IgE antibodies, FcεRIα, and IgE-mediated local anaphylaxis can limit snake venom toxicity

IgE antibodies, FcεRIa, and IgE-mediated local anaphylaxis can limit snake venom toxicity Philipp Starkl, PhD,a* Thomas Marichal, DVM, PhD,a* Nicolas Gaudenzio, PhD,a Laurent Lionel Reber, PhD,a Riccardo Sibilano, PhD,a Mindy Tsai, DMSc,a and Stephen Joseph Galli, MDa,b Stanford, Calif Background: Type 2 cytokine–related immune responses associated with development of antigen-specific IgE antibodies can contribute to pathology in patients with allergic diseases and to fatal anaphylaxis. However, recent findings in mice indicate that IgE also can enhance defense against honeybee venom. Objective: We tested whether IgE antibodies, IgE-dependent effector mechanisms, and a local anaphylactic reaction to an unrelated antigen can enhance defense against Russell viper venom (RVV) and determined whether such responses can be influenced by immunization protocol or mouse strain. Methods: We compared the resistance of RVV-immunized wild-type, IgE-deficient, and Fcer1a-deficient mice after injection of a potentially lethal dose of RVV. Results: A single prior exposure to RVV enhanced the ability of wild-type mice, but not mice lacking IgE or functional FcεRI, to survive challenge with a potentially lethal amount of RVV. Moreover, IgE-dependent local passive cutaneous anaphylaxis in response to challenge with an antigen not naturally present in RVV significantly enhanced resistance to the venom. Finally, we observed different effects on resistance to RVV or honeybee venom in BALB/c versus C57BL/6 mice that had received a second exposure to that venom before challenge with a high dose of that venom. Conclusion: These observations illustrate the potential benefit of IgE-dependent effector mechanisms in acquired host defense against venoms. The extent to which type 2 immune responses

From the Departments of aPathology and bMicrobiology & Immunology, Stanford University School of Medicine. *These authors contributed equally to this work. P.S. was supported by a Max Kade Fellowship of the Max Kade Foundation and the Austrian Academy of Sciences and a Schroedinger Fellowship of the Austrian Science Fund (FWF): J3399-B21. T.M. was supported by a Marie Curie International Outgoing Fellowship for Career Development (grant no. 299954). N.G. was supported by fellowships from the French ‘‘Fondation pour la Recherche Medicale FRM.’’ L.L.R. was supported by the Arthritis National Research Foundation (ANRF) and National Institutes of Health (NIH) grant K99AI110645. R.S. was supported by a fellowship from the Lucile Packard Foundation for Children’s Health and the Stanford NIH/NCRR CTSA (award no. UL1 RR025744). This work was supported by NIH grants AI023990, CA072074, and AI070813 (to S.J.G.) and the Department of Pathology of Stanford University. Disclosure of potential conflict of interest: P. Starkl has received research support from the Austrian Science Fund (FWF) and Max Kade Foundation/Austrian Academy of Sciences. S. J. Galli has received research support from the National Institutes of Health and Stanford University. The rest of the authors declare that they have no relevant conflicts of interest. Received for publication November 4, 2014; revised June 25, 2015; accepted for publication August 5, 2015. Corresponding author: Stephen Joseph Galli, MD, Stanford University School of Medicine, Department of Pathology, 300 Pasteur Dr L-235, Stanford, CA 94305-5324. E-mail: [email protected]. 0091-6749/$36.00 Ó 2015 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaci.2015.08.005

against venoms can decrease pathology associated with envenomation seems to be influenced by the type of venom, the frequency of venom exposure, and the genetic background of the host. (J Allergy Clin Immunol 2015;nnn:nnn-nnn.) Key words: Acquired resistance, allergy, Daboia russelii, Russell viper, FcεRIa, honeybee, IgE, toxin hypothesis, venom, mast cells, type 2 immunity

Venoms are complex mixtures of toxic molecules1-3 used by many different animal species to fulfill functions of deterrence, defense, and/or predation.4,5 Millions of years of coevolution with venomous animals have allowed certain mammals, including those that eat or are the prey of venomous creatures, to develop innate defense mechanisms that can increase their basal (or ‘‘innate’’) resistance against venoms and their toxins. Such specialized defense strategies include producing circulating serum proteins that efficiently neutralize venom components6,7 and conserving mutations, such as those in proteins targeted by toxins,8 which confer increased resistance to that toxin. We and others have been interested in the possibility that mast cells (MCs) can represent an important component of the innate defense of vertebrates to animal venoms. MCs populate virtually all vascularized mammalian tissues.9-11 When appropriately activated, MCs can release cytoplasmic granules containing a broad spectrum of preformed mediators into the surrounding tissues.11 Notably, many components of animal venoms can induce such MC degranulation,12 and some mediators stored in MC granules have the ability to neutralize the toxicity of components of animal venoms.13-17 Higginbotham and Karnella15 showed that venom of the honeybee and the highly poisonous18,19 Russell viper14 can induce degranulation of mouse MCs in vivo14,15 and that the toxicity of those venoms was significantly reduced on their ex vivo incubation with heparin, the serglycin proteoglycan stored in MC cytoplasmic granules.20 More recently, mice deficient in MCs or certain MC-associated proteases were used to show that MCs can importantly contribute to innate host defense against venoms13,16,21 or toxic venom components13,16,17 of honeybees,16,21 2 scorpions,13 and various reptiles.13,16,17 Because IgE antibodies can enhance MC sensitivity and responsiveness against specific antigens and in light of evidence that MCs can enhance innate resistance to venoms,13,16,17,21 Profet,22 Metz et al,16 and Palm et al23 speculated that IgE antibodies might also play a protective role in acquired resistance to venoms. However, it is well known that human subjects and other mammals with IgE antibodies to venom components from honeybees,2,24 reptiles,25-29 or other animals30-33 can exhibit anaphylaxis, a catastrophic and potentially fatal acute allergic reaction, on subsequent venom exposure.34,35 Such observations suggested that the development of an acquired TH2 (or type 2) immune response and associated IgE directed against venom 1

2 STARKL ET AL

Abbreviations used BMCMC: Bone marrow–derived cultured mast cell BV: Honeybee venom bvPLA2: Honeybee venom phospholipase A2 DNP: Dinitrophenol DNP-HSA: Dinitrophenol-conjugated human serum albumin MC: Mast cell OVA: Ovalbumin PAF: Platelet-activating factor RVV: Russell viper venom WT: Wild-type

components probably would increase rather than decrease the pathology associated with envenomation. Recently, our group21 and Palm et al36 reported that the development of a type 2 immune response to honeybee venom (BV)21 or honeybee venom phospholipase A2 (bvPLA2)36 could increase the resistance of mice (as quantified by body temperature,21,36 survival,21 or both) against a near-lethal dose challenge of whole BV21 or bvPLA2.36 This effect was dependent on the high-affinity IgE receptor FcεRIa21,36 and IgE antibodies.21 In addition, we also observed that injection of mice with sublethal amounts of Russell viper venom (RVV), a snake venom of high clinical relevance,18,19 induced a type 2 immune response that enhanced the survival of mice injected with a potentially lethal amount of that venom.21 In the present study we aimed to define the importance of IgE antibodies, FcεRIa, FcεRIa1 IgE effector cells, and local IgE-mediated MC activation in the orchestration of systemic resistance against RVV. In addition, we evaluated the influence of repeated exposure to venom and the genetic background of the host on acquired protection against challenge with a potentially lethal amount of RVV or BV.

METHODS Mice All animal care and experiments were carried out in accordance with current National Institutes of Health guidelines and with approval of the Stanford University Institutional Animal Care and Use Committee. Age-matched 5- to 7-week-old wild-type (WT) C57BL/6J or BALB/cJ female mice were purchased from Jackson Laboratories (Sacramento, Calif). All transgenic mouse strains were bred and housed with the respective (in house-bred) control mice in the Stanford animal facilities under specific pathogen-free conditions. Details regarding transgenic strains can be found in the Methods section in this article’s Online Repository at www.jacionline.org.

Reagents Russell viper (Daboia russelii) venom was obtained from Sigma (St Louis, Mo; lots SLBB5602V and SLBK7058V). Details regarding additional reagents can be found in the Methods section in this article’s Online Repository.

Venom injections Briefly, mice were shaved at the injection sites 24 hours before injection and were consistently treated in the morning (without anesthesia) with subcutaneous injections of 50 mL of PBS alone or containing indicated amounts of RVV or BV. Additional details regarding injections, mouse handling, quantification of scratching behavior, and descriptions of experiments involving serum transfer or multiple exposure to venoms before

J ALLERGY CLIN IMMUNOL nnn 2015

high-dose venom challenge are provided in the Methods section in this article’s Online Repository.

Other methods Detailed descriptions of the following methods are provided in the Methods section in this article’s Online Repository: histology and assessment of MC degranulation; analysis of skin and white blood cells by using flow cytometry; measurement of RVV-specific IgG1 and IgE, BV-specific IgG1, bvPLA2-specific IgE, and total IgE antibody levels; anti–dinitrophenol-conjugated human serum albumin (DNP-HSA)–specific IgE-dependent passive cutaneous anaphylaxis; antibody-mediated neutrophil depletion; and generation and degranulation analysis of bone marrow–derived cultured mast cells (BMCMCs).

Statistical analysis Statistical tests were performed with GraphPad PRISM 6 software (GraphPad Software, La Jolla, Calif). The 2-tailed Student t (unpaired), Mann-Whitney, Mantel-Cox, or x2 tests were performed as noted in the figure legends. Unless otherwise specified, data are shown as means 6 SEMs.

RESULTS MCs rapidly degranulate on injection of RVV and contribute to enhanced innate resistance to RVV Injection of RVV subcutaneously into naive C57BL/6 WT mice elicited intense scratching at that site (data not shown), rapid degranulation of skin MCs (Fig 1, A), local hemorrhage (Fig 1, B), and tissue infiltration with neutrophils and basophils (Fig 1, C and D), whereas the small numbers of eosinophils at such sites were not significantly different in sites injected with RVV versus PBS (data not shown). Systemically, RVV injection induced an increased percentage of blood neutrophils (see Fig E1 in this article’s Online Repository at www.jacionline.org) and marked hypothermia (Fig 1, E). However, almost all mice appeared to recover fully within 24 hours (Fig 1, E and F). Pretreatment of C57BL/6 and BALB/c mice with the H1-antihistamine triprolidine but not with the platelet-activating factor (PAF) receptor antagonist CV-6209 significantly decreased RVVinduced hypothermia without affecting mortality (Fig. 1, G and H, and see Fig E2 in this article’s Online Repository at www. jacionline.org). However, in C57BL/6 mice combined treatment with the antihistamine and PAF receptor antagonist did not protect against RVV-induced hypothermia and significantly increased RVV-induced mortality (Fig 1, G and H), whereas such treatment decreased hypothermia but did not influence mortality in RVV-injected BALB/c mice (see Fig E2). These findings suggest that there might be strain-dependent differences in the mechanisms contributing to responses to RVV in naive mice. We next evaluated the possible contributions of MCs, basophils, and neutrophils to the type 2 humoral response induced by RVV. Injection of a sublethal dose of RVV in basophil-deficient Mcpt8-Crehet; DTAfl/2 mice or MC- and basophil-deficient Cpa3-Cre1; Mcl-1fl/fl mice, which are markedly deficient in MCs and have an approximately 75% reduction in blood basophil counts,37 induced serum IgG1 and IgE antibody levels not significantly different from those in the corresponding littermate control mice (see Fig E3, A-E, in this article’s Online Repository at www.jacionline.org). Interestingly, anti–GR-1–treated neutrophil-depleted C57BL/6 mice had similar levels of IgG1

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

FIG 1. RVV can induce local MC degranulation, recruitment of innate inflammatory cells, and hypothermia. A and B, Toluidine blue–stained (Fig 1, A) and hematoxylin and eosin–stained (Fig 1, B) back skin sections. Fig 1, A, shows the extent of MC degranulation (mean 1 SD). C and D, Flow cytometric plots (Fig 1, C) and quantification (Fig 1, D; mean 1 SD from 3 mice representative of 2 experiments) of CD451 skin cells. E and F, Temperature (Fig 1, E) and survival (Fig 1, F) after RVV injection. G and H, Temperature (right panel magnifies the area in the dashed box; Fig 1, G) and survival (Fig 1, H) of RVV-treated mice pretreated with antihistamine (ie, triprolidine), PAF receptor antagonist (ie, CV-6209), or both. P values were determined as follows: x2 test (Fig 1, A), Student t test (Fig 1, D, E, and G), and Mantel-Cox test (Fig 1, H). Symbols in Fig 1, E, show comparison of the group in that color with vehicle-treated mice for that time point. Fig 1, EH, show data pooled from 2 to 3 experiments. *P < .05, **P < .01, and ***P < .001.

STARKL ET AL 3

4 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

A

Assess survival

Inject 2x 25 μg RVV Day 0

C, F

1 D, G Assess scratching

B, E Measure body temperature

B

C

D

% Survival

Scratching attempts (min 35- 45)

C57BL/6 Cpa3-Cre+; Mcl-1+/+ C57BL/6 Cpa3-Cre+; Mcl-1fl/fl 12 0 100 10 -2 80 8 -4 60 ** 6 -6 40 4 ** ** -8 20 2 *** -10 0 0 01 3 6 1234567 0 1 3 6 24 Hours after injection Hours Days of RVV (2x 25 μg) after injection of RVV (2x 25 μg)

F

C57BL/6 KitW-sh/W-sh 100 80 60 40 20 0

% Survival

Temp (°C)

0 -2 -4 -6 -8 -10

0 1 3 6 24 Hours after injection of RVV (2x 25 μg)

***

G

C57BL/6 Kit+/+

*

01

3 6 1234567 Hours Days after injection of RVV (2x 25 μg)

Scratching attempts (min 35-45)

Temp (°C)

E

7 Sacrifice

12 10 8 6 4 2 0

***

FIG 2. MCs can contribute to innate resistance and behavioral responses to RVV. Experimental outline (A), body temperature (B and E), survival (C and F), and scratching attempts (D and G) of MC-deficient Cpa3-Cre1; Mcl-1fl/fl (Fig 2, B-D) and KitW-sh/W-sh (Fig 2, E-G) mice and corresponding control mice after RVV injection. In Fig 2, D and G, data are shown as individual values and mean 6 SEM. P values were determined as follows: Student t test (Fig 2, B, D, E, and G) and Mantel-Cox test (Fig 2, C and F). Data were pooled from 2 to 4 experiments (n 5 5-21 per group). *P < .05, **P < .01, and ***P < .001.

antibodies but significantly higher levels of IgE antibodies than the isotype control antibody-treated mice (see Fig E3, F-H). These results provide evidence that neither MCs, basophils, nor neutrophils are necessary for induction of a type 2 humoral immune response to RVV. We next evaluated the possible contribution of MCs to innate resistance against RVV by testing 2 different types of MC-deficient mice (Fig 2, A). C57BL/6-KitW-sh/W-sh mice virtually lack MCs (but exhibit moderately increased numbers of blood basophils38) because of a mutation in c-kit, the gene encoding stem cell factor receptor.39,40 The MC deficiency of C57BL/6-Cpa3-Cre1; Mcl-1fl/fl mice is independent of c-kit and accompanied by decreased blood basophil numbers.37 We found that each type of MC-deficient mouse exhibited significantly more susceptibility to RVV toxicity, as assessed based on extent of hypothermia (Fig 2, B-E), survival (Fig 2, C-F), or both, than did the corresponding control mice. MC-deficient mice also exhibited an almost complete absence of the scratching that was elicited in control mice (Fig 2, D-G).

IgE- and FcεRIa-dependent effector mechanisms contribute to increased survival of mice challenged with RVV To assess whether IgE antibodies can contribute to acquired host resistance to RVV,21 we injected a low dose of RVV into IgEdeficient C57BL/6-Igh72/2 and IgE-sufficient C57BL/6-Igh71/1 mice (Fig 3, A). RVV induced RVV-specific IgG1 antibodies in both Igh72/2 and Igh71/1 mice (Fig 3, B) but no detectable serum IgE in Igh72/2 mice (Fig 3, C). When challenged subcutaneously with a potentially lethal dose of RVV 3 weeks after their first exposure to RVV, C57BL/6-Igh71/1 mice, but not C57BL/6-Igh72/2 mice, exhibited enhanced resistance to the RVV-induced hypothermia and mortality (Fig 3, D and E). The same was true for the comparison between IgE-deficient and IgE-sufficient BALB/c mice (see Fig E4 in this article’s Online Repository at www.jacionline.org). Serum transfer studies also supported a critical role for IgE antibodies in acquired resistance to RVV; enhanced protection could be transferred passively to naive C57BL/6 mice (Fig 3, F)

STARKL ET AL 5

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

28 Sacrifice

D

F

-4

100 80 60 40 20 0

Collect serum 21

Recipient naive C57BL/6

Heat (1h at 56°C) or anti-IgE treatment Challenge with high dose RVV Transfer I, J serum

Day 0

I

21

20

Temp (°C)

*

0 1 3 6 24 Hours after injection of RVV (2x 37.5 μg)

*

2500 n.d. n.d. PBS RVV PBS RVV Igh-7+/+ Igh-7-/-

0

01

3 6 1234567 Hours Days after injection of RVV (2x 37.5 μg)

H

10 6

***

10 4 10 2 10

0

PBS RVV

0.8

***

0.6 0.4 0.2 0.0

PBS RVV

28 Sacrifice

PBS-serum RVV-serum Heated RVV-serum Anti-IgE RVV-serum *

10000 5000

**

G

Antibodies G, H

***

E

0 1 3 6 24 Hours after injection of RVV (2x 37.5 μg)

Inject low dose RVV or PBS

15000

PBS RVV PBS RVV Igh-7+/+ Igh-7-/-

(6h, 24h) * (6h)

Donor C57BL/6

0 -2 -4 -6 -8 -10 -12

10

0

*

-6

Day 0

10 2

RVV-IgG1 (AU)

Temp (°C)

-2

-8

10 4

Igh-7+/+ PBS Igh-7+/+ RVV Igh-7-/- PBS Igh-7-/- RVV

0

***

RVV-IgE (Abs450nm)

14 21 B, C Measure serum antibodies

***

(1.5h, 3h, 6h) (1.5h) (1.5h)

J 100 80 60 40 20 0

% Survival

Day 0

Challenge with high dose RVV D, E

*

Total IgE (ng/ml)

Inject RVV (2x 12.5 μg) or PBS

C

10 6

% Survival

B RVV-IgG1 (AU)

A

*

**

***

01

3 6 1234567 Hours Days after injection of RVV (2x 37.5 μg)

FIG 3. IgE can contribute to acquired resistance to RVV. A, Outline of experiments with IgE-deficient (Igh-72/2) and control (Igh-71/1) C57BL/6 mice (Fig 3, B-E). B and C, Serum RVV-specific IgG1 (Fig 3, B) and total IgE (Fig 3, C). D and E, Body temperature (Fig 3, D) and survival (Fig 3, E). F, Outline of serum transfer experiments in C57BL/6 mice (Fig 3, G-J). G and H, Serum RVV-specific IgG1 (Fig 3, G) and total IgE (Fig 3 H). I and J, Body temperature (Fig 3, I) and survival (Fig 3, J). Data were pooled from 3 to 4 experiments (n 5 9-25 per group). In Fig 3, B, C, G and H, data are shown as individual values and mean 6 SEM. P values were determined as follows: Mann-Whitney test (Fig 3, B, C, G, and H), Student t test (Fig 3, D and I), and Mantel-Cox test (Fig 3, E and J). *P < .05, **P < .01, and ***P < .001. n.d., Not detectable.

by injecting them with 250 mL of serum collected from RVV-exposed WT donor mice (RVV-serum) that contained significantly increased levels of RVV-specific IgG1 and IgE antibodies (Fig 3, G and H, respectively) but not with the same amount of serum obtained from PBS mock-immunized mice (PBS serum; Fig 3, F, I, and J). Moreover, RVV serum from WT mice lost its protective potential when the contained IgE antibodies were neutralized either by heating, which destroys the ability of IgE to bind to FcεRI and induce passive cutaneous anaphylaxis without affecting the function of other antibody isotypes,41,42 or treatment with an anti-IgE antibody (Fig 3, F, I, and J).

The immune functions of IgE are primarily mediated by effector cells, including MCs and basophils, which express FcεRI.43,44 Both C57BL/6-Fcer1a2/2 mice, which lack the IgE-binding component of FcεRI (ie, FcεRIa), and WT animals had similar type 2 humoral responses after subcutaneous injection of RVV (Fig 4, A-C), but enhanced resistance to RVV challenge could only be detected in mice expressing the complete IgE receptor (Fig 4, D-E). Furthermore, in passive immunization experiments C57BL/6-Cpa3-Cre1; Mcl-1fl/fl mice exhibited no difference in survival after RVV challenge whether they had received untreated RVV serum from C57BL/6 WT mice, which contained functionally active venom-specific IgE antibodies, or

6 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

B

Inject low dose RVV or PBS

14 21 B, C Measure serum antibodies

28 Sacrifice

10

***

4

10 0

PBS RVV PBS RVV Fcer1a+/+ Fcer1a-/-

**

(1h, 3h) ** (6h) (1h, 3h)

0 1 3 6 24 Hours after injection of RVV (2x 37.5 μg)

0 -2 -4 -6 -8 -10 -12

10000 8000 6000 4000 2000 0

100 80 60 40 20 0

% Survival

Temp (°C)

*

.07

PBS RVV PBS RVV Fcer1a+/+ Fcer1a-/-

*

01

***

*

*

3 6 1234567 Hours Days after injection of RVV (2x 37.5 μg)

Challenge with Tranfer of C57BL/6 WT high dose RVV-serum or PBS-serum into RVV C57BL/6-Cpa3-Cre+;Mcl-1fl/fl mice G, H 0

Day -1

Temp (°C)

E

Fcer1a+/+ PBS Fcer1a+/+ RVV Fcer1a-/- PBS Fcer1a-/- RVV

F

G

***

10 2

D 0 -2 -4 -6 -8 -10

C

.66

PBS-serum RVV-serum .12 .12

.10

0 1 3 6 24 Hours after injection of RVV (2x 25 μg)

H

7 Sacrifice 100 80 60 40 20 0

% Survival

Day 0

10 6

Total IgE (ng/ml)

Challenge with high dose RVV D, E

RVV-IgG1 (AU)

A

.08

01

3 6 1234567 Hours Days after injection of RVV (2x 25 μg)

FIG 4. FcεRIa and FcεRIa-bearing cells can contribute to acquired resistance to RVV. A, Outline of experiments with Fcer1a2/2 and control (Fcer1a1/1) C57BL/6 mice (Fig 4, B-E). B and C, Serum RVV-specific IgG1 (Fig 4, B) and total IgE (Fig 4, C) levels. D and E, Body temperature (Fig 4, D) and survival (Fig 4, E). F, Outline of serum transfer experiments involving MC-deficient C57BL/6 mice (Fig 4, G and H). G and H, Body temperature (Fig 4, G) and survival (Fig 4, H). Data were pooled from 3 experiments (n 5 9-17 per group). In Fig 4, B and C, data are shown as individual values and mean 6 SEM. P values were determined as follows: Mann-Whitney test (Fig 4, B and C), Student t test (Fig 4, D and G), and Mantel-Cox test (Fig 4, E and H). *P < .05, **P < .01, and ***P < .001.

control serum from PBS mock-sensitized C57BL/6 WT mice (Fig 4, F-H). Taken together, these results demonstrate that IgE antibodies and FcεRIa-bearing effector cells contribute importantly to the acquired resistance of RVV-immunized mice against a high-dose RVV challenge.

Local anaphylactic reaction to an unrelated antigen can increase survival of mice challenged with a potentially lethal amount of RVV Immunization with bvPLA2, which represents approximately 10% of the dry weight of whole BV,45 can reduce the toxicityrelated hypothermia induced by subsequent challenge of mice with a high dose of the same allergen in an antibody- and FcεRIa-dependent manner.36 However, it is not clear whether an IgE response to a single constituent of an animal venom would

be able to enhance resistance to the entire group of toxins contained in that venom. To investigate this, we used a well-characterized mouse anti-dinitrophenol (DNP) IgE mAb,46 which can sensitize mouse MCs to degranulate in response to challenge in vivo with DNP-HSA.47,48 Specifically, we passively sensitized WT C57BL/6 and BALB/c mice against DNP-HSA by means of subcutaneous injection of anti-DNP IgE (or with anti-DNP IgG1 or IgG2b as controls) or mock sensitized them with saline and then challenged the mice subcutaneously at the same site 24 hours later by injecting a mixture of RVV and DNP-HSA (Fig 5, A). We used amounts of anti-DNP IgE and DNP-HSA that were able to induce a local increase in vascular permeability at the DNP-HSA injection site without resulting in systemic hypothermia and showed that the amount of DNP-HSA used did not by itself influence the toxicity of RVV (see Fig E5 in this article’s Online Repository at www.jacionline.org). We found

STARKL ET AL 7

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

A Inject C57BL/6 mice with anti-DNP IgE, IgG1, IgG2b or saline

Challenge with high dose RVV and DNP-HSA

Day -1

Assess survival

0

1

7 Sacrifice

Measure body temperature

B 0 -2 -4 -6 -8 -10

Saline-mock-sensitized

Temp (°C)

-5 -10 -15

0 1 3 6 24 Hours after injection of RVV and DNP-HSA (2x [37.5 RVV i 0.5 DNP-HSA]) mixed with

Vehicle vs.

IgE vs. IgG2b

IgG1

IgG2b

*

.82

IgE

IgG1

15 30 60 90 180

*

.91

.18

.76

**

**

***

***

.39

**

*

***

***

.09

**

.19

***

**

*

.14

.08

**

***

anti-DNP-IgG1 sensitized anti-DNP-IgG2b sensitized

0 15 30 60 90 Minutes after injection of RVV and DNP-HSA (2x [37.5 RVV DNP-HSA]) mixed with 0.5

Comparisons of body temperature after injection of RVV and DNP-HSA in various groups of mice. Min

anti-DNP-IgE sensitized

C 100 80 60 40 20 0

% Survival

Temp (°C)

0

***

***

01

3 6 1234567 Hours Days after injection of RVV and DNP-HSA (2x [37.5 μg RVV mixed with 0.5 μg DNP-HSA])

FIG 5. IgE-dependent passive cutaneous anaphylaxis to an irrelevant antigen can increase resistance to a potentially lethal challenge with RVV. A, Experimental outline. B and C, Body temperature (Fig 5, B) and survival (Fig 5, C) of C57BL/6 mice treated with 3 subcutaneous injections of saline alone or containing 50 ng of anti-DNP IgE, IgG1, or IgG2b antibody and challenged 18 hours later with 2 subcutaneous injections, each containing 37.5 mg of RVV and 0.5 mg of DNP-HSA. Data were pooled from 2 to 5 independent experiments (n 5 10-25 per group). P values were determined as follows: Student t test (Fig 5, B) and Mantel-Cox test (Fig 5, C). *P < .05, **P < .01, and ***P < .001.

that presensitization with anti-DNP IgE significantly increased the resistance of C57BL/6 (Fig 5, B and C) or BALB/c (see Fig E5, H and I) mice to challenge with a potentially lethal amount of RVV admixed with DNP-HSA. However, presensitization of mice with anti-DNP IgG1 or IgG2b, DNP-specific IgG isotypes with the capacity to activate effector cells through Fcg receptors,49 not only did not increase protection but also resulted in increased hypothermia at early time points compared with vehicle-treated or IgE-sensitized mice (Fig 5, B and C). Compared with passive sensitization with a 10-fold higher amount of anti-ovalbumin (OVA) IgE, anti-DNP IgE significantly enhanced the survival of IgE-deficient mice challenged with a potentially lethal amount of RVV admixed with DNP-HSA (see Fig E6 in this article’s Online Repository at www.jacionline. org). By contrast, IgE-deficient mice passively sensitized with a 10:1 mixture of anti-OVA IgE and anti-DNP IgE exhibited a level of survival that was intermediate between that seen in mice receiving either anti-DNP IgE alone or anti-OVA IgE alone (see Fig E6, C). This result suggests that the effect on survival of antigen-specific IgE in this model might depend on the proportion

of antigen-specific versus non–antigen-specific IgE on FcεRIbearing effector cells. These findings show that local tissue responses mediated by IgE and antigen can enhance host resistance against RVV, even when that antigen is not a native constituent of the venom, and are consistent with the general idea that the host needs only to generate an IgE response against a limited number of components of a complex venom (perhaps as few as 1 component) to manifest enhanced acquired resistance to that venom.

Influence of venom type, genetic background, and venom exposure protocol on the protective effects of type 2 immune responses IgE-associated type 2 immune responses induced by a single exposure to BV21 or RVV (this study) can increase the resistance of C57BL/6 or BALB/c mice to challenge with a potentially lethal amount of that venom. However, in nature some animals are exposed to the same venom more than twice. To analyze the potential effects of multiple venom exposures on acquired

8 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

A

Day 0

10 6 10 4

D *** *** ***

***

10 2 10 0

Day 0 Day 21

PBS RVV RVV PBS RVV RVV PBS PBS RVV PBS PBS RVV

C57BL/6

35 42 Measure serum antibodies B-G

4000 3000

.15

F *** *** ***

*** **

2000 1000 0

Day 0 Day 21

PBS RVV RVV PBS RVV RVV PBS PBS RVV PBS PBS RVV

BALB/c

C57BL/6

0 -2 -4 -6 -8 -10

0 1 3 6 24 Hours after injection of RVV **

(1h, 3h, 6h)

*

RVV-PBS

BALB/c

RVV-RVV

01

*

3 6 1234567 Hours Days after injection of RVV

100 80 60 40 20 0

**

(6h)

** (1h,

3h)

*** (6h)

.16

% Survival

.07

0 1 3 6 24 Hours after injection of RVV

(0.5h, 1h, 1.5h)

K

% Survival

100 80 60 40 20 0

PBS RVV RVV PBS RVV RVV PBS PBS RVV PBS PBS RVV

C57BL/6

PBS-PBS

I

* (3h)

J

*** *

BALB/c (Panels I, K)

RVV-RVV

Temp (°C)

0 -2 -4 -6 -8 -10

Day 0 Day 21

*** ***

Temp (°C)

H

RVV-PBS

G *** ***

1.0 0.8 0.6 0.4 0.2 0.0

BALB/c

C57BL/6 (Panels H, J) PBS-PBS

48 Sacrifice

E RVV-IgE (Abs450)

C *** ***

Total IgE (ng/ml)

RVV-IgG1 (AU)

B

21 PBS PBS RVV

PBS RVV RVV

/ /

Challenge with high dose RVV H-K

Injection 2 low dose RVV or PBS

Injection 1 low dose RVV or PBS

***

01

*

3 6 1234567 Hours Days after injection of RVV

FIG 6. Influence of genetic background and immunization regimen on acquired resistance to RVV. A, Experimental outline. B-G, Serum RVV-specific IgG1 (Fig 6, B and C), total IgE (Fig 6, D and E), and RVV-specific IgE (Fig 6, F and G) levels. H and I, Body temperature. J and K, Survival. Data were pooled from 3 to 4 experiments (n 5 11-16 per group). In Fig 6, B-G, data are shown as individual values and mean 6 SEM. P values were determined as follows: Mann-Whitney test (Fig 6, B-G), Student t test (Fig 6, H and I), or Mantel-Cox test (Fig 6, J and K). *P < .05, **P < .01, and ***P < .001.

resistance to that venom, we injected mice subcutaneously with RVV (or PBS as a control) once at day 0 after which some RVV-injected mice received a second subcutaneous RVV injection on day 21 (or got PBS as a control), and then all mice were challenged with a high dose of RVV at day 42 (Fig 6, A). C57BL/6 and BALB/c mice that had received 2 prior RVV injections (RVV-RVV mice) had significantly higher levels of RVV-specific IgG1 (Fig 6, B and C), total IgE (Fig 6, D and E), and RVV-specific IgE (Fig 6, F and G) at day 35 than the mice that received only a single RVV injection (RVV-PBS mice). Both RVV-RVV and RVV-PBS C57BL/6

mice had less hypothermia on RVV challenge than control mice that had received 2 mock immunizations with PBS (PBS-PBS mice; Fig 6, H). However, although survival of C57BL/6 mice injected once or twice with RVV was similar, the survival of the RVV-RVV mice did not quite achieve statistical significance versus that in the pooled PBS-PBS group (P 5 .07; Fig 6, J). Among BALB/c mice, animals injected once or twice with RVV were significantly more resistant than the PBS-PBS control mice to the hypothermia and mortality induced by challenge with a potentially lethal dose of RVV (Fig 6, I-K).

STARKL ET AL 9

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

A

Day 0

10

2

***

D ** .08

*

*

***

10 0 PBS BV BV PBS PBS BV

PBS BV BV PBS PBS BV

C57BL/6

BALB/c

H

Day 0 Day 21

C57BL/6 (Panels H, J) BV-PBS PBS-PBS

.79 .94

*** .34 ***

.86

PBS BV BV PBS PBS BV

PBS BV BV PBS PBS BV

C57BL/6

BALB/c

BV-BV

F

G

0.8

** **

0.6

**

0.4

.10

.43

0.2

*

0.0

Day 0 Day 21

PBS BV BV PBS PBS BV

PBS BV BV PBS PBS BV

C57BL/6

BALB/c

BALB/c (Panels I, K) BV-PBS BV-BV PBS-PBS

I

0 -2 -4 -6 -8 -10 -12 -14

Temp (°C)

0 -2 -4 -6 -8 -10 -12 -14

E

5000 4000 3000 2000 1000 0

48 Sacrifice

Temp (°C)

Day 0 Day 21

35 42 B-G Measure serum antibodies

0

*

1 3 6 24 Hours after injection of BV (4x 200 μg) (0.25h)

**

(0.25h, 1h, 1.5h, 3h)

*

*

(0.5h)

J

0

1 3 6 24 Hours after injection of BV (5x 200 μg)

(1.5h)

*

(1h, 3h)

**

(1.5h)

K

% Survival

100 80 60 40 20 0

** **

01 3 6 1234567 Hours Days after injection of BV (4x 200 μg)

100 80 60 40 20 0

.13

% Survival

10

4

C Total IgE (ng/ml)

BV-IgG1 (AU)

B 10 6

21 PBS PBS BV

PBS BV BV

/ /

Challenge with high dose BV H-K

Injection 2 low dose BV or PBS

PLA2-IgE (Abs450nm)

Injection 1 low dose BV or PBS

**

.10

01 3 6 1234567 Hours Days after injection of BV (5x 200 μg)

FIG 7. Influence of genetic background and immunization regimen on acquired resistance to BV. A, Experimental outline. B-G, Serum BV-specific IgG1 (Fig 7, B and C) and total IgE (Fig 7, D and E) on day 35 from the same mice shown in Fig 7, H-K and bvPLA2-specific IgE (Fig 7, F and G) from separate groups of mice on day 42. H and I, Body temperature. J and K, Survival. Data were pooled from 3 experiments (n 5 9-11 per group). In Fig 7, B-G, data are shown as values from the individual mice in the 3 different experiments performed and mean 6 SEM (B-E) or as values for the pooled sera obtained from separate groups of mice in each of the 3 different experiments performed and mean 6 SEM. P values were determined as follows: Mann-Whitney test (Fig 7, B-G), Student t test (Fig 7, H and I), and Mantel-Cox test (Fig 7, J and K). *P < .05, **P < .01, and ***P < .001.

We also tested the consequences of a second exposure to BVon responses to challenge with a potentially lethal amount of BV in C57BL/6 versus BALB/c mice (Fig 7, A). Serum levels of BV-specific IgG1 (Fig 7, B and C), total IgE (Fig 7, D and E), and bvPLA2-specific IgE (Fig 7, F and G) were significantly higher in the sera of C57BL/6 mice that had received 2 exposures to BV (BV-BV mice) compared with those of mice that had

received only 1 exposure (BV-PBS mice), whereas the differences in antibody levels in the 2 corresponding groups of BALB/c mice were statistically significant only in the case of bvPLA2-specific IgE. C57BL/6 mice that had been injected once with BV before potentially lethal challenge showed significantly increased survival compared with PBS-PBS control animals, confirming our prior findings,21 but this was not true for the C57BL/6

10 STARKL ET AL

mice, which were injected twice with BV before high-dose BV challenge (Fig 7, J). Moreover, these BV-BV C57BL/6 mice exhibited a decrease in body temperature in response to high-dose BV challenge that was significantly more profound than that observed in the BV-PBS mice over the entire first 3 hours of the response and that was even significantly worse than that of the PBS-PBS mice at 15 minutes after BV challenge (Fig 7, H). In contrast to results with C57BL/6 mice, in BALB/c mice challenged with high-dose BV, hypothermia was not significantly exacerbated in BV-BV versus BV-PBS mice (Fig 7, I), and survival was significantly enhanced by 2 BV exposures, whereas the effect on survival did not reach statistical significance in the BV-PBS mice (P 5 .1; Fig 7, K). Notably, the strain-dependent differences observed in the responses to high-dose BV in mice immunized once or twice with low-dose BV did not appear to reflect differences in the ability of IgE antibodies from these mice to sensitize MCs to degranulate in response to BV challenge in vitro (see Fig E7 in this article’s Online Repository at www.jacionline.org).

DISCUSSION Antigen-specific IgE antibodies and FcεRI-expressing effector cells constitute a sensitive, specific, and powerful module of acquired immunity that can respond within minutes to exposure to small amounts of antigen by initiating local or systemic inflammatory reactions.9,11 It appears plausible that this rapid and efficient but also potentially dangerous effector mechanism evolved primarily to operate in situations that represent a substantial threat for the organism. In her ‘‘toxin hypothesis of allergy,’’ Margie Profet22 proposed that toxins and venoms represent examples of such substantial threats and that ‘‘allergic reactions’’ originally evolved as immune defense mechanisms against such noxious substances. Recently, our laboratory21 and others36 provided in vivo experimental evidence that IgE antibodies can indeed contribute to protective immunity in mice against either whole BV21 or the potentially toxic BV enzyme bvPLA2.36 The results of the current study indicate that acquired IgE-mediated immune resistance is not restricted to BV but can also be deployed as a potent adaptive immune defense mechanism against a reptile venom of high clinical relevance.19 Notably, the immunization and challenge doses of RVV used in this study (25 and 50-100 mg, respectively), in relation to the body weight of a mouse, are similar to the amounts of venom to which a human might be exposed if bitten by a Russell viper.19 Taken together, our findings support the idea that IgE antibodies and FcεRIa-bearing effector cells might constitute part of a general defense strategy against animal venoms in addition to having roles in host responses to certain macroparasites.50 We also found that a local IgE-dependent reaction to an unrelated antigen (ie, DNP-HSA) not ordinarily contained in RVV can enhance the survival of mice subjected to challenge with a potentially lethal amount of RVV. Thus our results support the conclusion that mounting an IgE-dependent reaction to a single antigen can be sufficient to enhance host resistance to the complex mixture of toxins contained in the venom.51 The effector mechanisms involved in such enhanced resistance remain to be defined but might include MC-mediated venom detoxification13-17 and local dilution and interference with the systemic spread of the toxins.22,23

J ALLERGY CLIN IMMUNOL nnn 2015

Since it was discovered that IgE antibodies can mediate anaphylactic reactions,52-55 the development of IgE antibodies specific for certain antigens, including components of venoms,25,26,29,31,33,56,57 has primarily been regarded as a risk factor for the development of deleterious IgE-mediated hyperreactivity on subsequent antigen exposure. Yet a recent survey of more than 7000 German adults58,59 showed a prevalence of 22.6% for sensitization (ie, having specific serum IgE antibodies) against Hymenoptera (wasp and bee) venom in the general public,58 whereas the lifetime prevalence of diagnosed insect venom allergy in that group is only 2.8%.59 Indeed, it is well known from other studies that the vast majority (approximately 80%) of patients with demonstrable IgE antibodies specific for Hymenoptera venoms have no history of manifesting systemic reactions to such venoms56,60 and that the presence of antigen-specific IgE antibodies, taken in isolation, is not predictive of severe clinical reactivity to the recognized antigens.61-66 Therefore it is possible that in some human subjects the presence of anti-venom IgE antibodies might be beneficial, for instance through decreasing venom toxicity and tissue damage on subsequent venom exposure. It is thought that multiple factors, such as differences in pathogen exposure during childhood, the characteristics of the host’s microbiome, and many other environmental influences, as well as genetic background and the nature and frequency of exposure to potential allergens, can contribute to the variation in individual susceptibilities to clinical allergies.67-70 Here we compared the resistance of C57BL/6 and BALB/c mice to RVV or BV after 1 versus 2 sublethal exposures to the same venom. In contrast to BALB/c mice, C57BL/6 mice that were immunized twice with BV rapidly developed increased hypothermia on subsequent BV challenge. Importantly, such twice-immunized C57BL/6 mice, in striking contrast to singly immunized C57BL/6 mice or twice-immunized BALB/c mice, did not exhibit enhanced resistance against high-dose BV challenge. Taken together, our data indicate that, depending on the mouse strain and type of venom, a second exposure to venom can either increase (BV in BALB/c mice) or eliminate (BV in C57BL/6 mice) the enhanced protection to venom challenge observed after a single exposure to that venom. Although many factors might contribute to such strain-dependent differences, including genetically determined differences in end-organ sensitivity to MC-derived mediators,71-74 our data suggest that such factors probably do not include differences in the ability of IgE antibodies from these mice to sensitize MCs to degranulate in response to BV challenge. Our findings are consistent with the hypothesis that the coevolution of mammals with venomous animals provided positive evolutionary pressure to conserve IgE antibodies and IgE effector cells as survival advantages. However, it seems likely that sustaining the beneficial functions of this ‘‘allergy module’’ of immunity critically requires regulatory mechanisms that can keep this potentially dangerous effector mechanism under tight control. Therefore we speculate that anaphylaxis represents only the most extreme end of a spectrum of acquired TH2 immunity to venom and that appropriately regulated TH2 immune responses can actually enhance resistance, rather than susceptibility, to venoms. In fact, the occurrence of potentially dangerous allergic TH2 responses in some subjects might represent the price paid to maintain the benefits of IgE-associated TH2 immune responses

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

for the species. For example, beekeepers, who are frequently exposed to bee venom, can exhibit high levels of BV-specific IgG and IgE antibodies, which are associated in some of these subjects with the danger of anaphylaxis.75 However, in many beekeepers exposure to multiple bee stings as the season progresses induces development of BV-specific, IL-10– producing, inducible type 1 regulatory T (TR1) cells, which suppress T-cell responses to BV in vitro and might contribute in vivo to the observed reduction in cutaneous late-phase responses to bee stings that occur as the beekeeping season progresses.76 Mechanisms of antigen-induced, regulatory T cell–dependent immune tolerance also are thought to contribute to the success of venom-specific immunotherapy in patients with Hymenoptera venom allergy.77 Therefore it is tempting to speculate that IgE-dependent enhanced resistance to the toxicity of BV might represent an initial phase of a beneficial adaptive immune response to BV that, in subjects frequently exposed to the venom, can be supplemented or supplanted by regulatory T cell–dependent immune tolerance to BV, one important function of which is to restrain the development of an overly excessive and therefore potentially dangerous IgE response to BV. We thank all members of the Galli laboratory for helpful discussions, Hans C. Oettgen and Mitchell Grayson for generously providing IgE-deficient mice, and Chen Liu and Mariola Liebersbach for excellent technical assistance.

Key messages d

IgE and IgE effector mechanisms can limit RVV toxicity in mice.

d

A local anaphylactic reaction elicited by an unrelated antigen at the site of RVV injection can increase resistance against that venom.

d

The extent of IgE-associated acquired resistance to venom can be influenced by venom type, mouse genetics, and the number of exposures to that venom.

REFERENCES 1. Berger BJ, Bhatti AR. Snake venom components and their cross-reactivity: a review. Biochem Cell Biol 1989;67:597-601. 2. Habermann E. Bee and wasp venoms. Science 1972;177:314-22. 3. Meier J, White J. Handbook of clinical toxicology of animal venoms and poisons. Boca Raton (FL): CRC Press; 1995. 4. Fry BG, Roelants K, Champagne DE, Scheib H, Tyndall JD, King GF, et al. The toxicogenomic multiverse: convergent recruitment of proteins into animal venoms. Annu Rev Genomics Hum Genet 2009;10:483-511. 5. Brodie ED 3rd. Toxins and venoms. Curr Biol 2009;19:R931-5. 6. Perales J, Neves-Ferreira AG, Valente RH, Domont GB. Natural inhibitors of snake venom hemorrhagic metalloproteinases. Toxicon 2005;45:1013-20. 7. Voss RS, Jansa SA. Snake-venom resistance as a mammalian trophic adaptation: lessons from didelphid marsupials. Biol Rev Camb Philos Soc 2012;87:822-37. 8. Rowe AH, Xiao Y, Rowe MP, Cummins TR, Zakon HH. Voltage-gated sodium channel in grasshopper mice defends against bark scorpion toxin. Science 2013;342:441-6. 9. Galli SJ, Grimbaldeston M, Tsai M. Immunomodulatory mast cells: negative, as well as positive, regulators of immunity. Nat Rev Immunol 2008;8:478-86. 10. Kitamura Y. Heterogeneity of mast cells and phenotypic change between subpopulations. Annu Rev Immunol 1989;7:59-76. 11. Galli SJ, Kalesnikoff J, Grimbaldeston MA, Piliponsky AM, Williams CM, Tsai M. Mast cells as ‘‘tunable’’ effector and immunoregulatory cells: recent advances. Annu Rev Immunol 2005;23:749-86.

STARKL ET AL 11

12. Metz M, Grimbaldeston MA, Nakae S, Piliponsky AM, Tsai M, Galli SJ. Mast cells in the promotion and limitation of chronic inflammation. Immunol Rev 2007;217:304-28. 13. Akahoshi M, Song CH, Piliponsky AM, Metz M, Guzzetta A, Abrink M, et al. Mast cell chymase reduces the toxicity of Gila monster venom, scorpion venom, and vasoactive intestinal polypeptide in mice. J Clin Invest 2011;121:4180-91. 14. Higginbotham RD. Mast cells and local resistance to Russell’s viper venom. J Immunol 1965;95:867-75. 15. Higginbotham RD, Karnella S. The significance of the mast cell response to bee venom. J Immunol 1971;106:233-40. 16. Metz M, Piliponsky AM, Chen CC, Lammel V, Abrink M, Pejler G, et al. Mast cells can enhance resistance to snake and honeybee venoms. Science 2006;313: 526-30. 17. Schneider LA, Schlenner SM, Feyerabend TB, Wunderlin M, Rodewald HR. Molecular mechanism of mast cell mediated innate defense against endothelin and snake venom sarafotoxin. J Exp Med 2007;204:2629-39. 18. Simpson ID, Norris RL. Snakes of medical importance in India: is the concept of the ‘‘Big 4’’ still relevant and useful? Wilderness Environ Med 2007;18:2-9. 19. Warrell DA. Snake venoms in science and clinical medicine. 1. Russell’s viper: biology, venom and treatment of bites. Trans R Soc Trop Med Hyg 1989;83:732-40. 20. Wernersson S, Pejler G. Mast cell secretory granules: armed for battle. Nat Rev Immunol 2014;14:478-94. 21. Marichal T, Starkl P, Reber LL, Kalesnikoff J, Oettgen HC, Tsai M, et al. A beneficial role for immunoglobulin E in host defense against honeybee venom. Immunity 2013;39:963-75. 22. Profet M. The function of allergy: immunological defense against toxins. Q Rev Biol 1991;66:23-62. 23. Palm NW, Rosenstein RK, Medzhitov R. Allergic host defences. Nature 2012; 484:465-72. 24. Bilo BM, Rueff F, Mosbech H, Bonifazi F, Oude-Elberink JN. Diagnosis of Hymenoptera venom allergy. Allergy 2005;60:1339-49. 25. de Medeiros CR, Barbaro KC, Lira MS, Franca FO, Zaher VL, Kokron CM, et al. Predictors of Bothrops jararaca venom allergy in snake handlers and snake venom handlers. Toxicon 2008;51:672-80. 26. Kopp P, Dahinden CA, Mullner G. Allergic reaction to snake venom after repeated bites of Vipera aspis. Clin Exp Allergy 1993;23:231-2. 27. Reimers AR, Weber M, Muller UR. Are anaphylactic reactions to snake bites immunoglobulin E-mediated? Clin Exp Allergy 2000;30:276-82. 28. Schmutz J, Stahel E. Anaphylactoid reactions to snakebite. Lancet 1985;2:1306. 29. Wadee AA, Rabson AR. Development of specific IgE antibodies after repeated exposure to snake venom. J Allergy Clin Immunol 1987;80:695-8. 30. deShazo RD, Butcher BT, Banks WA. Reactions to the stings of the imported fire ant. N Engl J Med 1990;323:462-6. 31. Leynadier F, Hassani Y, Chabane MH, Benguedda AC, Abbadi MC, Guerin L. Allergic reactions to North African scorpion venom evaluated by skin test and specific IgE. J Allergy Clin Immunol 1997;99:851-3. 32. Portier MM, Richet C. De l’action anaphylactique de certains venims. C R Soc Biol 1902;54:170-2. 33. Togias AG, Burnett JW, Kagey-Sobotka A, Lichtenstein LM. Anaphylaxis after contact with a jellyfish. J Allergy Clin Immunol 1985;75:672-5. 34. Beyer K, Eckermann O, Hompes S, Grabenhenrich L, Worm M. Anaphylaxis in an emergency setting—elicitors, therapy and incidence of severe allergic reactions. Allergy 2012;67:1451-6. 35. Lee JK, Vadas P. Anaphylaxis: mechanisms and management. Clin Exp Allergy 2011;41:923-38. 36. Palm NW, Rosenstein RK, Yu S, Schenten DD, Florsheim E, Medzhitov R. Bee venom phospholipase A2 induces a primary type 2 response that is dependent on the receptor ST2 and confers protective immunity. Immunity 2013;39: 976-85. 37. Lilla JN, Chen CC, Mukai K, BenBarak MJ, Franco CB, Kalesnikoff J, et al. Reduced mast cell and basophil numbers and function in Cpa3-Cre; Mcl-1fl/fl mice. Blood 2011;118:6930-8. 38. Piliponsky AM, Chen CC, Grimbaldeston MA, Burns-Guydish SM, Hardy J, Kalesnikoff J, et al. Mast cell-derived TNF can exacerbate mortality during severe bacterial infections in C57BL/6-KitW-sh/W-sh mice. Am J Pathol 2010; 176:926-38. 39. Duttlinger R, Manova K, Chu TY, Gyssler C, Zelenetz AD, Bachvarova RF, et al. W-sash affects positive and negative elements controlling c-kit expression: ectopic c-kit expression at sites of kit-ligand expression affects melanogenesis. Development 1993;118:705-17. 40. Reber LL, Marichal T, Galli SJ. New models for analyzing mast cell functions in vivo. Trends Immunol 2012;33:613-25. 41. Prouvost-Danon A, Binaghi RA, Abadie A. Effect of heating at 56 degrees C on mouse IgE antibodies. Immunochemistry 1977;14:81-4.

12 STARKL ET AL

42. Strait RT, Morris SC, Finkelman FD. IgG-blocking antibodies inhibit IgE-mediated anaphylaxis in vivo through both antigen interception and FcgRIIb cross-linking. J Clin Invest 2006;116:833-41. 43. Kinet JP. The high-affinity IgE receptor (FcεRI): from physiology to pathology. Annu Rev Immunol 1999;17:931-72. 44. Rivera J, Fierro NA, Olivera A, Suzuki R. New insights on mast cell activation via the high affinity receptor for IgE. Adv Immunol 2008;98:85-120. 45. Habermann E, Walsch P, Breithaupt H. Biochemistry and pharmacology of the cortoxin complex. II. Possible interrelationships between toxicity and organ distribution of phospholipase A, crotapotin and their combination. Naunyn Schmiedebergs Arch Pharmacol 1972;273:313-30. 46. Liu FT, Bohn JW, Ferry EL, Yamamoto H, Molinaro CA, Sherman LA, et al. Monoclonal dinitrophenyl-specific murine IgE antibody: preparation, isolation, and characterization. J Immunol 1980;124:2728-37. 47. Wershil BK, Wang ZS, Gordon JR, Galli SJ. Recruitment of neutrophils during IgE-dependent cutaneous late phase reactions in the mouse is mast cell-dependent. Partial inhibition of the reaction with antiserum against tumor necrosis factor-alpha. J Clin Invest 1991;87:446-53. 48. Schafer B, Piliponsky AM, Oka T, Song CH, Gerard NP, Gerard C, et al. Mast cell anaphylatoxin receptor expression can enhance IgE-dependent skin inflammation in mice. J Allergy Clin Immunol 2013;131:541-8, e1-9. 49. Bruhns P. Properties of mouse and human IgG receptors and their contribution to disease models. Blood 2012;119:5640-9. 50. Artis D, Maizels RM, Finkelman FD. Forum: immunology: allergy challenged. Nature 2012;484:458-9. 51. Risch M, Georgieva D, von Bergen M, Jehmlich N, Genov N, Arni RK, et al. Snake venomics of the Siamese Russell’s viper (Daboia russelli siamensis) —relation to pharmacological activities. J Proteomics 2009;72:256-69. 52. Ishizaka K, Ishizaka T, Hornbrook MM. Physico-chemical properties of human reaginic antibody. IV. Presence of a unique immunoglobulin as a carrier of reaginic activity. J Immunol 1966;97:75-85. 53. Johansson SG, Bennich H. Immunological studies of an atypical (myeloma) immunoglobulin. Immunology 1967;13:381-94. 54. Stanworth DR. The discovery of IgE. Allergy 1993;48:67-71. 55. Stanworth DR, Humphrey JH, Bennich H, Johansson SG. Specific inhibition of the Prausnitz-Kustner reaction by an atypical human myeloma protein. Lancet 1967;2:330-2. 56. Bilo BM, Bonifazi F. Epidemiology of insect-venom anaphylaxis. Curr Opin Allergy Clin Immunol 2008;8:330-7. 57. Russo AJ, Calton GJ, Burnett JW. The relationship of the possible allergic response to jellyfish envenomation and serum antibody titers. Toxicon 1983;21:475-80. 58. Haftenberger M, Laussmann D, Ellert U, Kalcklosch M, Langen U, Schlaud M, et al. [Prevalence of sensitisation to aeraoallergens and food allergens: results of the German Health Interview and Examination Survey for Adults (DEGS1)] [Article in German]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2013;56:687-97. 59. Langen U, Schmitz R, Steppuhn H. [Prevalence of allergic diseases in Germany: results of the German Health Interview and Examination Survey for Adults (DEGS1)] [Article in German]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2013;56:698-706.

J ALLERGY CLIN IMMUNOL nnn 2015

60. Sturm GJ, Heinemann A, Schuster C, Wiednig M, Groselj-Strele A, Sturm EM, et al. Influence of total IgE levels on the severity of sting reactions in Hymenoptera venom allergy. Allergy 2007;62:884-9. 61. Bousquet J, Anto JM, Bachert C, Bousquet PJ, Colombo P, Crameri R, et al. Factors responsible for differences between asymptomatic subjects and patients presenting an IgE sensitization to allergens. A GA2LEN project. Allergy 2006; 61:671-80. 62. Golden DB, Marsh DG, Kagey-Sobotka A, Freidhoff L, Szklo M, Valentine MD, et al. Epidemiology of insect venom sensitivity. JAMA 1989;262:240-4. 63. Nicolaou N, Poorafshar M, Murray C, Simpson A, Winell H, Kerry G, et al. Allergy or tolerance in children sensitized to peanut: prevalence and differentiation using component-resolved diagnostics. J Allergy Clin Immunol 2010;125:191-7, e1-13. 64. Schafer T, Przybilla B. IgE antibodies to Hymenoptera venoms in the serum are common in the general population and are related to indications of atopy. Allergy 1996;51:372-7. 65. Sturm GJ, Kranzelbinder B, Schuster C, Sturm EM, Bokanovic D, Vollmann J, et al. Sensitization to Hymenoptera venoms is common, but systemic sting reactions are rare. J Allergy Clin Immunol 2014;133:1635-43.e1. 66. Hamilton RG. Allergic sensitization is a key risk factor for but not synonymous with allergic disease. J Allergy Clin Immunol 2014;134:360-1. 67. Blumenthal MN. Genetic, epigenetic, and environmental factors in asthma and allergy. Ann Allergy Asthma Immunol 2012;108:69-73. 68. Holgate ST. Genetic and environmental interaction in allergy and asthma. J Allergy Clin Immunol 1999;104:1139-46. 69. Holloway JW, Yang IA, Holgate ST. Genetics of allergic disease. J Allergy Clin Immunol 2010;125(suppl):S81-94. 70. McGowan EC, Bloomberg GR, Gergen PJ, Visness CM, Jaffee KF, Sandel M, et al. Influence of early-life exposures on food sensitization and food allergy in an inner-city birth cohort. J Allergy Clin Immunol 2015;135:171-8. 71. Arumugam M, Ahrens R, Osterfeld H, Kottyan LC, Shang X, Maclennan JA, et al. Increased susceptibility of 129SvEvBrd mice to IgE-Mast cell mediated anaphylaxis. BMC Immunol 2011;12:14. 72. Brandt EB, Strait RT, Hershko D, Wang Q, Muntel EE, Scribner TA, et al. Mast cells are required for experimental oral allergen-induced diarrhea. J Clin Invest 2003;112:1666-77. 73. Hart PH, Grimbaldeston MA, Swift GJ, Jaksic A, Noonan FP, Finlay-Jones JJ. Dermal mast cells determine susceptibility to ultraviolet B-induced systemic suppression of contact hypersensitivity responses in mice. J Exp Med 1998; 187:2045-53. 74. Vaz NM, de Souza CM, Hornbrook MM, Hanson DG, Lynch NR. Sensitivity to intravenous injections of histamine and serotonin in inbred mouse strains. Int Arch Allergy Appl Immunol 1977;53:545-54. 75. Muller UR. Bee venom allergy in beekeepers and their family members. Curr Opin Allergy Clin Immunol 2005;5:343-7. 76. Meiler F, Zumkehr J, Klunker S, Ruckert B, Akdis CA, Akdis M. In vivo switch to IL-10-secreting T regulatory cells in high dose allergen exposure. J Exp Med 2008;205:2887-98. 77. Ozdemir C, Kucuksezer UC, Akdis M, Akdis CA. Mechanisms of immunotherapy to wasp and bee venom. Clin Exp Allergy 2011;41:1226-34.

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

METHODS Mice All animal care and experiments were carried out in accordance with current National Institutes of Health guidelines and with approval of the Stanford University Institutional Animal Care and Use Committee. For experiments involving WT mice (Figs 1; 3, F-J; 5-7; E1; E2; E3, F-H; and E5) and production of serum from WT mice (Figs 3, F-J, and 4, F-H), agematched 5- to 7-week-old C57BL/6J or BALB/cJ female mice were purchased from Jackson Laboratories (Sacramento, Calif) and housed in the Research Animal Facilities at Stanford University for at least 7 days before starting the experiments. All transgenic mouse strains were bred and housed with respective (in-house bred) control mice in the Stanford Animal facilities under specific pathogen-free conditions. C57BL/6-Fcer1a2/2 mice were originally provided by Jean-Pierre Kinet (Harvard Medical School, Boston, Mass). Igh72/2 and Igh71/1 control mice on the BALB/ c background were described previouslyE1 and generously provided by Hans C. Oettgen (Harvard Medical School, Boston, Mass). Igh72/2 mice were backcrossed 8 generations on the C57BL/6J background, and littermate control mice were generously provided by Mitchell Grayson (Medical College of Wisconsin, Milwaukee, Wis). MC-deficient C57BL/6-KitW-sh/W-sh mice and MC- and basophil-deficient C57BL/6Cpa31-Cre;Mcl-1fl/fl mice were previously described.E2,E3 We originally obtained previously described B6.129S4-Mcpt8tm1(cre)Lky/JE4 (referred to in Fig E3 as Mcpt8-Cre) and B6.129P2-Gt(ROSA)26Sortm1(DTA)Lky/JE4 (referred to in Fig E3 as DTAfl) mice from Jackson Laboratories. Homozygous Mcpt8-Cre mice were bred at Stanford with DTAfl/2 mice to generate basophil-deficient Mcpt8-Cre1/2; DTAfl/2 and basophil-sufficient Mcpt8-Cre1/2; DTA2/2 littermates (blood basophil levels were analyzed for all mice before the start of immunization). Age-matched male and female transgenic and control mice (with comparable sex distribution within the groups) were used for experiments.

Reagents Russell viper (Daboia russelii) venom was obtained from Sigma (Carlsbad, Calif) and resuspended in sterile endotoxin-free PBS (Gibco, Grand Island, NY) at 10 mg/mL. Two different RVV batches were used throughout this study that appeared to differ in their toxin composition (assessed by using PAGE and Coomassie blue staining, data not shown) and toxicity: lot SLBB5602V (country of origin, China) and lot SLBK7058V (country of origin, Pakistan). Challenge with SLBK7058V resulted in notably lower mortality than injection of the same dose of SLBB5602V. Lot SLBB5602V was used for all experiments except those in Figs 1, G and H; E2; E3; and E6, which used batch SLBK7058V. Honeybee (Apis mellifera) venom (lot 12071006HB) was obtained from ALK-Abello Source Materials (Spring Mills, Pa). Batches of freeze-dried complete BV were resuspended in PBS at 4 mg/mL. BV and RVV were stored in aliquots at 2208C. Dispase was from Gibco. Collagenase A and DNAse I were from Roche (Mannheim, Germany). Aqua Dead Cell Stain Kit and propidium iodide were obtained from Life Technologies (Grand Island, NY). V450 (Pacific Blue equivalent)–conjugated anti-mouse CD49b (DX-5), fluorescein-conjugated anti-mouse CD117 (KIT; clone 2B8), and allophycocyanin-cyanin7–conjugated anti-mouse Ly6G (clone 1A8; used for skin cell analysis) antibodies and phycoerythrin-Texas Red–conjugated streptavidin were from BD Biosciences (San Jose, Calif). Allophycocyaninconjugated anti-mouse CD45 (clone 30-F11) and anti-mouse Ly6G (Gr-1; fluorescein-conjugated clone RB6-8C5 used for blood cell analysis in Fig E1), allophycocyanin-conjugated clone 1A8-Ly6g (used for analysis of neutrophil depletion), and allophycocyanin-conjugated anti-mouse F4/80 (clone BM8) antibodies were from eBioscience (San Diego, Calif). Phycoerythrin-conjugated anti-mouse FcεRIa (clone MAR-1) antibody and fluorescein-conjugated anti-mouse CD11b (clone M1/70 used in combination with 1A8-Ly6g for analysis of neutrophil depletion) were from BioLegend (San Diego, Calif). Biotin-conjugated anti-mouse Siglec-F (clone ES22-10D8) antibody was from Miltenyi Biotec (Bergisch Gladbach, Germany). Rat anti-mouse IgE (clone R35-92) and rat IgG1 isotype controls (clone R3-34), used for experiments involving serum pre-treatment, were

STARKL ET AL 12.e1

obtained from BD PharMingen (San Jose, Calif) and dialyzed twice against PBS before use in in vivo experiments. Clones and sources of antibodies used in ELISA experiments are stated in the respective sections below. Dinitrophenyl (DNP)-specific IgE (a-DNP clone ε26E5) was kindly provided by Dr Fu-Tong Liu (University of California-Davis). DNP-specific mouse IgG1 (clone 109.3) and IgG2b (clone 10.12) were obtained from Santa Cruz Biotechnology (Santa Cruz, Calif) and dialyzed twice against PBS before use in in vivo experiments. The specificity of dialyzed anti-DNP IgG antibodies for DNP-HSA (see below) was verified by means of ELISA (data not shown). Anti-OVA IgE (clone E-C1) was from Chondrex (Redmond, Wash). PAF receptor antagonist CV-6209 was obtained from Santa Cruz Biotechnology, resuspended at 3.3 mg/mL in sterile saline, and stored in aliquots at 2208C until use. Triprolidine hydrochloride, dinitrophenyl30-40-conjugated human serum albumin, and other chemicals were obtained from Sigma.

Venom injections and active immunization Six- to 8-week-old C57BL/6 and BALB/c WT or transgenic mice were shaved at the injection sites (back skin for RVVand back and belly skin for BV challenge) 24 hours before injection and consistently treated in the morning (without anesthesia) by administering 2 subcutaneous injections of 50 mL of PBS alone or containing either 250 mg of RVV/mL (ie, 2 3 12.5 mg), which was used for standard injections of mice before high-dose RVV exposure; 500 mg of RVV/mL (ie, 2 3 25 mg), which was used for comparing the responses of MC-deficient mice and corresponding MC-sufficient control animals (Fig 2) and for testing the effect of immune serum transfer into C57BL/6-Cpa3-Cre1; Mcl-1fl/fl mice (Fig 4); or 750 mg RVV/mL (ie, 2 3 37.5 mg), which was used for challenges with a potentially lethal dose of RVV. Scratching behavior was quantified between 35 and 45 minutes after RVV injection as the number of scratching attempts with the hind legs per individual mouse. For experiments involving active ‘‘immunization’’ to RVV, mice were challenged 21 days after immunization. Body temperature was measured immediately before challenge and at indicated time intervals after challenge (in surviving mice) with a rectal thermometer. For experiments investigating the effect of 2 sequential exposures to RVV on subsequent resistance to high-dose RVV challenge (Fig 6), C57BL/6 and BALB/c mice were injected as follows: (1) on each of days 0 and 21 with 2 subcutaneous injections of 50 mL of PBS (control mice, referred to as PBS-PBS); (2) on day 0 with 2 subcutaneous injections of 12.5 mg of RVV in 50 mL of PBS and on day 21 with 2 subcutaneous injections of 50 mL of PBS (referred to as RVV-PBS mice); or (3) on each of days 0 and 21 with 2 subcutaneous injections of 12.5 mg of RVV in 50 mL of PBS (referred to as RVV-RVV mice). At day 42, all mice were challenged with 2 3 37.5 mg of BV in 50 mL of PBS. If in a particular independent challenge experiment shown in Fig 6 more than 50% of the control mice injected only with PBS on days 0 and 21 exhibited a decrease in body temperature of less than 48C for C57BL/6 mice or less than 38C for BALB/c mice by 1 hour after injection of 2 3 37.5 mg of RVV (we occasionally observed such variation in the response, which might be due to the greater age and weight of these mice compared with mice challenged 21 days after initial injections), all the mice in that experiment then received a single additional subcutaneous injection of 25 mg of RVV in 50 mL of PBS in the back skin. We used this approach to achieve sufficient mortality in the mock-immunized (ie, PBS-PBS) control animals to allow identification of potential resistance in the 2 groups of RVV-immunized mice. As in our prior study,E6 all venom injections were performed by a team of 2 experimenters working together (to enable better control of the mice during the injections), and experiments were designed and performed to allow the injection of all animals in any single experiment within minutes. For analysis of venom-induced antibody responses, blood was collected at day 14 (in experiments involving challenge after 21 days) or day 35 (in experiments involving challenge after 42 days) from the retro-orbital veins of anesthetized mice. In experiments investigating the effects of 2 sequential exposures to BV on subsequent resistance to high-dose BV challenge (Fig 7), C57BL/6 and BALB/c mice were injected as follows: (1) on each of days 0 and 21 with 2 subcutaneous injections into the back skin, each consisting of 50 mL of PBS (control mice, referred to as

12.e2 STARKL ET AL

PBS-PBS); (2) on day 0 with 2 subcutaneous injections of 200 mg of BV for C57BL/6 mice or 1 3 100 mg of BV for BALB/c mice, with each injection in 50 mL of PBS, and on day 21 with 2 subcutaneous injections into the back skin, each consisting 50 mL of PBS (referred to as BV-PBS mice); or (3) on each of days 0 and 21 with 2 subcutaneous injections of 200 mg of BV for C57BL/6 mice or 1 3 100 mg of BV for BALB/c mice, each injection in 50 mL of PBS (referred to as BV-BV mice). These BV amounts were selected because we showed previously that they resulted in significantly increased resistance to challenge with a potentially lethal BV dose 21 days later.E6 At day 42, C57BL/6 mice were challenged subcutaneously with 4 3 200 mg of BV, with each of the 4 injections in 50 mL of PBS (3 in back skin and 1 in belly skin), and BALB/c mice were challenged with 5 3 200 mg of BV, with each of the 5 injections in 50 mL of PBS (4 in back skin and 1 in belly skin). These challenge protocols caused death in approximately 70% to 90% of 6- to 8week-old animals, as determined in previous dose-finding experiments.E6 Body temperature was measured immediately before challenge and at indicated time intervals after challenge (in surviving mice) with a rectal thermometer.

Passive immunization: Serum preparation and transfer For experiments presented in Fig 3, F-J, and Fig 4, F-H, 6- to 8-week-old female WT C57BL/6J donor mice were immunized with 2 subcutaneous injections (in the back skin) of 12.5 mg of RVV in PBS or with PBS alone. Three weeks later, blood was collected, and sera of PBS- and RVV-injected mice were recovered after centrifugation in Serum-Gel Microtubes (Sarstedt, Neumbrecht, Germany). After quantification of antibody levels (as described above), sera were pooled and placed in aliquots for later use in treatment groups of 5 mice (approximately 1.3 mL per group) and stored at 2208C. Aliquots of pooled serum derived from donor mice that had been injected with either PBS or RVV are referred to as PBS-serum or RVV-serum, respectively. On average, the amount of serum obtained from 2 donor mice at this age (ie, a total of approximately 750 mL) was sufficient for intravenous transfer into 3 recipient naive mice. For experiments shown in Fig 3, I and J, sera were modified immediately before transfer, as described previouslyE6: (1) serum from PBS mock-immunized mice (PBS-serum) was supplemented with a 1:10 volume (50 mg/mL final concentration) of the rat IgG1 isotype control antibody (clone R3-34; BD PharMingen; dialyzed overnight against 2 3 5 L of PBS to remove sodium azide); (2) serum from RVV-immunized mice (RVV-serum) was supplemented with a 1:10 volume of dialyzed rat IgG1 isotype control antibody and not further treated before transfer (RVV-serum); (3) RVV-serum was supplemented with a 1:10 volume of dialyzed rat IgG1 isotype control antibody and heated 60 minutes at 568C in a water bath (heated BV-serum) to specifically ablate IgE functionE6-E10; and (4) RVV-serum was supplemented with a 1:10 volume of rat anti-mouse IgE antibody (clone R35-92 [a clone commonly used to neutralize cell-mediated IgE function in vivoE6,E11,E12], BD PharMingen; dialyzed overnight against 2 3 5 L of PBS to remove sodium azide) for 30 minutes at room temperature (anti-IgE RVV-serum). Aliquots of these modified sera (250 mL per mouse) were transfused intravenously through the retro-orbital vein into anesthetized 10- to 11-week-old female C57BL/6 recipient naive mice. On the following day, all recipient mice were challenged with 2 subcutaneous injections of 37.5 mg of RVV in PBS. We measured body temperature at the indicated times and monitored survival over 1 week. For experiments involving serum transfer into MC-deficient recipient mice (Fig 4, F-H), naive C57BL/6-Cpa31-Cre;Mcl-1fl/fl mice received serum generated from C57BL/6J WT mice and were challenged with 2 3 25 mg of RVV in PBS.

Histology and assessment of MC degranulation Back skin specimens were removed carefully to avoid stretching or compressing the specimens, which can increase artifactual changes in MC morphology, and were fixed with 10% formalin and embedded in paraffin, and 4-mm sections were stained with 0.1% Toluidine blue or hematoxylin and eosin for histologic examination. Images were captured with a Nikon

J ALLERGY CLIN IMMUNOL nnn 2015

E1000 M microscope (Belmont, Calif) by using a Spotflex camera and Spot version 5.1 software (Diagnostic Instruments, Sterling Heights, Mich). MC degranulation is expressed as the percentage of skin MCs examined in which greater than 50% (‘‘extensive’’ degranulation of that cell), 10% to 50% (‘‘moderate’’ degranulation of that cell), or less than 10% (‘‘none,’’ which is indicative of no evidence of significant degranulation of that cell) of the cytoplasmic granules of individual MCs exhibited morphologic evidence of degranulation (ie, alterations in the staining characteristics, size, or distribution of the granules).E6,E13 Data are presented as bar graphs of the percentage of MCs exhibiting the various extents of degranulation (mean 1 SEM), and the differences between results were examined for statistical significance by using the x2 test.

ELISAs After each incubation step in the ELISAs described below, plates are washed 3 to 5 times with PBS containing 0.05% Tween. For detection of venom-specific IgG1 serum antibodies, Nunc MaxiSorp ELISAs plates (Thermo Scientific, Waltham, Mass) were coated with whole RVV or BV diluted to 5 mg/mL PBS at 48C overnight, followed by blocking with 1% BSA in PBS for at least 2 hours at room temperature. Sera were diluted in PBS containing 1% BSA and incubated in the blocked wells for 2 hours at 378C or overnight at 48C. We detected bound IgG1 using a biotinylated detection antibody (rat anti-mouse IgG1 [clone A85-1, BD PharMingen; incubated for 1 hour at room temperature]), followed by incubation with horseradish peroxidase–conjugated streptavidin (BD PharMingen) for 20 minutes at room temperature and detection of the reaction product after using supersensitive TMB substrate (Sigma). Antibody titers were calculated by plotting the serum dilution that produced half-maximal signal of a reference serum. When the detected signal produced a corresponding titer of less than or equal to 1, an arbitrary value of 1 was assigned. For determination of absolute amounts of total serum IgE, we coated ELISA plates with rat anti-mouse IgE (clone R35-72, BD Biosciences, at 2 mg/ mL), followed by blocking, as described above. We incubated coated and blocked wells with serial dilutions of purified mouse IgE (BD PharMingen) as standard and 1:10 diluted sera (in PBS 1% BSA) for 2 hours at 378C or overnight at 48C. Bound IgE was detected by using biotinylated anti-mouse IgE (clone R35-118, BD PharMingen) and reagents, as described above. To quantify RVV- or bvPLA2-specific IgE, we substituted biotinylated RVV or bvPLA2 (see below) for biotinylated anti-mouse IgE.

Generation and analysis of degranulation BMCMCs MCs were generated from bone marrow cells of female age-matched C57BL/ 6 or BALB/c mice by means of (simultaneous) culture for 10 weeks in Walter and Eliza Hall Institute-conditioned, IL-3–containing medium until an MC purity of approximately 95% was reached, as previously described.E6 BMCMCs were sensitized overnight with serum pools (diluted 1:20 in medium) derived from 3 independent experiments (the same serum pools were used in Fig 7, F and G, for assessment of phospholipase A2–specific IgE), followed by 1 hour of stimulation without or with BV at the indicated concentrations in Tyrode buffer, to analyze the sensitization potential of immune sera (Fig E6). Percentage release of the MC granule-stored enzyme b-hexosaminidase in the supernatants of BMCMCs was measured, as previously described.E6

Pharmacologic blocking of histamine and PAF in mice exhibiting RVV-mediated hypothermia Potential contributions of histamine and PAF to the hypothermia observed on RVV injection were analyzed by using specific neutralization of these mediators according to a previously described method,E14 with slight modifications: the H1-receptor–specific antihistamine triprolidine was solubilized at 1 mg/mL in sterile saline and filter sterilized, and 200 mL was injected into mice intraperitoneally 1 hour before venom injections. The PAF receptor antagonist CV-6209 was diluted at 330 mg/mL in saline, and 200 mL was injected into the retro-orbital vein of anesthetized mice 30 minutes before venom injection; particular care was taken that all mice were exposed to

STARKL ET AL 12.e3

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

the isoflurane anesthesia for the same amount of time (ie, 2.5-3 minutes). All mice were allowed to recover from anesthesia (with exposure to an infrared lamp for 5 minutes directly after anesthesia) for the same period of time. Recovery from the anesthesia by all mice (ie, reaching a body temperature of at least 378C) was verified before injecting 2 3 37.5 mg of RVV (lot SLBK7058 V) and monitoring temperature and survival, as described above.

Antibody-mediated neutrophil depletion For assessment of the potential role of neutrophils in the type 2 immune response against RVV, C57BL/6 mice were injected intraperitoneally with 150 mg of anti–GR-1 antibody (clone RB6-8C5; BioXCell, West Lebanon, NH) for neutrophil depletionE15,E16 or an isotype control antibody (clone LTF-2, BioXCell) in 100 mL of PBS or with PBS alone (vehicle). These injections were started 3 days before immunization with 2 3 12.5 mg of RVV and continued every other day until collection of serum 14 days after immunization. Starting with day 1 after immunization, a drop of blood was collected from the tails of the mice, and levels of neutrophils were assessed as described above immediately before antibody injection to confirm (sustained) neutrophil depletion by means of flow cytometry with anti-mouse Ly-6G (clone 1A8-Ly6g) combined with anti-mouse CD11b antibodies. In the experiments shown in Fig E3, F and G, neutrophil counts in the blood of all groups of mice treated with anti–GR-1 antibody ranged from undetectable to less than 1% of blood cells at all times tested. In the PBS-treated or isotype control antibody–treated mice, blood levels of neutrophils ranged from 2% to 30% over the time period tested (with the highest levels seen on days 1 or 3 after immunization with RVV).

Biotinylation of RVV and bvPLA2 RVV and bvPLA2 (each at a stock concentration of 4 mg/mL in PBS; Sigma) were mixed 1:1 with (1)-Biotin N-hydroxysuccinimide ester (stock concentration of 2 mg/mL in dimethyl sulfoxide) and incubated at room temperature for 2 hours. One volume of 0.1 mol/L TRIS (pH 7) was added for 10 minutes to quench the reaction. The reaction mixture was dialyzed 2 3 overnight at 48C against PBS by using Slide-A-Lyzer Dialysis Cassettes (2K MWCO; Pierce, Rockford, Ill), followed by desalting with Zeba Spin Desalting Columns (7K MWCO, Pierce). Biotinylated proteins were stored at 2208C. Optimal dilutions (resulting in highest signal-to-noise ratio) of biotinylated proteins and horseradish peroxidase–conjugated streptavidin for detection of specific IgE antibodies in ELISA experiments (described above) were determined in pilot experiments by using control serum from naive mice and serum pools with high IgE content (derived from mice that had been immunized and challenged with RVV or BV).

DNP-HSA–specific and IgE-dependent passive cutaneous anaphylaxis A model of IgE-dependent passive cutaneous anaphylaxis was established in the back skin of C57BL/6 and BALB/c mice to test the effects of a local IgE-dependent anaphylactic reaction against an irrelevant antigen on resistance of mice to a potentially lethal challenge with RVV. Six- to 8-week-old C57BL/6J and BALB/cJ mice were sensitized with 3 subcutaneous injections of 50 mL of saline alone or containing 50 ng of anti-DNP IgE antibody into the back skin. For experiments shown in Fig E5, mice were anesthetized 18 hours later and injected intravenously with 200 mL of PBS containing 1% Evan blue dye and challenged 30 minutes later by means of 2 subcutaneous injections of 50 mL of PBS containing either 0.1 mg or 0.5 mg of DNP-HSA. Changes in body temperature were monitored during the first 30 minutes. Specimens of 1 cm2 in size of back skin, including the subcutaneous injection sites, were collected 1 hour after challenge and death of the mice, minced into 4 equally sized pieces, and incubated in 300 mL of dimethyl sulfoxide in a 48-well plate for 90 minutes on a shaker at room temperature. The absorption of the supernatant at 620 nm was analyzed to assess levels of extracted Evan blue dye. For experiments shown in Figs 5; E5, G-I; and E6, 18 hours after sensitization with 3 3 50 ng of anti-DNP IgE antibody (or in Fig 5 with the same amount of an anti-DNP IgG1 or anti-DNP IgG2b antibody or in Fig E6 with 3 3 50 ng of anti-DNP

IgE antibody alone or mixed with 500 ng of anti-OVA IgE, with 3 3 500 ng anti-OVA IgE alone, or with saline alone as a mock sensitization), mice were challenged with 2 subcutaneous injections, each consisting of 50 mL of PBS containing 37.5 mg of RVV and 0.5 mg of DNP-HSA.

Treatment of skin and blood samples for flow cytometric analysis For assessment of skin immune cell infiltration after RVV injection (Fig 1, C and D), skin samples of 1 cm2, including the injection sites, were dissected after death of the mice and transferred into a Dispase solution (2.5 U/mL in Hank buffered saline solution) for 90 minutes at 378C. The epidermis was removed with forceps, and the dermis was cut into small pieces and digested for 60 minutes at 378C on a shaker at 200 rpm in 2 mL of DMEM medium containing 20 mmol/L HEPES, 396 U/mL DNAse I, and 1 mg/mL collagenase A. Cell aggregates were further mechanically disrupted by passing them through an 18-gauge needle, and the cell suspension was filtered through a 70-mm cell strainer. After red blood cell lysis with ACK buffer, skin dermal cells and white blood cells were stained with the Aqua Dead Cell Stain Kit or propidium iodide, respectively, to label dead cells, followed by incubation with an anti-mouse CD16/CD32 (clone 2.4G2, BD Biosciences) antibody to reduce nonspecific binding before staining. Cells were kept at 48C during processing. We analyzed white blood cells with an Accuri C6 flow cytometer (BD Biosciences) and skin dermal samples with an LSR-II flow cytometer (BD Biosciences) at the Stanford University Flow Cytometry Core Facility. Skin neutrophils were defined as CD451Ly6G1Siglec-F2 cells, blood neutrophils as CD451Ly6GhighF4/802 cells, skin eosinophils as CD451Ly6GintSiglec-F1 cells, and skin basophils as CD451FcεRIa1CD49b1c-Kit2 cells. Blood monocytes were defined as CD451F4/80intLy6Gint cells. Results were analyzed with FlowJo software (Tree Star, Ashland, Ore).

Statistical analysis Statistical tests were performed with GraphPad PRISM 6 software (GraphPad Software, San Diego, Calif). Two-tailed Student t tests (unpaired), Mann-Whitney tests, Mantel-Cox tests, or x2 tests were performed as noted in the respective figure legends. Unless otherwise specified, data are shown as means 6 SEMs. REFERENCES E1. Oettgen HC, Martin TR, Wynshaw-Boris A, Deng C, Drazen JM, Leder P. Active anaphylaxis in IgE-deficient mice. Nature 1994;370:367-70. E2. Lilla JN, Chen CC, Mukai K, BenBarak MJ, Franco CB, Kalesnikoff J, et al. Reduced mast cell and basophil numbers and function in Cpa3-Cre; Mcl-1fl/fl mice. Blood 2011;118:6930-8. E3. Duttlinger R, Manova K, Chu TY, Gyssler C, Zelenetz AD, Bachvarova RF, et al. W-sash affects positive and negative elements controlling c-kit expression: ectopic c-kit expression at sites of kit-ligand expression affects melanogenesis. Development 1993;118:705-17. E4. Sullivan BM, Liang HE, Bando JK, Wu D, Cheng LE, McKerrow JK, et al. Genetic analysis of basophil function in vivo. Nat Immunol 2011;12:527-35. E5. Liu FT, Bohn JW, Ferry EL, Yamamoto H, Molinaro CA, Sherman LA, et al. Monoclonal dinitrophenyl-specific murine IgE antibody: preparation, isolation, and characterization. J Immunol 1980;124:2728-37. E6. Marichal T, Starkl P, Reber LL, Kalesnikoff J, Oettgen HC, Tsai M, et al. A beneficial role for immunoglobulin E in host defense against honeybee venom. Immunity 2013;39:963-75. E7. Prouvost-Danon A, Binaghi RA, Abadie A. Effect of heating at 56 degrees C on mouse IgE antibodies. Immunochemistry 1977;14:81-4. E8. Strait RT, Morris SC, Finkelman FD. IgG-blocking antibodies inhibit IgE-mediated anaphylaxis in vivo through both antigen interception and FcgRIIb cross-linking. J Clin Invest 2006;116:833-41. E9. Ishizaka T, Helm B, Hakimi J, Niebyl J, Ishizaka K, Gould H. Biological properties of a recombinant human immunoglobulin ε-chain fragment. Proc Natl Acad Sci U S A 1986;83:8323-7. E10. Wakayama H, Hasegawa Y, Kawabe T, Saito H, Kikutani H, Shimokata K. IgG-mediated anaphylaxis via Fcg receptor in CD40-deficient mice. Clin Exp Immunol 1998;114:154-60.

12.e4 STARKL ET AL

E11. Amiri P, Haak-Frendscho M, Robbins K, McKerrow JH, Stewart T, Jardieu P. Anti-immunoglobulin E treatment decreases worm burden and egg production in Schistosoma mansoni-infected normal and interferon g knockout mice. J Exp Med 1994;180:43-51. E12. Haak-Frendscho M, Saban R, Shields RL, Jardieu PM. Anti-immunoglobulin E antibody treatment blocks histamine release and tissue contraction in sensitized mice. Immunology 1998;94:115-21. E13. Metz M, Piliponsky AM, Chen CC, Lammel V, Abrink M, Pejler G, et al. Mast cells can enhance resistance to snake and honeybee venoms. Science 2006;313:526-30.

J ALLERGY CLIN IMMUNOL nnn 2015

E14. Strait RT, Morris SC, Yang M, Qu XW, Finkelman FD. Pathways of anaphylaxis in the mouse. J Allergy Clin Immunol 2002;109:658-68. E15. Jonsson F, Mancardi DA, Kita Y, Karasuyama H, Iannascoli B, Van Rooijen N, et al. Mouse and human neutrophils induce anaphylaxis. J Clin Invest 2011;121: 1484-96. E16. Khodoun MV, Kucuk ZY, Strait RT, Krishnamurthy D, Janek K, Clay CD, et al. Rapid desensitization of mice with anti-FcgammaRIIb/FcgammaRIII mAb safely prevents IgG-mediated anaphylaxis. J Allergy Clin Immunol 2013;132: 1375-87.

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

FIG E1. Subcutaneous RVV induces systemic neutrophilia. A, Flow cytometric plots of blood neutrophils (CD451Ly6GhighF4/802). PI, Propidium iodide. B, Percentages (means 1 SDs, n 5 3) of blood neutrophils at indicated times after injection. Numbers indicate percentages of gated cells in total subpopulation. Data in Fig E1, A, are from 1 mouse per panel and representative of 3 mice per group. P values were determined as follows: Student t test. **P < .01.

STARKL ET AL 12.e5

12.e6 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

A

Inject triprolidine or PBS i.p. Minute -60

Anesthetize and inject Challenge with CV-6209 high dose RVV Assess survival or PBS i.v. -30 -5 0 1

Day 0

7 Sacrifice

Measure body temperature

B

-5

-10

100 80 60 40 20 0

0 20 40 60 90 6 24 Minutes Hours after challenge of pre-treated mice with 2x 37.5 μg RVV

** *** *** *** ** * * * *

-1

***

-2 -3 -4 0

20

40 60 90 Minutes after challenge of pre-treated mice with 2x 37.5 μg RVV

2x 37.5 μg RVV + Vehicle

% Survival

C

0 Temp (°C)

Temp (°C)

0

2x 37.5 μg RVV + Triprolidine 2x 37.5 μg RVV + CV-6209 2x 37.5 μg RVV + Triprolidine + CV-6209

01

3 6 1234567 Hours Days after challenge of pre-treated mice with 2x 37.5 μg RVV

FIG E2. Effects of triprolidine or CV-6209 pretreatment on responses to RVV in BALB/c mice. A, Experimental outline. i.p., Intraperitoneal; i.v., intravenous. B, Temperature (right panel magnifies area in dashed box). C, Survival. P values were determined as follows: Student t test (Fig E2, B) and Mantel-Cox test (Fig E2, C). Data were pooled from 2 experiments. Symbols in Fig E2, B, indicate comparison of the group in that color with vehicle-treated mice for that time point. *P < .05, **P < .01, and ***P < .001.

STARKL ET AL 12.e7

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

A Inject low dose RVV or PBS 14 Measure serum antibodies

Day 0

C 10 5 10 4 10 3 10 2 10 1 10 0

.13 ***

**

.31

2500 2000 1500 1000 500 0

Total IgE (ng/ml)

RVV-IgG1 (AU)

B

PBS RVV PBS RVV

** ***

PBS RVV PBS RVV DTA -/DTA fl/Mcpt8-Cre +/-

DTA -/DTA fl/Mcpt8-Cre +/-

D

E .66

10 10 4 10 3 10 2 10 1 10 0

***

***

Total IgE (ng/ml)

RVV-IgG1 (AU)

5

1000

PBS RVV PBS RVV +/+

Mcl-1 Mcl-1 Cpa3-Cre +

F

.78

.05

**

500 0

PBS RVV PBS RVV Mcl-1 +/+ Mcl-1 fl/fl Cpa3-Cre +

fl/fl

Inject C57BL/6 mice with low dose RVV or PBS

14 Measure serum antibodies

Day -1 0 1 3 5 7 9 11 13 PBS Assess blood neutrophils and inject i.p.

Isotype anti-GR-1

H

I.p. injection/immunization

V

V

RV

PB

1/ R-

S/

RV

S

V

an

an

ti-

G

R-

yp

1/

e/

PB

ot

S/

Is

PB

10 2

RV

RV V

10 0

10 3

Is

10

1

G

10 2

*** *

ti-

10 3

***

10 4

e/

Total IgE (ng/ml)

***

yp

ns

PB

10 4

S

RVV-IgG1 (AU)

***

ot

G

I.p. injection/immunization

FIG E3. Assessment of roles of basophils, MCs, and neutrophils in type 2 humoral responses to RVV. A, Experimental outline (Fig E3, B-E). B-E, Serum RVV-specific IgG1 (Fig E3, B and D) and total IgE (Fig E3, C and E) levels: in Fig E3, B and C, basophil-deficient Mcpt8-Cre1/2; DTAfl/2 and control basophil-sufficient Mcpt8-Cre1/2; DTA2/2 mice were used, and in Fig E3, D and E, MC- and basophil-deficient Cpa3-Cre1; Mcl-1fl/fl and corresponding control mice were used. F, Experimental outline (Fig E3, G and H). i.p., Intraperitoneal. G and H, Serum RVV-specific IgG1 (Fig E3, G) and total IgE (Fig E3, H) levels in neutrophil-depleted anti–GR-1–treated (see the Methods section for extent of neutrophil depletion), isotype antibody control–treated, and PBS-treated C57BL/6 mice. Data are pooled from 2 to 4 experiments (n 5 6-18 per group). In Fig E3, B-E, G and H, data are shown as individual values and mean 6 SEM. P values were determined as follows: Mann-Whitney test. *P < .05, **P < .01, and ***P < .001. ns, Not significant (P > .05).

12.e8 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

B

14 21 B, C Measure serum antibodies

28 Sacrifice

10 4

***

10 3 10 2 10 1 10 0

PBS RVV PBS RVV Igh-7+/+

D Temp (°C)

***

Igh-7-/-

10000 8000 6000 4000 2000 0

***

n.d. n.d. PBS RVV PBS RVV Igh-7+/+

Igh-7-/-

E 0 -2 -4 -6 -8 -10 -12 -14

Igh-7+/+ PBS Igh-7+/+ RVV Igh-7-/- PBS Igh-7-/- RVV * *

0 1 3 6 24 Hours after injection of RVV (2x 37.5 μg)

(1.5h, 6h) (1.5h, 6h)

100 80 60 40 20 0

% Survival

Day 0

C

.88

Total IgE (ng/ml)

Inject low dose RVV or PBS

Challenge with high dose RVV D, E

RVV-IgG1 (AU)

A

**

01

***

3 6 1234567 Hours Days after injection of RVV (2x 37.5 μg)

FIG E4. IgE contributes to acquired resistance to RVV-induced toxicity in BALB/c mice. A, Outline of experiments with IgE-deficient (Igh-72/2) and Igh-71/1 mice. B and C, Serum RVV-specific IgG1 (Fig E4, B) and total IgE (Fig E4, C) levels. D and E, Body temperature (Fig E4, D) and survival (Fig E4, E). Data are pooled from 3 experiments (n 5 16-19 per group). In Fig E4, B and C, data are shown as individual values and mean 6 SEM. P values were determined as follows: Mann-Whitney test (Fig E4, B-C), Student t test (Fig E4, D), and Mantel-Cox test (Fig E4, E). *P < .05, **P < .01, and ***P < .001. n.d., Not detectable.

STARKL ET AL 12.e9

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

B 1.5

1

Abs 620nm

Temp (°C)

2 0 -1 -2

0 15 30 Minutes after DNP-HSA injection

DNP-HSA (2x 0.5 μg)

Anti-DNP IgE

DNP-HSA (2x 0.5 μg)

C

*

D 1.5

2 Temp (°C)

A

Challenge

Saline

1.0 0.5

1

Abs 620nm

C57BL/6 BALB/c

Sensitization

0 -1 -2

0.0

0 15 30 Minutes after DNP-HSA injection

E

1.0

**

0.5 0.0

F 2x 37.5 μg RVV + PBS 2x 37.5 μg RVV + 0.5 μg DNP-HSA 100 80 60 40 20 0

% Survival

Temp (°C)

0 -5

-10 -15

G

0 1 3 6 24 Hours after RVV injection

01

Challenge with Inject BALB/c mice with high dose RVV and anti-DNP IgE DNP-HSA or saline Day -1

0

3 6 1234567 Hours Days after RVV injection

Assess survival 1

7 Sacrifice

Measure body temperature

H

I Saline-mock-sensitized Anti-DNP IgE-sensitized

0 1 3 6 24 Hours after injection of RVV and DNP-HSA (2x [37.5 μg RVV mixed with 0.5 μg DNP-HSA])

100 80 60 40 20 0

% Survival

*

Temp (°C)

0 -2 -4 -6 -8 -10

*

01

3 6 1234567 Hours Days after injection of RVV and DNP-HSA (2x [37.5 μg RVV mixed with 0.5 μg DNP-HSA])

FIG E5. Passive cutaneous anaphylaxis to an irrelevant antigen can increase resistance to a potentially lethal challenge with RVV. A-D, C57BL/6 and BALB/c mice received 3 subcutaneous injections of 50 mL of saline alone or containing 50 ng of anti-DNP IgE antibody on the shaved back skin. Eighteen hours later, mice were injected intravenously with 200 mL of PBS containing 1% Evan blue dye and challenged 30 minutes later by means of 2 subcutaneous injections of 50 mL of PBS containing 0.5 mg of DNP-HSA administered to the site of prior injection of anti-DNP IgE or saline. Fig E5, A and C, Body temperature after DNP-HSA challenge. Fig E5, B and D, Extravasation of intravascular fluid into the skin at the site of DNP-HSA challenge 1 hour after challenge, as assessed based on levels of Evan blue dye. E and F, Body temperature (Fig E5, E) and survival (Fig E5, F) of C57BL/6 mice after 2 subcutaneous injections of 37.5 mg of RVV mixed with 0.5 mg of DNP-HSA or vehicle. G, Experimental outline for Fig E5, H and I. H and I, Body temperature (Fig E5, H) and survival (Fig E5, I) of BALB/c mice treated with 3 subcutaneous injections of 50 mL of saline alone or containing 50 ng of anti-DNP IgE antibody and challenged 18 hours later with 2 subcutaneous injections, each of 50 mL of PBS containing 37.5 mg of RVV and 0.5 mg of DNP-HSA. Data are pooled from 2 to 3 experiments (n 5 6-15 per group). In Fig E5, B and D, data are shown as individual values and mean 6 SEM. P values were determined as follows: Student t test (Fig E5, B, D, and H) and Mantel-Cox test (Fig E5, I). *P < .05 and **P < .01.

12.e10 STARKL ET AL

J ALLERGY CLIN IMMUNOL nnn 2015

A Inject C57BL/6 Igh7-/- mice with 3x 50 μl PBS containing Challenge with high dose either 50 ng anti-DNP IgE, RVV and 500 ng anti-OVA IgE, DNP-HSA or both Day -1

Assess survival

0

1

7 Sacrifice

Measure body temperature

B

C anti-DNP-IgE sensitized anti-DNP-IgE + anti-OVA-IgE sensitized anti-OVA-IgE sensitized

-5 *

-10 -15

0 1 3 6 24 Hours after injection of RVV and DNP-HSA (2x [37.5 RVV DNP-HSA]) mixed with 0.5

(3 h)

100 80 60 40 20 0

% Survival

Temp (°C)

0

*

01

3 6 1234567 Hours Days after injection of RVV and DNP-HSA (2x [37.5 μg RVV mixed with 0.5 μg DNP-HSA])

FIG E6. Eliciting passive cutaneous anaphylaxis–mediated resistance against RVV depends on using an antigen-specific IgE. A, Experimental outline. B and C, Body temperature (Fig E6, B) and survival (Fig E6, C) of IgE-deficient (Igh-72/2) C57BL/6 mice treated with 3 subcutaneous injections of saline alone or containing 50 ng of anti-DNP IgE alone, 50 ng of anti-DNP IgE mixed with 500 ng of anti-OVA IgE, or 500 ng of anti-OVA IgE antibody alone and challenged 18 hours later with 2 subcutaneous injections, each containing 37.5 mg of RVV and 0.5 mg of DNP-HSA. Data are pooled from 3 independent experiments (n 5 7 per group). P values were determined as follows: Student t test (Fig E6, B) and Mantel-Cox test (Fig E6, C). *P < .05.

STARKL ET AL 12.e11

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

at tre

un

μg 1

ed

l

l

/m

/m

l

μg

/m

5

μg 10 BV

Serum source Immunization PBS-PBS C57BL/6 BV-PBS BV-BV PBS-PBS BALB/c BV-PBS BV-BV C57BL/6 BALB/c BMCMCs

*** ** *** **

Stimulation with 5 μg/ml BV % β-hexosamindase release

% β-hexosamindase release

Stimulation with 10 μg/ml BV 50 40 30 20 10 0

ns ns

Stimulant

Stimulant

B

*** ** ** **

BV

% β-hexosamindase release

at

/m

tre un

μg 1

50 40 30 20 10 0

ed

l

l /m μg

/m

5

μg BV

10 BV

BALB/c-derived BMCMCs Serum source Immunization PBS-PBS C57BL/6 BV-PBS BV-BV PBS-PBS BALB/c BV-PBS BV-BV

* ** * **

BV

50 40 30 20 10 0

* ** ns ns ns ns

l

% β-hexosamindase release

C57BL/6-derived BMCMCs

BV

A

50 40 30 20 10 0 C57BL/6 BALB/c BMCMCs

FIG E7. BV induces similar degranulation responses in BMCMCs derived from C57BL/6 or BALB/c mice after their sensitization with BV-immune serum derived from either C57BL/6 or BALB/c mice. A and B, BMCMCs were sensitized overnight with pooled immune serum collected on day 42 from PBS-PBS–, BV-PBS–, or BV-BV–immunized C57BL/6 or BALB/c mice (these were the same sera used for measurements of phospholipase A2–specific IgE in Fig 7, F-G). Percentage of b-hexosaminidase released was measured 1 hour after exposure to BV. Fig E7, B, shows selected data from Fig E7, A, to allow side-by-side comparison of results from C57BL/6 and BALB/c BMCMCs; none of the responses to any of the concentrations of BV tested were significantly different in the BMCMCs derived from the 2 mouse strains. Data (mean 6 SEM, n 5 3) were pooled from 3 independent experiments by using the same batch of BMCMCs tested on different days but each of the 3 experiments with a different batch of pooled serum from each of the 3 independent immunizations. P values were determined as follows: Student t test. *P < .05, **P < .01, and ***P < .001. ns, Not significant (P > .05).