularly on day 28.
intraperitoneally (arrows) after 15 (8
We did not detect IgE antibodies specific for
mice), 21 (7 mice) or 28 (14 mice) days.
© 2001 Nature Publishing Group http://immunol.nature.com

The majority of mice developed allergic PLPp(139–151) with our enzyme-linked immunosorbent
symptoms and died when challenged at assay (ELISA) method. However, total IgE, which was
days 21 and 28, but not day 15. not detectable in the sera of strain-, age- and gender-
matched normal mice, was elevated in 33% of mice at day
15 (3 of 8), 29% of mice at day 21 (2 of 7) and 21% of
mice at day 28 (3 of 14) (Fig. 4c). Total IgE concentration
To confirm that anaphylactic shock was triggered specifically by did not increase during the course of EAE. Taken together, these find-
PLPp(139–151), we challenged some mice with a control peptide of 13 ings suggest that the allergic reactions that can be expressed in these
amino acids 28 days after EAE induction. The absence of allergic reac- mice after the acute phase of EAE are mediated by IgG1 antibodies.
tions, even after repeated challenges with the control peptide, suggest-
ed that the allergic response in mice with EAE was specific for Characterization of the self-allergen and its adjuvants
PLPp(139–151). (Anaphylaxis developed in 10 of 14 mice after chal- An allergic response to self-PLPp(139–151) could also be detected
lenge with this peptide versus 0 of 12 after challenge with the control when the immunization did not induce clinical EAE. Mice were immu-
peptide; P<0.0002 with Fisher’s exact test; Table 1 and Fig. 2a,b.) In nized, via the intraperitoneal (i.p.) route, with PLPp(139–151) in
addition, mass spectroscopic analysis of the peptides indicated a single incomplete Freund’s adjuvant (IFA) or in complete Freund’s adjuvant
peak at the appropriate molecular size, which rules out aggregation of (CFA). Challenge after 4 weeks caused anaphylactic shock and death in
these molecules (data not shown). 100% of the mice immunized with the peptide in IFA (5 of 5) and 75%
of the mice immunized with the peptide in CFA (6 of 8) (Table 1).
Analysis of the humoral response In many of our experiments immunization and challenge was carried
The humoral response was studied at different time-points during the out with a PLPp(139–151) amino acid sequence that included a substi-
course of EAE by analyzing serum that had been collected a few hours tution, at position 140, of the native amino acid, cysteine, with serine
before peptide challenge (Fig. 4). Titers of IgG antibody to PLPp (Cys140→Ser140). This substitution is commonly used as a “pathogenic
(139–151) increased during the course of EAE. Among these antibod- sequence” to induce EAE25. However, to evaluate whether the allergic
ies, and in all the mice, titers of IgG1 exceeded those of IgG2a (Fig. responses can be induced to the “true-self-antigen”, we also examined
4a,b). Although only one of eight mice (12.5 %) had an elevated titer mice in which EAE was induced with the peptide that contained the
of IgG1 antibodies to PLPp(139–151) at day 15, 71% (5 of 7) or 79% native cysteine at position 140, referred to hereafter as native
(11 of 14) of the mice developed high titers of this antibody isotype by PLPp(139–151). Challenge with this soluble self-peptide 3 or 4 weeks
day 21 or 28, respectively. Of the mice with elevated titers of IgG1 after the induction of EAE, elicited allergic responses in 45% of the
antibodies to PLPp(139–151), 60% (3 of 5) exhibited allergic reactions mice (13 of 29) and death in 24% (7 of 29). This indicated that allergic

Figure 2. The reactions that occur after

challenge with PLPp(139–151) are asso-
a b
ciated with a drop in body temperature
and an increase in enhanced respiratory
pause. EAE was induced in SJL/J female mice
with PLPp(139–151) or native PLPp(139–151)
3–4 weeks before challenge with the respec-
tive peptides or, as a negative control, with a
control peptide. As an additional negative con-
trol, naïve SJL/J female mice that had not been
immunized with the peptide were challenged
with PLPp(139–151). As a positive control for
IgE-dependent anaphylaxis, SJL/J female mice
that had received anti-DNP IgE 24 h before
were challenged with DNP-HSA. Data are
mean±s.e.m. values and the number of mice in
each group is shown. (a) Body temperature
was recorded at multiple intervals after the
challenge. (****P<0.001 by ANOVA versus
either negative-control group; ***P<0.005 ver-
sus either negative-control group; ††††P<0.001
versus positive-control group; †P<0.05 versus
positive-control group.) (b) The average
enhanced respiratory pause (Penh, a measure
of airway resistance) at 3 min intervals after
challenge is expressed for each time-point as a percentage of baseline Penh values (baseline is defined as 100%). (***P<0.005 by ANOVA versus either negative-control group;
*P<0.05 versus either negative-control groups, n.s., nonsignificant differences, where P>0.05 among the three indicated groups.)

http://immunol.nature.com • march 2001 • volume 2 no 3 • nature immunology 217

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

Table 1. Incidence of allergic reactions in mice challenged with various peptides

Mouse strain Peptide of primary Immunization protocol Number of mice Peptide used for Number of mice with
a
immunization adjuvant route with EAE challenge allergic reactions at the
(100 µg) challengeb
SJL/J PLPp(139–151) CFA s.c. 13 of 14 PLPp(139–151) 10 of 14e
SJL/J PLPp(139–151) CFA s.c. 11 of 12 Control peptide 0 of 12
SJL/J PLPp(139–151) CFA i.p. 1 of 8 PLPp(139–151) 6 of 8
© 2001 Nature Publishing Group http://immunol.nature.com

SJL/J PLPp(139–151) IFA i.p. 0 of 5 PLPp(139–151) 5 of 5

SJL/J Native PLPp(139–151) CFA s.c. 27 of 30 Native PLPp(139–151) 13 of 29
SJL/J PLPp(139–151) CFA s.c. 0 of 10 PLPp(139–151) 0 of 10
Leu145→Lys145 Leu145→Lys145
SJL/J MOG(35–55) CFA s.c. 0 of 10 MOG(35–55) 0 of 10
0 of 10 MOG(35–55)d 0 of 10
SJL/J PLPp(178–191) CFA s.c. 13 of 15 PLPp(178–191) 0 of 14
C57BL/6 MOG(35–55) CFA s.c. 0 of 20 MOG(35–55)d 1 of 20f, 2 of 20g, 16 of 20h
C57BL/6 PLPp(139–151) CFA s.c. 0 of 10 PLPp(139–151) 0 of 10
PL/J MBPAc(1–11) CFA s.c. 8 of 10 MBPAc(1–11)d 0 of 7
Ptx i.vc
PL/J MBPAc(1–11) CFA s.c. 0 of 10 MBPAc(1–11)d 0 of 10
a
Unless otherwise specified, challenge occurred 4 weeks after immunization with i.p. injection of 100 µg of the different peptides in PBS. Absence of allergic signs was
b

confirmed by three weekly challenges. cBordetella pertussis toxin (Ptx) was given as an adjuvant in a 200 ng dose at days 0 and 2 after immunization. dChallenge by i.p. injec-
tion with 500 µg of the peptide in PBS. eP<0.0002 by Fisher’s Exact test challenge with PLPp(139–151) versus challenge with control peptide. fChallenge at 4 weeks.
g
Challenge at 5 weeks. hChallenge at 6 weeks. (s.c., subcutaneous; i.p. intraperitoneal; i.v., intravenous.)

responses to “true-self” could still induce anaphylaxis, albeit at a some- hypersensitivity reactions when challenged with the same peptide 28
what lower frequency compared to the modified PLPp(139–151) days after immunization (Table 1).
sequence (Table 1). In addition, the key pathophysiological features of To determine whether another self-antigen could induce immediate
these reactions are strikingly similar to those observed in IgG1- hypersensitivity in another strain of mice, we explored the induction
FcγRIII–associated passive systemic anaphylaxis8,24. A marked drop in of allergic responses in H-2b mice that had been immunized with
body temperature, a significant increase in airway resistance and little myelin oligodendroglial glycoprotein (MOG). C57BL/6 (H-2b) mice,
or no morphological evidence of mast cell degranulation were observed but not SJL/J (H-2s) mice, are susceptible to EAE induced by MOG,
in mice that developed anaphylaxis in response to native which binds to I-Ab (ref. 29). In two consecutive experiments, SJL/J
PLPp(139–151) (Figs. 2a,b and 5). and C57BL/6 mice were immunized with the peptide MOG(35–55)
To induce an allergic type-TH2 response, as well as a TH1 response and 4 weeks later, at a time when self-tolerance to MOG has been
capable of eliciting EAE, a peptide antigen that is complexed with a broken and anti-MOG responses are present (data not shown), chal-
major histocompatibility (MHC) molecule must be presented to T lenged with the same peptide. MOG did not induce clinical EAE
cells23,26. In SJL/J mice (H-2s), PLPp(139–151) binds with high affinity to unless Bordetella Pertussis was used as an adjuvant. No SJL/J mice
the I-As MHC class II molecule27. PLPp(139–151)-reactive T cells are (0 of 20) developed any allergic reactions after challenge with MOG
normally found in naïve SJL/J mice, probably because of a lack of thymic (Table 1). However, in the C57BL/6 mice, 5% (1 of 20) developed
deletion of these autoreactive T cells28. The residues at positions 145 anaphylaxis after the first challenge. This increased to 10% (2 of 20)
(leucine) and 148 (proline) of PLPp(139–151) are important for MHC after the second injection and to 80% (16 of 20) after the third chal-
class II (I-As)-binding. The Leu145→Lys145 analog of PLPp(139–151) is lenge with a 0.5 mg dose of the peptide. Of the mice with anaphylax-
unable to bind to I-As and to stimulate T cells hybridomas specific to is, 25% died (5 of 20).
PLPp(139–151)25. As might be expected, mice immunized with the Compared to SJL/J mice, C57BL/6 are less sensitive to the effects of
Leu145→Lys145 analog of PLPp(139–151) did not express immediate histamine and perhaps to other mast cell–derived mediators13,18,30. Strain

a b c

Figure 3. Histopathology of tissue sections obtained from a 12-week-old SJL/J mouse that was killed when it became moribund 20 min after apparent
anaphylaxis. EAE had been induced, with PLPp(139–151), 4 weeks previously.The mouse was then challenged with 100 µg of i.p. PLPp(139–151).There was extensive vas-
cular congestion in all tissues examined.Tissues were stained with hematoxylin and eosin and are shown at magnification ×300. (a) Marked pulmonary congestion and focal
emphysematous changes in the lung. (b) Sinusoidal as well as vascular congestion in a liver section. (c) Vascular congestion in cardiac muscle.

218 nature immunology • volume 2 no 3 • march 2001 • http://immunol.nature.com

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

Figure 4.Titers of anti-PLPp(139–151)

of the IgG1 isotype progressively
increased during the course of EAE
a b c
compared to IgG2a and to total IgE.
Sera were collected from mice at different
time-points after EAE induction, 2–3 h
before challenge with PLPp(139–151).
Antibodies to PLPp(139–151) of isotypes
(a) IgG1 and (b) IgG2a as well as (c) total
IgE were measured by ELISA. Each data
© 2001 Nature Publishing Group http://immunol.nature.com

point represents the antibody titer of an

individual mouse, at the time-point indicat-
ed, at a dilution of 1:100 for IgG1 and IgG2a
and of 1:20 for IgE. (nms, normal mouse
serum.)

variation in these mice could explain why anaphylactic shock appeared whether there was a correlation between the ability a self-antigen to
in C57BL/6 mice only after repeated challenges with a higher dose of induce thymic tolerance and the ability to induce anaphylaxis. To eval-
MOG(35–55). Nevertheless, these results clearly show that another uate this possibility, we tested two self-antigens that are expressed in
self-peptide, MOG(35–55), was able to induce immediate hypersensi- thymus and induce some thymic tolerance. PLPp(178–191) is expressed
tivity in a mouse strain (C57BL/6) that can present this antigen to the in the thymi of SJL/J mice, where it induces some degree of thymic tol-
immune system. Immunization of C57BL/6 mice with PLPp(139–151) erance28. We found that although PLPp(178–191) binds to I-As and
in CFA did not induce immediate hypersensitivity, as it did in SJL/J induces EAE, it did not induce anaphylaxis (Table 1 and Fig. 6).
mice, as shown by the absence of allergic reactions (0 of 10 mice) after Similarly myelin basic protein (MBP) acetylated 1–11 peptide,
challenge with repeated doses of PLPp(139–151). These experiments MBPAc(1–11) is expressed in the thymi of PL/J mice31,32. With peptide-
also rule out the possibility that anaphylaxis developed because of MHC multimers it has been shown that MBPAc(1–11)–specific T cells
some contaminant in the antigen preparation. This is because are below the limit of detection in PL/J mice, which suggests that they
PLPp(139–151), though able to induce anaphylaxis in SJL/J mice, was have been deleted by thymic selection33. However, after immunization
not able to do so in C57BL/6 mice. with MBPAc(1–11), some residual T cells undergo rapid expansion and
can be easily detected in the periphery. T cell interaction with
Role of thymic tolerance in allergy to self
The peptides tested, PLPp(139–151) and MOG(35–55), are self-anti-
gens that are found in the myelin sheath. PLPp(139–151) is not
a
expressed in the thymus and thus does not produce thymic, or central,
tolerance. There is no evidence that MOG is expressed in the thymus of
mice (C. Bernard, personal communication). We therefore considered

Figure 5. Mice with EAE that develop anaphylactic shock after challenge
with native PLPp(139–151) exhibit a much lower mast cell degranulation
compared to mice that developed IgE-dependent anaphylaxis. SJL/J female
mice in which EAE had been induced 3 weeks previously with native PLPp(139–151) Figure 6. Absence of allergic reactions after challenge with MBPAc(1–11)
were challenged via the i.p. route with this same peptide (EAE PLP139–151) or with or with PLPp(178–191) is confirmed by the absence of changes in body
a control peptide (EAE control peptide). For IgE-dependent anaphylaxis, SJL/J female temperature. (a) PL/J female mice immunized with MBPAc(1–11) with (Ptx) or
mice that had received anti-DNP IgE 24 h previously were challenged with DNP-HSA without (no Ptx) pertussis toxin (which was used as an adjuvant) were challenged
(IgE anti-DNP DNP-HSA). As control, tissues from strain-, age- and gender-matched with MBPAc(1–11) or with a control peptide. (b) SJL/J female mice immunized with
naïve mice were collected (control). Epon-embedded Giemsa-stained sections that PLPp(178–191) were challenged with PLPp(178–191) or with a control peptide. (a,b)
were 1 µm thick were examined to assess the extent of mast cell degranulation in Data were obtained from measurement of body temperature during the third week-
the tissues (ears). Data are the mean±s.e.m. values from five mice per group. ly challenge and the number of mice in each group are shown. (n.s, nonsignificant dif-
(***P<0.001 by the χ2 test data from IgE–anti-DNP–sensitized, DNP-challenged mice ferences, where P>0.05, as shown by ANOVA tests comparing challenge with myelin-
versus data from any of the other groups of mice.) specific peptides versus challenge with control peptide.)

http://immunol.nature.com • march 2001 • volume 2 no 3 • nature immunology 219

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

preparations required purification by affinity

Table 2.Treatment of EAE with cyproheptadine
chromatography with a mouse-derived mono-
Treatment Incidence Disease score Disease onset Peak clonal antibody to factor VIII or to von
(%) (at day 13)a (day)a disease severitya Willebrand factor42,43. The reactions to factor
Cyproheptadine b
80 (8/10) c
0.8±0.4 d
13±0.8 e
1.8±0.4 f VIII may have been due to a human anti-
PBS 100 (10/10) 2.4±0.4 11.3±0.5 3.1±0.3 mouse response or to preexistent “natural”
a
antibodies to mouse44. In addition, in subjects
Data shown as mean± s.e.m values. bCyproheptadine was administered i.p., 15 mg/kg/day in 0.5 ml of PBS.
c
Numbers in parentheses are the number of sick animals of the total tested. dP=0.01, eP=0.05, fP=0.03. All P val- that are deficient in factor VIII, the factor VIII
© 2001 Nature Publishing Group http://immunol.nature.com

ues are in comparison to the PBS-treated group as assesed by Student’s t test. itself may have been “foreign”.
Our findings show that allergic responses,
which are typically TH2- mediated, can be
MBPAc(1–11) is a low-avidity interaction that permits escape from cen- induced in mice with EAE, despite the well known TH1 bias of this dis-
tral tolerance34–36. Immunization with MBPAc(1–11) effectively trig- ease. Indeed, even CFA, which is well known to polarize the immune
gered EAE but did not induce anaphylaxis (Table 1 and Fig. 6). Thus, response towards TH1, could be used to induce immediate hypersensitiv-
with two examples of self-antigens, PLPp(178–191) and MBPAc(1–11), ity. As an aside, a paucity of allergic disease in MS patients, compared to
which are expressed in thymus and do induce some degree of thymic tol- a control cohort, has been reported45. We found that allergic responses
erance, there appears to be resistance to the induction of anaphylaxis, could not be elicited 2 weeks after EAE induction, during the peak of dis-
even after self-tolerance is broken and EAE is induced. ease. Instead, they appeared only after 3 weeks, during the recovery from
the acute attack or at about the time of the first relapse, when TH2
Potential role of histamine and serotonin responses are known to occur38. This suggests that immediate hypersen-
Suppression of EAE with inhibitors of mast cell degranulation or with sitivity reactions that are strong enough to produce anaphylaxis general-
antagonists of histamine and serotonin (which can be derived from ly occur late in the course of this EAE model. On the other hand, the find-
mast cells or other sources) has been reported in a passive EAE model ings that treatment with cyproheptadine could significantly reduce mean
in rats12 and in an active EAE model in mice, where Bordetella pertus- disease score at day 13, reduce mean disease severity and delay mean day
sis was used as an adjuvant11. We decided to investigate the potential of disease onset raise the possibility that allergic mechanisms may pro-
involvement of histamine or serotonin in the development of mote the development and severity of EAE. Nevertheless, cyprohepta-
PLPp(139–151)–induced EAE in our model (in which Bordetella per- dine treatment may have effects on EAE that are unrelated to allergic
tussis was not used). Ten SJL/J mice were treated daily with cyprohep- mechanisms, for example, the inhibition of actions of histamine or sero-
tadine, an anti-histamine–anti-serotonin drug. Amelioration of acute tonine that is derived from sources other than mast cells.
clinical disease was observed in the cyproheptadine-treated group, In addition to their implications for understanding the pathogenesis
compared with the control PBS-treated group (Table 2). The onset of of EAE in mice, our findings offer a cautionary tale with regard to
disease was delayed compared to the control PBS group (P<0.05) and efforts to develop immune therapies for MS that aim to shift a patho-
mean peak disease-severity (P<0.03) and mean disease score (P<0.01) genic TH1 response towards TH2 immunity. Indeed, in a recent phase II
were reduced. clinical trial in which MS patients received an altered peptide ligand
(APL) from an epitope of myelin basic protein, repeated injections of
Discussion the APL induced immediate hypersensitivity reactions in 9% of the
The findings reported here show that, after recovery from EAE, mice subjects46. Although none of these patients exhibited anaphylaxis, the
can exhibit severe allergic reactions, including fatal anaphylaxis, to a findings support our data in that TH2 immunity to self or “altered self”
self-myelin peptide. The mechanisms that account for the development molecules may produce a spectrum of undesirable reactions, including
of such responsiveness are not fully understood. It has been reported allergy46 and, in marmoset EAE, a worsening of disease47. In MS
that, during the course of EAE, there is a significant decline in TH1-type patients, use of APLs may reduce disease activity on magnetic reso-
cytokine secretion, although TH2-type cytokine secretion rises37. IL-4, nance (MR) scans. Allergic reactions were not associated with a wors-
in particular, peaks during the remission of EAE38. We hypothesize that ening of MS46.
the elevated secretion of IL-4 during the remission phase of EAE pro- Myelin antigens, including PLP, are expressed in the thymus of nor-
motes the production of IgG1 antibodies to PLPp(139–151) that capa- mal mice during development and tolerance to them may develop at this
ble of mediating anaphylaxis39. site28. It is worth noting that only the DM-20 isoform of PLP is
Allergy is classically defined as an immunological reaction to a for- expressed in the thymus of SJL/J mice and the PLPp(139–151) fragment
eign antigen5. Immediate hypersensitivity reactions have been report- is expressed in the white matter of the brain. During EAE an immune
ed in response to the administration of peptide competitors, which response to this epitope, which lies outside of the region encoded by
contain some D–amino acids, for I-A in the treatment of insulin-depen- DM-2028,48,49, is elicited. Hence tolerance to this epitope may not have
dent diabetes in the nonobese diabetic (NOD) mouse26. The peptides
used, however, were not self-peptides. There are reports of anaphylax-
in response to “self-antigens” such as insulin and factor VIII. Table 3. Myelin antigens that cause anaphylaxis
However, in each of these reports, it appears that a nonself molecule
Antigens Causes Expressed in Causes
may actually have triggered anaphylaxis. Anaphylaxis in response to
EAE thymus anaphylaxis
human insulin preparations that had been modified by protamine have
been reported40,41. However, radioallergosorbent tests revealed that the PLPp(139–151) Yes No Yes
MOG(35–55) Yes No Yes
IgE reaction was directed to protamine and not to insulin40,41. PLPp(178–191) Yes Yes No
Anaphylaxis has also been reported in response to factor VIII, admin- MBPAc(1–11) Yes Yes No
istered as either the purified or the recombinant protein. But both

220 nature immunology • volume 2 no 3 • march 2001 • http://immunol.nature.com

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

developed in the thymus, which may explain why profound allergic Elicitation of passive systemic anaphylaxis. Mice received an i.p. injection of anti-DNP
reactions can develop to this self-constituent49. In support of this hypoth- IgE (20 µg) dissolved in HMEM (200 µl, Gibco-BRL, Gaithersburg, MD) that contained
PIPES buffer (0.47 g/l, Sigma) and were challenged 24 h later by an i.v. injection of DNP-
esis, two antigensMBPAc (1–11) and PLPp(178–191), which are ex- HSA (200 µg, Sigma) in saline (200 µl)8.
pressed in the thymus and induce central tolerancedid not induce ana-
phylaxis, even after self-tolerance was broken and EAE was induced Determination of airway responsiveness. Airway responses were assessed using a single-
chamber whole-body plethysmograph (Buxco, Troy, NY) as described55. Enhanced respira-
(Table 3). tory pause (Penh) was used as the measure of airway resistance. Mice were placed individ-
Ehrlich’s “horror autotoxicus” may thus involve allergy to self, as ually in the plethysmograph, baseline readings were then taken and averaged for 10 min.
well as autoimmunity to self-antigens that escape thymic tolerance. After challenge with various peptides or i.p. DNP-HSA, readings were taken and averaged
every 3 min for 30 min. For each time point the Penh values measured during each 3-min
© 2001 Nature Publishing Group http://immunol.nature.com

Recognition of self-molecules may lead therefore to organ-specific sequence are expressed as a percentage of baseline Penh values.
autoimmunity, systemic autoimmunity, protective autoimmunity50 and
allergy. Temperature measurement. Baseline temperature was established for each animal using a
rectal probe (Physitemp, Clifton, NJ). Rectal temperature was recorded in each mouse 5 min
after each challenge with i.p. peptide or i.v. DNP-HSA.
Methods
Immunization protocol. EAE was induced in 8- to 12-week-old SJL/J female mice (The Acknowledgements
Jackson Laboratory, Bar Harbor, ME) with PLPp(139–151) as described51 but without the We thank Z-S.Wang for technical assistance and H. McDevitt,T. Staehelin, A. Pedotti, P.
use of Bordetella pertussis toxin. Mice were scored as follows: O, healthy; 1, tail weakness Decamilli, P. Ghezzi and L. Stark for critical reading of the manuscript. Supported (in part)
or paralysis; 2, paraparesis (incomplete paralysis of one or two hind limbs or plegia of one by a postdoctoral fellowship from the National Multiple Sclerosis Society (to R. P.) and
hind limb); 3, paraplegia extending to the thorax; 4, forelimb weakness, paralysis with hind support from the National Institutes of Health and the Phil N. Allen Fund.
limb paraparesis or paraplegia; 5, moribund or dead animal. Blood was collected from the
tails of mice 2, 3 and 4 weeks after immunization and analyzed for antibody responses.
Received 7 December 2000; accepted 19 January 2001.
Mice were challenged 2, 3 or 4 weeks after i.p. immunization with 100 µg of
PLPp(139–151) in PBS and observed for 1 h to evaluate symptoms of hypersensitivity. The 1. Vandenbark, A. A. et al. Specificity of T lymphocyte lines for peptides of myelin basic protein. J.
same protocol was used to immunize and challenge SJL/J, C57BL/6 and PL/J mice (The Immunol. 135, 229–233 (1985).
2. Brosnan, C. F., Selmaj, K. & Raine, C. S. Hypothesis: a role for tumor necrosis factor in immune-medi-
Jackson Laboratory) with the different myelin, analog or control peptides. In PL/J mice i.v.
ated demyelination and its relevance to multiple sclerosis. J. Neuroimmunol. 18, 87–94 (1988).
Bordetella pertussis toxin (200 ng) was given as an adjuvant at the moment of the immu- 3. Abbas, A. K., Murphy, K. M. & Sher, A. Functional diversity of helper T lymphocytes. Nature 383,
nization, as well as 2 days later, to induce EAE52. To evaluate the efficacy of EAE treatment 787–793 (1996).
with anti-histamine and/or anti-serotonin drugs, cyproheptadine (Sigma, St Louis, MO) at a 4. Bochner, B. S. & Lichtenstein, L. M. Anaphylaxis. N. Engl. J. Med. 324, 1785–1790 (1991).
dose of 15 mg/kg was injected daily via the i.p. route in 0.5 ml of PBS. Injections started 5. Galli, S. J. & Lantz, C. S. in Fundamental Immunology (ed. Paul,W. E.) 1137–1184 (Lippincott-Raven
on day 2 after EAE was induced in SJL/J mice with PLPp(139–151). All animal protocols Press, Philadelphia, 1999).
6. Ishizaka,T. & Ishizaka, K. Activation of mast cells for mediator release through IgE receptors. Prog.
were approved by the Division of Laboratory Medicine at Stanford and conformed to
Allergy 34, 188–235 (1984).
National Institutes of Health guidelines. 7. Mota, I. & Wong, D. Homologous and heterologous passive cutaneous anaphylactic activity of mouse
antisera during the course of immunization. Life Sci. 8, 813–820 (1969).
Measurement of peptide-specific serum Ig responses. Peptide-specific IgG1, IgG2a and 8. Miyajima, I. et al. Systemic anaphylaxis in the mouse can be mediated largely through IgG1 and
IgE antibodies were measured in mice sera samples by ELISA. Briefly, for IgG1 and IgG2a FcγRIII. Assessment of the cardiopulmonary changes, mast cell degranulation, and death associated
ELISA, 96-wells microtiter plates (Immunolon, Dynex, Chantilly, VA) were coated with active or IgE- or IgG1-dependent passive anaphylaxis. J. Clin. Invest. 99, 901–914 (1997).
9. Liblau, R. S., Singer, S. M. & McDevitt, H. O.Th1 and Th2 CD4+ T cells in the pathogenesis of organ-
overnight at 4 °C with PLPp(139–151) (0.1 ml) that was diluted in NaHCO3 buffer (0.1 M
specific autoimmune diseases. Immunol.Today 16, 34–38 (1995).
at pH 8.0) at a concentration of 0.010 mg/ml. The plates were blocked with 10% fetal calf 10. Lafaille, J. J. et al. Myelin basic protein-specific T helper 2 (Th2) cells cause experimental autoimmune
serum in PBS buffer for 2 h. Samples were diluted in blocking buffer to 1:100 for IgG1 and encephalomyelitis in immunodeficient hosts rather than protect them from the disease. J. Exp. Med.
IgG2a ELISA and to 1:20 for IgE ELISA, then incubated for 2 h at room temperature. 186, 307–312 (1997).
Antibody binding was tested by the addition of alkaline phosphatase–conjugated mono- 11. Linthicum, D. S. Development of acute autoimmune encephalomyelitis in mice: factors regulating the
clonal goat anti–mouse IgG1, IgG2a or IgE (Southern Biotechnology Associates, effector phase of the disease. Immunobiology 162, 211–220 (1982).
12. Dietsch, G. N. & Hinrichs, D. J.The role of mast cells in the elicitation of experimental allergic
Birmingham, AL), each at a 1:1000 dilution in blocking buffer. Enzyme substrate was added
encephalomyelitis. J. Immunol. 142, 1476–1481 (1989).
and plates were read at 405 nm on a microplate reader. Antibody titer was considered ele- 13. Bebo, B. F. Jr, Lee, C. H., Orr, E. L. & Linthicum, D. S. Mast cell-derived histamine and tumour necrosis
vated if optical density>0.1. factor: differences between SJL/J and BALB/c inbred strains of mice. Immunol. Cell. Biol. 74, 225–230
(1996).
Measurement of total IgE production. Total IgE was measured by sandwich ELISA 14. Brenner,T., Soffer, D., Shalit, M. & Levi-Schaffer, F. Mast cells in experimental allergic encephalomyelitis:
(PharMingen, San Diego, CA) according to the manufacturer’s instructions53. Serum sam- characterization, distribution in the CNS and in vitro activation by myelin basic protein and neu-
ropeptides. J. Neurol. Sci. 122, 210–213 (1994).
ples were analyzed in duplicate at a 1:20 dilution.
15. Toms, R.,Weiner, H. L. & Johnson, D. Identification of IgE-positive cells and mast cells in frozen sec-
tions of multiple sclerosis brains. J. Neuroimmunol. 30, 169–177 (1990).
Antigens. Peptides were synthesized on a model 9050 peptide synthesizer (Milligen, 16. Olsson,Y. Mast cells in plaques of multiple sclerosis. Acta Neurol. Scand. 50, 611–618 (1974).
Burlington, MA) by standard 9-fluorenylmethoxycarbonyl chemistry. Peptides were purified 17. Kruger, P. G. et al. Mast cells and multiple sclerosis: a light and electron microscopic study of mast
by HPLC and tested for endotoxin with a limulus amoebocyte lysate qualitative test cells in multiple sclerosis emphasizing staining procedures. Acta Neurol. Scand. 81, 31–36 (1990).
(BioWhittaker, Walkersville, MD). Endotoxin concentrations in all preparations were always 18. Linthicum, D. S. & Frelinger, J. A. Acute autoimmune encephalomyelitis in mice. II. Susceptibility is
controlled by the combination of H-2 and histamine sensitization genes. J. Exp. Med. 156, 31–40
<0.03 endotoxin units (EU)/ml. Size was confirmed by mass spectroscopy, which also indi-
(1982).
cated that no aggregates had formed. Peptides used for experiments were: PLPp(139–151), 19. Johnson, D., Seeldrayers, P. A. & Weiner, H. L.The role of mast cells in demyelination. 1. Myelin pro-
HSLGKWLGHPDKF; native PLPp(139–151), HCLGKWLGHPDKF; MOG(35–55), teins are degraded by mast cell proteases and myelin basic protein and P2 can stimulate mast cell
MEVGWYRSPFSRVVHLYRNGK; PLPp(178–191), NTWTTCQSIAFPSK; MBPAc(1–11), degranulation. Brain Res. 444, 195–198 (1988).
Ac-ASQKRPSQRHG; Leu145→Lys145 analog of PLPp(139–151), HSLGKWKGHPDKF; and 20. Dines, K. C. & Powell, H. C. Mast cell interactions with the nervous system: relationship to mecha-
control peptide, CAVPLKTQMSGSSFM. nisms of disease. J. Neuropathol. Exp. Neurol. 56, 627–640 (1997).
21. Secor,V. H., Secor,W. E., Gutekunst, C. A. & Brown, M. A. Mast cells are essential for early onset and
severe disease in a murine model of multiple sclerosis. J. Exp. Med. 191, 813–822 (2000).
Pathological studies. For routine histopathological studies, the lungs, liver and heart of 22. Linthicum, D. S., Munoz, J. J. & Blaskett, A. Acute experimental autoimmune encephalomyelitis in
mice that died after antigen challenge were stored in 10% formalin for 1 week. Sections mice. I. Adjuvant action of Bordetella pertussis is due to vasoactive amine sensitization and increased
were then obtained and stained with hematoxylin and eosin. For the quantification of mast vascular permeability of the central nervous system. Cell. Immunol. 73, 299–310 (1982).
cell degranulation, the presence of mast cells and their state of degranulation were assessed 23. Soloway, P. et al. Regulation of the immune response to peptide antigens: differential induction of
in 1-µm, Epon-embedded, Giemsa-stained sections54. Tissues were removed and fixed, as immediate-type hypersensitivity and T cell proliferation due to changes in either peptide structure
or major histocompatibility complex haplotype. J. Exp. Med. 174, 847–858 (1991).
described54, immediately after death. Death was induced by antigen challenge with the
24. Dombrowicz, D. et al. Absence of FcεRI α chain results in upregulation of FcγRIII-dependent mast
native PLPp(139–151) or IgE anti-DNP + DNP-HSA or, in mice that received control pep- cell degranulation and anaphylaxis. Evidence of competition between FcεRI and FcγRIII for limiting
tide or survived challenge with the native PLPp(139–151) or dinitrophenyl-conjugated amounts of FcR β and γ chains. J. Clin. Invest. 99, 915–925 (1997).
human serum albumin (DNP-HSA), by CO2 inhalation. As a normal control, tissues from 25. Kuchroo,V. K. et al. A single TCR antagonist peptide inhibits experimental allergic encephalomyelitis
strain-, age- and gender-matched normal mice killed by CO2 inhalation were also collected. mediated by a diverse T cell repertoire. J. Immunol. 153, 3326–3336 (1994).
Sections of the ear skin dermis were examined by an observer who was unaware of the iden- 26. Hurtenbach, U., Lier, E., Adorini, L. & Nagy, Z. A. Prevention of autoimmune diabetes in non-obese
diabetic mice by treatment with a class II major histocompatibility complex-blocking peptide. J. Exp.
tity of individual sections and evaluated for mast cell number and the extent of mast cell Med. 177, 1499–1504 (1993).
degranulation as described8,54. Mast cells were classified as extensively degranulated, mod- 27. Tuohy,V. K., Lu, Z., Sobel, R. A., Laursen, R. A. & Lees, M. B. Identification of an encephalitogenic
erately degranulated or normal as described8,54. determinant of myelin proteolipid protein for SJL mice. J. Immunol. 142, 1523–1527 (1989).

http://immunol.nature.com • march 2001 • volume 2 no 3 • nature immunology 221

 © 2001 Nature Publishing Group http://immunol.nature.com
A RTICLES

28. Anderson, A. C. et al. High frequency of autoreactive myelin proteolipid protein-specific T cells in Transfusion 28, 603 (1988).
the periphery of naive mice: mechanisms of selection of the self-reactive repertoire. J. Exp. Med. 191, 43. Shopnick, R. I. et al. Anaphylaxis after treatment with recombinant factor VIII. Transfusion 36, 358–361
761–770 (2000). (1996).
29. Encinas, J. A.,Weiner, H. L. & Kuchroo,V. K. Inheritance of susceptibility to experimental autoimmune 44. Davis, H. M. et al. Lack of immune response to mouse IgG in hemophilia A patients treated chroni-
encephalomyelitis. J. Neurosci. Res. 45, 655–669 (1996). cally with Monoclate, a monoclonal antibody affinity purified factor VIII preparation. Thromb.
30. Steinman, L. et al. Murine model for pertussis vaccine encephalopathy: linkage to H-2. Nature. 299, Haemost. 63, 386–391 (1990).
738–740 (1982). 45. Oro, A. S., Guarino,T. J., Driver, R., Steinman, L. & Umetsu, D.T. Regulation of disease susceptibility:
31. Pribyl,T. M. et al. The human myelin basic protein gene is included within a 179-kilobase transcrip- decreased prevalence of IgE- mediated allergic disease in patients with multiple sclerosis. J. Allergy
tion unit: expression in the immune and central nervous systems. Proc. Natl Acad. Sci. USA 90, Clin. Immunol. 97, 1402–1408 (1996).
10695–10699 (1993). 46. Kappos, L. et al. Induction of a non-encephalitogenic type 2 T helper-cell autoimmune response in
32. Zelenika, D., Grima, B. & Pessac, B. A new family of transcripts of the myelin basic protein gene: multiple sclerosis after administration of an altered peptide ligand in a placebo-controlled, random-
expression in brain and in immune system. J. Neurochem. 60, 1574–1577 (1993). ized phase II trial. Nature Med. 6, 1176–1182 (2000).
© 2001 Nature Publishing Group http://immunol.nature.com

33. Radu, C. G., Anderton, S. M., Firan, M.,Wraith, D. C. & Ward, E. S. Detection of autoreactive T cells in 47. Genain, C. P. et al. Late complications of immune deviation therapy in a nonhuman primate. Science
H-2(u) mice using peptide-MHC multimers. Int. Immunol. 12, 1553–1560 (2000). 274, 2054–2057 (1996).
34. Liu, G.Y. et al. Low avidity recognition of self-antigen by T cells permits escape from central toler- 48. Klein, L., Klugmann, M., Nave, K. A. & Kyewski, B. Shaping of the autoreactive T-cell repertoire by a
ance. Immunity. 3, 407–415 (1995). splice variant of self protein expressed in thymic epithelial cells. Nature Med. 6, 56–61 (2000).
35. Zamvil, S. et al. T-cell clones specific for myelin basic protein induce chronic relapsing paralysis and 49. Steinman, L. Multiple approaches to multiple sclerosis. Nature Med. 6, 15–16 (2000).
demyelination. Nature 317, 355–358 (1985). 50. Schwartz, M. & Cohen, I. R. Autoimmunity can benefit self-maintenance. Immunol.Today 21, 265–268
36. Zamvil, S. S. et al. T-cell epitope of the autoantigen myelin basic protein that induces (2000).
encephalomyelitis. Nature 324, 258–260 (1986). 51. Ruiz, P. J. et al. Suppressive immunization with DNA encoding a self-peptide prevents autoimmune
37. Jensen, M. A., Dayal, A. & Arnason, B. G. Cytokine secretion by δγ and αβ T cells in monophasic disease: modulation of T cell costimulation. J. Immunol. 162, 3336–3341 (1999).
experimental autoimmune encephalomyelitis. J. Autoimmun. 12, 73–80 (1999). 52. Waisman, A. et al. Suppressive vaccination with DNA encoding a variable region gene of the T-cell
38. Begolka,W. S.,Vanderlugt, C. L., Rahbe, S. M. & Miller, S. D. Differential expression of inflammatory receptor prevents autoimmune encephalomyelitis and activates Th2 immunity. Nature Med. 2,
cytokines parallels progression of central nervous system pathology in two clinically distinct models 899–905 (1996).
of multiple sclerosis. J. Immunol. 161, 4437–4446 (1998). 53. Spergel, J. M. et al. Epicutaneous sensitization with protein antigen induces localized allergic dermati-
39. Faquim-Mauro, E. L., Coffman, R. L., Abrahamsohn, I. A. & Macedo, M. S. Cutting edge: mouse IgG1 tis and hyperresponsiveness to methacholine after single exposure to aerosolized antigen in mice. J.
antibodies comprise two functionally distinct types that are differentially regulated by IL-4 and IL-12. Clin. Invest. 101, 1614–1622 (1998).
J. Immunol. 163, 3572–3576 (1999). 54. Wershil, B. K., Murakami,T. & Galli, S. J. Mast cell-dependent amplification of an immunologically non-
40. Kim, R. Anaphylaxis to protamine masquerading as an insulin allergy. Del. Med. J. 65, 17–23 (1993). specific inflammatory response. Mast cells are required for the full expression of cutaneous acute
41. Schernthaner, G. Immunogenicity and allergenic potential of animal and human insulins. Diabetes Care inflammation induced by phorbol 12-myristate 13- acetate. J. Immunol. 140, 2356–2360 (1988).
16, 155–165 (1993). 55. Williams, C. M. & Galli, S. J. Mast cells can amplify airway reactivity and features of chronic inflamma-
42. Bove, J. R. & McIntosh, S. Anaphylactic reaction to purified anti-hemophilic factor concentrate. tion in an asthma model in mice. J. Exp. Med. 192, 455–462 (2000).

222 nature immunology • volume 2 no 3 • march 2001 • http://immunol.nature.com

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

View publication stats