Omeprazole inhibits IgE-mediated mast cell activation and allergic inflammation induced by ingested allergen in mice

Journal Pre-proof Omeprazole inhibits IgE-mediated mast cell activation and allergic inflammation induced by ingested allergen in mice Cynthia Kanagaratham, PhD, Yasmeen S. El Ansari, MMSc, Benjamin F. Sallis, BSc, Brianna-Marie A. Hollister, BA, Owen L. Lewis, BS, Samantha C. Minnicozzi, MD, Michiko K. Oyoshi, PhD, Rachel Rosen, MD, MPH, Samuel Nurko, MD, Edda Fiebiger, PhD, Hans C. Oettgen, MD, PhD PII:

S0091-6749(20)30342-0

DOI:

https://doi.org/10.1016/j.jaci.2020.02.032

Reference:

YMAI 14452

To appear in:

Journal of Allergy and Clinical Immunology

Received Date: 10 October 2019 Revised Date:

25 February 2020

Accepted Date: 26 February 2020

Please cite this article as: Kanagaratham C, El Ansari YS, Sallis BF, Hollister B-MA, Lewis OL, Minnicozzi SC, Oyoshi MK, Rosen R, Nurko S, Fiebiger E, Oettgen HC, Omeprazole inhibits IgEmediated mast cell activation and allergic inflammation induced by ingested allergen in mice, Journal of Allergy and Clinical Immunology (2020), doi: https://doi.org/10.1016/j.jaci.2020.02.032. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology.

1

Omeprazole inhibits IgE-mediated mast cell activation and allergic inflammation induced

2

by ingested allergen in mice

3 4

Cynthia Kanagaratham, PhD1,2, Yasmeen S. El Ansari, MMSc1,2, Benjamin F. Sallis, BSc1,

5

Brianna-Marie A. Hollister, BA1, Owen L. Lewis, BS1, Samantha C. Minnicozzi, MD1,2, Michiko K.

6

Oyoshi, PhD1,2, Rachel Rosen1,2, MD, MPH, Samuel Nurko1,2, MD, Edda Fiebiger, PhD1,2,* and

7

Hans C. Oettgen, MD, PhD1,2,*

8 9

1

10

2

11

*These authors contributed equally to this work.

Department of Pediatrics, Boston Children’s Hospital, Boston, MA, 02115, USA. Department of Pediatrics, Harvard Medical School, Boston, MA, 02115, USA.

12 13

Conflicts: RR is a consultant for Takeda. All other authors declare that they have no relevant

14

conflicts of interest.

15 16

Corresponding author:

17

[email protected]

18

1 Blackfan Circle

19

Boston, MA 02115

20

USA

21

Phone: 1 (617) 919-6842

22

Fax: 1 (617) 730-0528

23 24

Research in this report was funded by NIH, grants 1R01AI119918-05 (HCO), DK097112-05 (RR)

25

and 5T32AI007512-33 (SCM). CK is supported by a postdoctoral fellowship from the Fonds de

26

recherche du Québec - Santé. EF was supported by a Bridge Grant from the Research Council

27

of Boston Children's Hospital, an Emerging Investigator Award from FARE, a Senior Research

28

Award from the Crohn's and Colitis Foundation, and an unrestricted gift from the Mead Johnson

29

Nutrition Company. This work was further supported by an NIH grant of the Harvard Digestive

30

Diseases Center (P30DK034854, Core C). Additional generous support was provided by a Sean

31

N. Parker Center Seed Grant, the Bunning Family Foundation, the Nanji Family Fund for Food

32

Allergy Research, the Rao Chakravorti Family Fund, and the Christine Olsen and Robert Small

33

Food Allergy Research Fund.

2

34

Abstract

35

Background: Patients with eosinophilic esophagitis have increased numbers of mucosal mast

36

cells. Administration of the proton pump inhibitor omeprazole can reduce both esophageal mast

37

cell and eosinophil numbers and attenuate type 2 inflammation in these subjects.

38

Objective: Given that maintenance of an acidic environment within granules is important for

39

mast cell homeostasis, we sought to evaluate the effects of omeprazole on mast cell functions

40

including development, IgE:FcεRI-mediated activation and responses to food allergen.

41

Methods: Mast cell degranulation, cytokine secretion and early signaling events in the FcεRI

42

pathway, including protein kinase phosphorylation and Ca2+ flux, were measured following IgE

43

crosslinking in murine bone marrow-derived mast cells and human cord blood-derived mast

44

cells. The effects of omeprazole on these responses were investigated as was its impact on

45

mast cell-dependent anaphylaxis and food allergy phenotypes in vivo.

46

Results: Murine and human mast cells treated with omeprazole exhibited diminished

47

degranulation and release of cytokines and histamine in response to allergen. In murine mast

48

cells, phosphorylation of protein kinases, ERK and SYK, was decreased. Differentiation of mast

49

cells from bone marrow progenitors was also inhibited. IgE-mediated passive anaphylaxis was

50

blunted in mice treated with omeprazole as was allergen-induced mast cell expansion and mast

51

cell activation in the intestine in a model of food allergy.

52

Conclusion: Our findings suggest that omeprazole targets pathways important for the

53

differentiation and activation of murine mast cells and for the manifestations of food allergy and

54

anaphylaxis.

55 56 57 58

3

59 60

Key Messages:

61

-Omeprazole blocks IgE-mediated mast cell degranulation, and prostaglandin D2 and cytokine

62

production in response to allergen as well as IgE-mediated hypersensitivity in vivo.

63

-Pretreatment with omeprazole results in decreased phosphorylation of protein kinases and

64

inhibits calcium flux into the cytosol.

65

-In a food allergy model, omeprazole inhibits mast cell expansion, type 2 allergic inflammation

66

and hypersensitivity responses to ingested allergen.

67 68

Capsule Summary:

69

Omeprazole inhibits IgE-mediated mast cell activation and allergic responses to food allergen.

70 71

Key Words: food allergy, anaphylaxis, mast cell, omeprazole, proton pump inhibitor

72 73

Abbreviations: OVA, ovalbumin; MCPT-1, mouse mast cell protease; EoE, eosinophilic

74

esophagitis; PPI, proton pump inhibitor

75

4

76 77

Introduction Food allergy is a major health problem in industrialized countries and its prevalence has

78

been increasing over the past decades1, 2. There is a lack of understanding of the mechanisms

79

leading to increased immunological sensitivity to foods in affected subjects and there are no

80

cures for the disease. Patients are advised to practice allergen avoidance and manage acute

81

reactions that arise after accidental exposures using injectable epinephrine2, 3. Insight into the

82

cellular and molecular pathways regulating hypersensitivity reactions to foods are badly needed

83

and may lead to innovative strategies for treatment. In its most common and severe form, food

84

allergy is mediated by the recognition of food antigens by food-specific IgE antibodies.

85

Reactions are triggered upon IgE:allergen-mediated activation of mast cells which harbor the

86

high affinity IgE receptor, FcεRI, on their cell surface4. Crosslinking of the receptor leads to

87

degranulation of mast cells and release of inflammatory mediators that lead to clinical

88

phenotypes including gastrointestinal responses (oral pruritus, abdominal pain, vomiting,

89

diarrhea) and systemic anaphylaxis including vasodilation and vascular leak with tissue edema.

90

Together, these physiologic changes can lead to decreased blood pressure with eventual shock

91

and to respiratory failure3.

92

The prevalence of chronic inflammatory forms of immunologically-mediated food

93

sensitivity is increasing. Many of these can belong to the category of eosinophilic

94

gastrointestinal disorders. The most common is eosinophilic esophagitis (EoE), which is

95

characterized by an elevated number of eosinophils in the esophagus and is one of the leading

96

causes of food impaction and esophageal strictures5-7. There is clearly overlap between the two

97

types of food sensitivity. For instance, in patients affected by EoE, tissue mast cell homeostasis

98

is dysregulated, and IgE-mediated food allergies are often also present8. Increased mast cell-

99

associated transcripts, including, for example, CPA3 and TPSB2, and mast cell numbers are

100

typically present in the esophageal tissue of EoE patients9-12. In a subset of EoE patients,

101

symptoms of esophageal dysfunction and eosinophilia are relieved following a course of PPI 5

102

therapy, giving rise to the term “PPI-responsive EoE”13, 14. Although PPIs are used primarily to

103

treat esophagitis and gastritis, they have also been shown to ameliorate symptoms, reverse

104

tissue inflammation, and normalize the allergic transcriptome this subset of patients. Moreover,

105

PPI treatment of subjects with PPI-responsive EoE results in a decrease in mast cell numbers

106

with RNA analysis showing a reduction in mast cell transcripts 11.

107

By covalently binding to the cysteine residues of H+/K+ ATPases of the parietal cells,

108

PPIs block the secretion of gastric acid, subsequently changing the pH of the cell and stomach

109

lumen16. Mast cells are known to be sensitive to changes in pH. The granules of the cell are

110

acidic compartments analogous to lysosomes17. Inhibition of proton pumps in mast cells by

111

bafilomycin A1, a selective inhibitor of vacuolar ATPase, affects the acidification of the granules

112

and consequently the processing and activity of granule contents18. Based on these

113

observations, we reasoned that PPIs might similarly affect the pH of mast cell compartments

114

and, consequently, the effector functions of the cells. We tested our hypothesis in in vitro and in

115

vivo, analyzing the effects of PPIs on the functions of cultured mast cells and in mast cell-

116

dependent mouse models of anaphylaxis and food allergy, respectively. Our results provide

117

evidence that PPIs interact with targets that mediate activating signals provided by FcεRI

118

ligation in mast cells, inhibit mediator release and attenuate the physiologic and inflammatory

119

responses to mast cell activation in vivo.

120

Methods

121

Murine mast cell culture. Bone marrow-derived mast cells (BMMCs) were cultured as

122

previously described19. Briefly, BMMCs were derived from bone marrow precursor cells of

123

BALB/c mice. Cells were cultured in RPMI-1640 medium supplemented with 10% FCS (Gibco),

124

100U/ml penicillin (Gibco), 100µg/ml streptomycin (Gibco), 1% Minimum Essential Medium non-

125

essential amino acids (Gibco), 10mM HEPES buffer (Gibco), 55μM 2-mercaptoethanol (Gibco),

126

10μg/ml gentamicin (Life Technologies), and 20ng/mL each of IL-3 and SCF (complete media).

6

127

Once cultures were mature (>90% of cells c-Kit+ FcεRIα+) cytokines were reduced to 10ng/ml

128

(gating strategy for mast cells is provided in Fig E1A). IgE-induced degranulation (LAMP-1

129

expression), cytokine release, calcium flux, protein tyrosine phosphorylation and changes in

130

granule and cytosolic pH were determined as detailed in the Online Repository.

131

Cord blood-derived human mast cell culture. Cord blood-derived mast cells (CBMCs) were

132

generously provided by the laboratory of Dr. J. Boyce (Brigham and Women’s Hospital, Boston,

133

MA). Human mast cells were derived from cord blood mononuclear cells as previously

134

described20. Mature human mast cells were always maintained and cultured in the presence of

135

SCF (100ng/mL). For activation experiments, human mast cells were primed with IL-4 (10ng/mL)

136

for three days in upregulate their surface expression of FcεRI. On the third day of IL-4

137

treatment cells were incubated with human myeloma IgE (100ng/ml; Chemicon International,

138

Tecaluma, CA). Unbound IgE was washed away and cells were stimulated with rabbit anti-

139

human IgE (100ng/ml). Details quantification of degranulation by measuring LAMP-1 expression,

140

and release of histamine, prostaglandin D2 and cytokines are detailed in the Online Repository.

141

Treatment of mast cell cultures with omeprazole. Omeprazole solid (Sigma, catalog number

142

O104) was first activated for 30 minutes in PBS at a pH of 4 to convert it to the active

143

sulfonamide form. An equal volume of complete media was added to the activated omeprazole

144

to neutralize the pH before adding it to cell cultures. Cells were treated with omeprazole at a

145

final concentration of 50µM as previously reported21-23. Drug vehicle (DMSO) was prepared in

146

the same manner, including acidification and neutralization, at equal volumes.

147

Mouse studies. All work was performed under protocols reviewed and approved by Boston

148

Children’s Hospital Institutional Animal Care and Use Committee. All mice used in this study

149

were bred and maintained under specific pathogen-free conditions in individually ventilated

150

cages. To induce food allergy, 6-8-week-old BALB/c mice were sensitized for three consecutive

151

weeks by intraperitoneal injection of 100µg of ovalbumin (OVA; Sigma, catalog number A5503)

7

152

adsorbed to 1.5mg of aluminum hydroxide (Imject Alum, Pierce, Rockford, IL) in 0.2mL of sterile

153

PBS. Following sensitization, animals were divided into two groups: omeprazole- or vehicle-

154

treated. Omeprazole treatment was administered intragastrically at a dose of 12.5mg/kg per

155

mouse (prepared in sterile PBS), as previously described23, and untreated animals were given

156

equal volumes of vehicle (DMSO). Two hours post treatment, the mice were intragastrically

157

challenged with 50mg of ovalbumin. Animals were re-challenged two days later and sacrificed

158

two hours following the second challenge. Tissues were collected for follow-up experiments as

159

described in the Online Repository. For passive systemic anaphylaxis experiments, female

160

BALB/c mice were treated with 1.6mg esomeprazole in 0.2mL (a formulation of the drug for

161

intravenous injection in humans) or with vehicle (PBS) for 4 consecutive days by intraperitoneal

162

injection. On the third day, mice were intraperitoneally sensitized with 10µg IgE anti-DNP (clone

163

SPE-7, Sigma-Aldrich, St. Louis, MO). Two hours after the final treatment mice were challenged

164

intraperitoneally with 75µg DNP-BSA. Body temperature was recorded every 5 minutes for 60

165

minutes via implanted transponders24.

166

Mast cell differentiation assay. To study the effects of omeprazole on the differentiation of

167

progenitor cells to BMMCs, cells were treated with 50µM of omeprazole. Media were changed

168

every three days and differentiation was assessed every six days by measuring the percentage

169

of c-Kit and FcεRIα double-positive cells by flow cytometry.

170

Interleukin 4 (IL-4)-induced mast cell expansion in the small intestine. Mice were injected

171

with complexes of 2µg of IL-4 (Peprotech) and 10µg of anti-IL-4 (Biolegend) on days 0, 3 and 6.

172

Mice were treated with esomeprazole daily from day -1 to day 6 by intraperitoneal injection.

173

Blood and small intestine were collected on day 7.

174

8

175

Results

176

Omeprazole blocks the degranulation of bone marrow-derived murine mast cells and cord

177

blood-derived human mast cells

178

To test the effects of omeprazole on mast cells, we first examined its effects on IgE-

179

mediated mast cell degranulation and cytokine production using cultured cells. Surface

180

expression of lysosomal-associated membrane protein 1 (LAMP-1) following stimulation with

181

antigen for 10 minutes was used as an indicator of the degree of mast cell degranulation. At

182

baseline, in the absence of antigen-specific IgE (IgE anti-TNP), exposure of BMMCs to antigen

183

(TNP-OVA) had no effect, and LAMP-1 expression was minimal (2.293% ± 0.5031%) (Fig 1A).

184

As expected, upon stimulation with TNP-OVA, mast cells sensitized with IgE anti-TNP exhibited

185

nearly a 30-fold increase in LAMP-1 expression on their surface (60.95% ± 5.118%). In contrast,

186

IgE-sensitized BMMCs preincubated with 50µM omeprazole for two hours prior to stimulation

187

with TNP-OVA exhibited markedly suppressed IgE-induced degranulation (8.775% ± 0.7180%

188

LAMP-1+). and histamine release (Fig 1B). A dose response analysis revealed that suppression

189

of degranulation by omeprazole was exerted in a dose-dependent manner over a range of 10 to

190

50µM in BMMCs (Fig E1B). The concentration of omeprazole used did not affect the viability of

191

murine mast cells (Fig E1C).

192

In order to test whether omeprazole similarly affects activation of human mast cells we

193

took advantage of human cord blood-derived mast cells. These were sensitized with myeloma

194

IgE and activated with anti-IgE in the presence or absence of omeprazole. As observed with

195

murine BMMC, LAMP-1 expression and histamine release were suppressed by omeprazole in a

196

dose dependent manner (Fig 2 A and B).

197

Though crosslinking of the IgE receptor is the best-known trigger of mast cell

198

degranulation and the one likely to mediate immunologically specific responses to foods,

199

stimulation through other receptors can also activate mast cells. We next tested if omeprazole

9

200

can block non-IgE receptor-mediated mast cell activation, such as through G-protein coupled

201

adenosine receptors. Stimulation of BMMCs with 100µM adenosine caused a 6-fold increase in

202

LAMP-1 expression compared to unstimulated cells (25.03% ± 1.313% vs 3.847% ± 0.777%,

203

Fig 1C). Treatment with omeprazole prior to adenosine stimulation restrained surface LAMP-1

204

at levels similar to those observed in unstimulated cells (3.310% ± 1.640%).

205

Upon IgE activation, transcription and translation of pro-inflammatory cytokines that

206

contribute to allergic inflammation also occurs in mast cells. To assess these transcriptional

207

changes, we used a predesigned NanoString panel composed of probes for the assessment of

208

genes related to Th2 inflammation25. Transcriptional profiling of allergen-stimulated, IgE-

209

sensitized BMMCs performed two hours following antigen exposure revealed the induction of a

210

number of genes, including some encoding cytokines associated with allergic inflammation (Fig

211

1D). Pretreatment with omeprazole at 50µM for two hours attenuated the IgE:FcεRI-induced

212

expression in BMMCs of pro-inflammatory genes such as, CCL5, IL-1β, IL-18, TGFβ, and IL-6

213

as well as IL-13 which is specifically related to type 2 inflammation. We further investigated the

214

effect of omeprazole on select mast-cell-specific cytokines that are known to be synthesized de

215

novo and released from the cells. Six hours after allergen challenge, IgE-sensitized BMMCs

216

produced significantly greater amounts of IL-4, IL-6, IL-13 and TNF-α than unstimulated cells

217

(Fig 1E-H). In cells pretreated with omeprazole, the production and release of these cytokines

218

was reduced. Likewise, in human mast cell cultures we observed a dose responsive decrease in

219

the secretion of PGD2 and cytokines, IL-5 and IL-13 (Fig 2C-E). Taken together, these results

220

show that omeprazole treatment blocks the immediate release of preformed mediators

221

contained in mast cell granules, the FcεRI-induced transcriptional program and the eventual

222

production and secretion of cytokines.

10

223 224

Omeprazole treatment dampens activating signaling pathways in mast cells To obtain a better understanding of how omeprazole attenuates mast cell activation, we

225

investigated early events in the well-characterized signaling cascade activated by FcεRI

226

crosslinking. This pathway drives both degranulation with release of preformed mediators of

227

anaphylaxis and the upregulation of inflammatory cytokine transcription and secretion4. To test

228

the effect of omeprazole on signaling, we stimulated IgE anti-TNP bound BMMCs via FcεRI

229

crosslinking with TNP-OVA for 2.5 to 30-minutes with or without pretreatment with omeprazole.

230

The degree of phosphorylation of upstream (SYK) and downstream (ERK) protein kinases was

231

measured to assess the strength of activation of the pathway. As previously extensively

232

described by others26, 27, crosslinking of the IgE receptor resulted in phosphorylation of both

233

SYK and ERK detectable as early as 2.5 minutes after stimulation (Fig 3A). Quantification of the

234

ratio in signal intensities of phosphorylated to total protein showed that the IgE-induced

235

phosphorylation was significantly weaker at all time points in omeprazole-treated cells following

236

antigen stimulation.

237

Another early signal critical for degranulation and cytokine production by mast cells is

238

increased cytosolic calcium. FcεRI crosslinking is associated with the release of calcium from

239

endoplasmic reticulum stores into the cytosol. This is necessary for granule fusion with the

240

plasma membrane and exocytosis of the contents28. Blocking calcium increase can disrupt the

241

full range of mast cell activation phenotypes following IgE receptor crosslinking. Using the Ca2+-

242

chelating fluorophore, Fluo 4, we observed a robust increase in cytosolic Ca2+ ([Ca2+]i)

243

immediately after antigen stimulation (Fig 3B). Compared to IgE-sensitized untreated BMMCs,

244

omeprazole-treated mast cells exhibited a dramatic suppression of increase in [Ca2+]i in a dose

245

responsive manner. These results show that omeprazole exerts suppressive effects on several

246

of the key signaling events elicited by cross-linking of FcεRI.

11

247 248

Omeprazole dampens passive IgE-mediated systemic anaphylaxis Systemic anaphylaxis is the most dramatic clinical outcome of an allergic reaction and is

249

caused by the multi-tissue effects of mediators released by activated mast cells and basophils

250

following crosslinking of the cell surface IgE receptor24, 29-31. Our in vitro data showed that

251

treatment of mast cells with omeprazole blocked their IgE:FcεRI-induced degranulation. Using a

252

model of passive systemic anaphylaxis, we assessed if the inhibition observed in vitro is

253

recapitulated in vivo and whether diminished mast cell activation in the presence of omeprazole

254

is accompanied by attenuation of the physiologic manifestations of anaphylaxis. Mice were

255

passively sensitized with anti-DNP IgE and challenged intraperitoneally with DNP-BSA.

256

Changes in core body temperature were measured as an indicator of vasodilation (with

257

increased cutaneous blood flow and cooling) and shock. As expected, untreated BALB/c mice

258

exhibited a significant drop in temperature following antigen challenge (Fig 4A). To investigate

259

the effects of proton pump inhibitors on anaphylaxis, we used esomeprazole which is a single (S)

260

enantiomer of omeprazole and available as an injectable formulation. Esomeprazole treatment

261

for 4 days prior to challenge significantly dampened the rate of temperature decline to less than

262

half of control. Release of the mast cell granule constituent protease, MCPT-1, and histamine

263

were also reduced to less than 50% that observed in esomeprazole-treated animals (Fig 4B

264

and C).

265

Treatment with omeprazole alters the pH in the cytosol and subcellular compartments of the

266

mast cell

267

Based on the known action of omeprazole as a proton pump inhibitor in gastric parietal

268

cells, we sought to determine whether it affects the pH of the acidic storage compartments and

269

cytosol of mast cells. Using Lysosensor Blue, a dye that accumulates in acidic intracellular

270

compartments, we assessed the effects of omeprazole on pH in mast cell granules. As a control

271

we treated BMMCs with bafilomycin, a blocker of V-ATPase that has previously been shown to

272

increase the pH of mast cell granules18. In comparison to vehicle-treated cells, the fluorescence 12

273

intensity of Lysosensor Blue in both omeprazole- and bafilomycin A1-treated cells was

274

significantly decreased, indicating that the intracellular granule compartment pH was increased

275

(Fig 5A).

276

In an alternate approach to detecting changes in the pH of intracellular vesicles, we

277

performed a FITC-dextran quenching experiment. FITC-dextran is taken up into endosomes

278

which fuse with acidic intracellular vesicles. The FITC signal is quenched by the acidic pH within

279

those compartments. Chloroquine, a blocker of endosome-lysosome fusion, was included as a

280

control for this experiment. Compared to untreated cells, both chloroquine- and omeprazole-

281

treated cells exhibited higher FITC fluorescence intensity than did untreated cells, consistent

282

with an increased pH in those normally acidic organelles (Fig 5B). As granules represent a

283

significant fraction of mast cell volume, we reasoned that inhibition of proton accumulation in

284

those compartments might be accompanied by an opposite effect, namely reduction of pH, in

285

the cytosol. We tested this hypothesis using pHrodoGreen, a dye that becomes trapped in the

286

cytosol and whose fluorescence is inversely proportional to the pH of the cytosol. The

287

pHrodoGreen assay showed that BMMCs treated with omeprazole had a more acidic cytosol

288

environment compared to untreated cells (Fig 5C). Notably, the cytosolic pH was unaffected by

289

pretreatment with bafilomycin. Furthermore, mast cells treated with bafilomycin have previously

290

been characterized as having altered morphology with swollen granules that have diminished

291

staining with May Grünwald/Giemsa (MGG) 18. In our hands, we observed similar changes (Fig

292

E2). This dramatic phenotype was not observed in omeprazole treated cells. However, unlike

293

the DMSO-treated control cells, they did exhibit a few punctate MGG negative regions (Fig E2).

294

These findings suggest that the effects of omeprazole on mast cell function might involve the

295

target of bafilomycin, the V-ATPase or a related proton pump.

296

Omeprazole blocks mast cell differentiation in vitro and in vivo

297

As our findings had shown that treatment with omeprazole affects the signaling

298

pathways and activation of mature mast cells, we next evaluated whether they might affect the

13

299

differentiation of precursor cells into mature mast cells. Culturing murine bone marrow stem

300

cells in the presence of IL-3 and SCF induces differentiation into mature mast cells. Mast cell

301

maturation in such cultures can be assessed by monitoring the appearance of c-Kit and FcεRIα

302

double-positive cells (Fig E1A). Under normal growth conditions, more than 80% (86.65% ±

303

1.350%) of cells in such cultures acquire the double-positive mast cell phenotype after five

304

weeks (Fig 6A). However, less than 15% (12.75% ± 0.75%) of cells cultured in the presence of

305

50µM of omeprazole were mature. In order to determine whether omeprazole might similarly

306

affect mast cell differentiation in vivo, we took advantage of an IL-4-driven model. We and

307

others have reported that exogenous administration of IL-4 as an immune complex with anti-IL4

308

(IL4C) to prolong the cytokine’s half-life (three injections over one week), elicits an expansion of

309

intestinal mast cells (Fig 6B)19, 32. Using this approach along with flow cytometric enumeration of

310

intestinal mast cells, we found that daily treatment with esomeprazole by intraperitoneal

311

injection decreased IL-4-induced mast cell expansion (Fig 6B). These in vivo data corroborate

312

our tissue culture findings in showing that omeprazole can block the maturation of precursor

313

cells into mast cells.

314

Treatment with omeprazole blocks allergic responses in a mast cell-dependent mouse model of

315

food allergy

316

In addition to driving immediate hypersensitivity reactions, mast cells are known to play

317

key roles in initiating and sustaining allergic inflammatory responses19, 31, 33, 34. They provide a

318

critical tissue source of cytokines that activate mucosal antigen presenting cells, as well as IL-4

319

which primes and consolidates T-helper 2 responses. Our analysis of the effects of omeprazole

320

on cultured mast cells showed a dramatic suppression of cytokine production following IgE-

321

mediated activation by allergen (Fig 1C-D). Our final aim was to investigate whether this effect

322

would be reflected in vivo by an altered response to food allergen. To test this possibility, we

323

adapted a well-characterized mast cell-dependent model of food allergy initially described by

14

324

Rothenberg and colleagues (Fig 7A)31. In this model, mice sensitized intraperitoneally with

325

allergen and then repeatedly enterally challenged to develop strong IgE responses and food

326

sensitivity. The response to repeated allergen challenge was accompanied by intestinal mast

327

cell expansion and a correlation has been demonstrated between the number of mast cells in

328

the small intestine and the severity of allergic reactions29.

329

Examination of chloroacetate esterase (CAE)-stained jejunal sections revealed a robust

330

expansion of mast cells following ovalbumin challenge in mice that had been sensitized to the

331

antigen (Fig 7D), but not in unsensitized controls (Fig 7B and C). This was reduced by

332

omeprazole treatment (Fig 7E). Enumeration of mast cells in sections from multiple mice

333

confirmed these observations (Fig 7F). Following allergen challenge by gavage, omeprazole-

334

treated mice also exhibited greatly reduced release of MCPT-1 into the serum (Fig 7G). The

335

increase in total and allergen specific IgE (OVA-IgE) in the serum following antigen challenge in

336

sensitized mice was attenuated in omeprazole treated mice (Fig 7 H and I). mRNA expression

337

of several cytokines associated with allergic inflammation, IL-4, IL-5, IL-13 and TNF-α, was also

338

reduced in omeprazole-treated, allergen-sensitized and challenged animals compared to

339

untreated animals (Fig 7 J-M). These findings indicate that omeprazole affects pathways

340

involved in mast cell differentiation and activation in response to ingested allergen and alters the

341

intestinal cytokine environment.

342

15

343 344

Discussion Our study provides strong evidence that the proton pump inhibitor omeprazole exerts

345

significant effects on pathways regulating mast cell homeostasis and function and can modulate

346

mast cell-dependent allergic disease phenotypes. We demonstrate that omeprazole attenuates

347

mast cell activation induced by FcεRI crosslinking and by adenosine, with decreased

348

degranulation and cytokine responses. Suppression occurs early in the signaling cascade with

349

decreased phosphorylation of the FcεRI-proximal tyrosine protein kinase SYK, as well as the

350

downstream serine protein kinase ERK, along with attenuation of increased [Ca2+]i. The

351

physiologic impact of the inhibitory effects of omeprazole on mast cell function were clearly

352

evident in in vivo models of anaphylaxis and mast cell-dependent food allergy. Furthermore, we

353

observe that omeprazole at 200µM can also block the activation of basophils, another major cell

354

involved in allergic phenotypes and anaphylaxis (Fig E3).

355

PPIs have been reported to have clinical benefit in eosinophilic esophagitis35. It was

356

initially postulated that the omeprazole responsiveness of some subjects with distal esophageal

357

eosinophilia indicated that their esophageal inflammation was triggered by acid reflux and not

358

food-induced type 2 inflammation. However, there is evidence that the mechanism of action of

359

omeprazole can be attributed to anti-type 2 inflammatory effects. For instance, oral

360

administration of omeprazole prior to allergen instillation into mouse airways reduced the

361

recruitment of inflammatory cells, including eosinophils21. In PPI-responsive EoE, there is now

362

general agreement, based largely on transcriptomic analyses, that omeprazole treatment

363

attenuates type 2 inflammation11. There is evidence that omeprazole impacts a range of

364

signaling intermediates. For instance, studies on epithelial cells have shown that omeprazole

365

decreases IL-4-, IL-13-, and IFN-ß-induced phosphorylation of STAT621.

366

The mechanism whereby omeprazole exerts its effects on mast cells remains to be

367

determined. On gastric parietal cells, omeprazole binds to and inhibits the gastric ATPase.

16

368

However, we have found that expression of this proton pump is quite low in mast cells. We

369

speculate that omeprazole could be binding to the V-ATPase, the target of bafilomycin A1.

370

Treatment with bafilomycin has previously been shown to alter mast cell morphology, inhibit the

371

maturation of mast cell granules and processing granule contents and impair mast cell

372

activation18. Furthermore, it has been reported that omeprazole binds to the catalytic sites of V-

373

ATPases purified from adrenal chromaffin granules and reconstituted into liposomes36. In our

374

hands, bafilomycin and omeprazole exerted similar effects on mast cell granule pH (Fig 5 A).

375

Yet, we have reason to speculate that additional targets for omeprazole might exist in mast cells

376

as omeprazole, but not bafilomycin, affects cytosolic pH (Fig 5C) and changes observed in

377

granule morphology are not identical by both drugs (Fig E2).

378

Several studies describe a link between the use of acid suppression and the

379

development of food allergies37. Elevating the pH of the stomach can decrease the activity of

380

stomach enzymes, such as pepsin, that are needed for protein digestion. It is hypothesized that

381

incomplete digestion might allow larger peptides harboring allergenic epitopes to enter the small

382

intestine and induce antigen sensitization38. In our studies, omeprazole had an anti-allergic

383

effect, but it was only administered once sensitization was established. Animals are sensitized

384

to either ovalbumin or with anti-DNP IgE via the intraperitoneal route, and so our experimental

385

design allows us to study the effects of omeprazole on allergen challenge-induced inflammation

386

and avoid its effects on sensitization.

387

Other research groups have also described anti-inflammatory effects of PPIs. For

388

example, omeprazole has been shown to decrease the migration of polymorphonuclear

389

neutrophils (PMN) towards the chemoattractant IL-839. Correlating with this decrease in

390

migration, a decrease in the expression of adhesion molecules CD11b and CD18 on PMNs has

391

also been observed40. Furthermore, production of IL-8 by epithelial cells following stimulation

392

with IL-1β is also inhibited39. These effects are beneficial in attenuating the inflammation

393

observed in infections, such as from Helicobacter pylori. In models of sepsis, injection with 17

394

omeprazole pre- or post-endotoxic shock can increase survival time by limiting cytokine

395

production23. Treatment of monocytes and animals with omeprazole inhibits TLR agonist-

396

induced secretion of IL-1β and TNF-α23.

397

Our studies establish that omeprazole can interfere with pathways involved in mast cell

398

differentiation and activation as well as allergic inflammation. These observations provide insight

399

into the beneficial effects of this agent in some patients with EoE. In future studies, it will be of

400

great interest to identify the specific pharmacologic target(s) of omeprazole in mast cells, and to

401

consider the development of higher affinity and more specific small molecule inhibitors.

402 403 404 405 406

Acknowledgements The authors are grateful to Dr. Joshua Boyce and Dr. Chunli (Lily) Feng for providing

407

human cord blood derived- mast cells. We are also grateful to Dr. Takashi Fujimura, Ms.

408

Catherine Lavallee for advice and assistance with animal studies, and to Ms. Yvonne Nguyen

409

for technical assistance.

18

410

Figure Legends

411

Figure 1. Omeprazole blocks IgE-induced mast cell degranulation, cytokine secretion and

412

transcriptional changes. A) Representative histogram and bar plots for LAMP-1 expression by

413

BMMCs following sensitization with IgE anti-TNP, treatment with drug vehicle (DMSO) or

414

omeprazole, and challenge with TNP-OVA for 10 minutes. B) Histamine release into culture

415

supernatant from IgE anti-TNP sensitized BMMCs following activation with TNP-OVA for 5

416

minutes. C) LAMP-1 expression of BMMCs following activation with adenosine. D) Heatmap of

417

transcriptional changes in IgE sensitized BMMCs pre-treated with and without omeprazole

418

followed by antigen stimulation for two hours. Direct mRNA counts as measured by the

419

nCounter® Digital Analyzer System (NanoString) were normalized to internal positive, negative

420

and housekeeping gene controls and presented as standard deviation from the row mean. E-H)

421

Release of cytokines (IL-4, IL-6, IL-13, TNF-α) by IgE-sensitized BMMCs following omeprazole

422

treatment and antigen stimulation for six hours. Statistical analysis by one-way analysis of

423

variance. Data are shown for one experiment representative of two independent experiments for

424

LAMP1, histamine and cytokine secretion assay; transcriptional profiling by NanoString was

425

performed once. *P<0.05, **P<0.01, ***P<0.001, and ****P<0.0001

426

Figure 2. Omeprazole blocks IgE-induced human mast cell degranulation and cytokine

427

secretion. Bar plots showing LAMP-1 expression (A) and histamine release (B) by human mast

428

cells following sensitization with IgE, treatment with drug vehicle (DMSO) or omeprazole, and

429

challenge with anti-IgE for 10 and 5 minutes, respectively. Secretion of PGD2 (C), IL-5 (D) and

430

IL-13 (E) by IgE-sensitized human following omeprazole treatment and stimulation with anti-IgE

431

for six hours. Statistical analysis by one-way analysis of variance. *P<0.05, **P<0.01,

432

***P<0.001, and ****P<0.0001.

433

Figure 3. Effect of omeprazole on calcium flux and signaling pathways downstream of FcεRI. A)

434

Representative SYK and ERK phosphorylation blots and compiled ratios of phospho

19

435

protein/total protein intensities in IgE-sensitized BMMCs treated or untreated (DMSO) with

436

omeprazole for two hours. Statistics calculated following two-way ANOVA. B) Mobilization of

437

calcium from intracellular stores following antigen stimulation of IgE-sensitized BMMCs treated

438

with omeprazole for two hours or untreated (DMSO). P-value calculated by two-way ANOVA

439

between DMSO and omeprazole (50 µM). Data are shown for one experiment each

440

representative of at least 2 independent experiments.

441

Figure 4. Effect of omeprazole on passive systemic anaphylaxis. A) Change in core body

442

temperature of mice treated with omeprazole or drug vehicle, sensitized with SPE-7, and

443

challenged with DNP-BSA. Statistical analysis by two-way analysis of variance. B) Serum

444

concentration of MCPT-1 in animals from PSA experiment at endpoint. C) Plasma histamine

445

concentration in animals 5 minutes after allergen challenge. Statistical analysis by unpaired t-

446

test. Data are from one experiment representative of 2 independent experiments with n = 4-5 for

447

temperature drop and MCPT1 levels; histamine levels are from one experiment with n = 3-4

448

animals. **P<0.01 and ***P<0.001

449

Figure 5. pH of cellular compartments following omeprazole treatment of BMMCs. A)

450

Measurement of changes in pH by of intracellular vesicles by Lysosensor blue following

451

treatment with omeprazole and bafilomycin. B) Loss of quenching of FITC-dextran in BMMCs

452

treated with omeprazole or chloroquine. C) Measurement of changes in cytosolic pH by

453

pHrodoGreen following treatment with omeprazole, chloroquine or bafilomycin. Statistical

454

analysis by one-way analysis of variance. Data are shown from one representative of two

455

experiments. MFI, mean fluorescence intensity. *P<0.05, **P<0.01, ***P<0.001 , and

456

****P<0.0001

457

Figure 6. Omeprazole inhibits differentiation of myeloid precursor cells to mast cells. A)

458

Maturation of precursor cells to BMMCs in the presence or absence of omeprazole. Statistical

459

analysis by two-way analysis of variance. B) Percentage of small intestinal mast cells among

460

CD45+ cells in mice intraperitoneally injected with immune complexes of IL-4 and anti-IL-4 (IL4C) 20

461

to induce mastocytosis and intragastrically treated with omeprazole. For in vitro mast cell

462

differentiation, data is from one experiment involving three independent cultures; the in vivo

463

experiment was repeated twice, and data are shown for one experiment. *P<0.05

464

Figure 7. Treatment with omeprazole blocks inflammation in a mouse model of ovalbumin-

465

induced food allergy. A) Experimental design (S: sensitization, C: challenge, PPI: proton pump

466

inhibitor). B-E) Representative CAE-stained histological sections of small intestine from mice

467

from each experimental group. F) Summary of average counts of mast cells per high power field.

468

G) Serum concentrations of MCPT-1 at the experimental endpoint. H and I) Serum total and

469

OVA-specific IgE concentrations at experimental endpoint. J-M) Expression of cytokine

470

transcripts in the small intestines of mice from each experimental group at study endpoint. UT:

471

Untreated, OM: Omeprazole, OVA: Ovalbumin. Data are shown are from one experiment

472

representative of three independent experiments with n = 3-6 per group. Statistical analysis by

473

one-way analysis of variance. *P<0.05, **P<0.01

474

21

475

References

476

1.

477 478

treatment. Annu Rev Med 2009; 60:261-77. 2.

479 480

Sicherer SH, Sampson HA. Food allergy: Epidemiology, pathogenesis, diagnosis, and treatment. J Allergy Clin Immunol 2014; 133:291-307; quiz 8.

3.

481 482

Sicherer SH, Sampson HA. Food allergy: recent advances in pathophysiology and

Renz H, Allen KJ, Sicherer SH, Sampson HA, Lack G, Beyer K, et al. Food allergy. Nat Rev Dis Primers 2018; 4:17098.

4.

483

Oettgen HC, Burton OT. IgE receptor signaling in food allergy pathogenesis. Curr Opin Immunol 2015; 36:109-14.

484

5.

Furuta GT, Katzka DA. Eosinophilic Esophagitis. N Engl J Med 2015; 373:1640-8.

485

6.

O'Shea KM, Aceves SS, Dellon ES, Gupta SK, Spergel JM, Furuta GT, et al.

486 487

Pathophysiology of Eosinophilic Esophagitis. Gastroenterology 2018; 154:333-45. 7.

488 489

Spergel J, Aceves SS. Allergic components of eosinophilic esophagitis. J Allergy Clin Immunol 2018; 142:1-8.

8.

Pelz BJ, Wechsler JB, Amsden K, Johnson K, Singh AM, Wershil BK, et al. IgE-

490

associated food allergy alters the presentation of paediatric eosinophilic esophagitis. Clin

491

Exp Allergy 2016; 46:1431-40.

492

9.

Hsu Blatman KS, Gonsalves N, Hirano I, Bryce PJ. Expression of mast cell-associated

493

genes is upregulated in adult eosinophilic esophagitis and responds to steroid or dietary

494

therapy. J Allergy Clin Immunol 2011; 127:1307-8 e3.

495

10.

Niranjan R, Mavi P, Rayapudi M, Dynda S, Mishra A. Pathogenic role of mast cells in

496

experimental eosinophilic esophagitis. Am J Physiol Gastrointest Liver Physiol 2013;

497

304:G1087-94.

498

11.

Wen T, Dellon ES, Moawad FJ, Furuta GT, Aceves SS, Rothenberg ME. Transcriptome

499

analysis of proton pump inhibitor-responsive esophageal eosinophilia reveals proton

500

pump inhibitor-reversible allergic inflammation. J Allergy Clin Immunol 2015; 135:187-97.

22

501

12.

502 503

Abe Y, Sasaki Y, Yagi M, Yaoita T, Nishise S, Ueno Y. Diagnosis and treatment of eosinophilic esophagitis in clinical practice. Clin J Gastroenterol 2017; 10:87-102.

13.

Liacouras CA, Furuta GT, Hirano I, Atkins D, Attwood SE, Bonis PA, et al. Eosinophilic

504

esophagitis: updated consensus recommendations for children and adults. J Allergy Clin

505

Immunol 2011; 128:3-20 e6; quiz 1-2.

506

14.

Lucendo AJ, Arias A, Molina-Infante J. Efficacy of Proton Pump Inhibitor Drugs for

507

Inducing Clinical and Histologic Remission in Patients With Symptomatic Esophageal

508

Eosinophilia: A Systematic Review and Meta-Analysis. Clin Gastroenterol Hepatol 2016;

509

14:13-22 e1.

510

15.

Dohil R, Newbury RO, Aceves S. Transient PPI responsive esophageal eosinophilia may

511

be a clinical sub-phenotype of pediatric eosinophilic esophagitis. Dig Dis Sci 2012;

512

57:1413-9.

513

16.

514 515

J Neurogastroenterol Motil 2013; 19:25-35. 17.

516 517

Johnson RG, Carty SE, Fingerhood BJ, Scarpa A. The internal pH of mast cell granules. FEBS Lett 1980; 120:75-9.

18.

518 519

Shin JM, Kim N. Pharmacokinetics and pharmacodynamics of the proton pump inhibitors.

Pejler G, Hu Frisk JM, Sjostrom D, Paivandy A, Ohrvik H. Acidic pH is essential for maintaining mast cell secretory granule homeostasis. Cell Death Dis 2017; 8:e2785.

19.

Burton OT, Darling AR, Zhou JS, Noval-Rivas M, Jones TG, Gurish MF, et al. Direct

520

effects of IL-4 on mast cells drive their intestinal expansion and increase susceptibility to

521

anaphylaxis in a murine model of food allergy. Mucosal Immunol 2013; 6:740-50.

522

20.

Ochi H, Hirani WM, Yuan Q, Friend DS, Austen KF, Boyce JA. T helper cell type 2

523

cytokine-mediated comitogenic responses and CCR3 expression during differentiation of

524

human mast cells in vitro. J Exp Med 1999; 190:267-80.

525 526

21.

Cortes JR, Rivas MD, Molina-Infante J, Gonzalez-Nunez MA, Perez GM, Masa JF, et al. Omeprazole inhibits IL-4 and IL-13 signaling signal transducer and activator of

23

527

transcription 6 activation and reduces lung inflammation in murine asthma. J Allergy Clin

528

Immunol 2009; 124:607-10, 10 e1.

529

22.

Zhang X, Cheng E, Huo X, Yu C, Zhang Q, Pham TH, et al. Omeprazole blocks STAT6

530

binding to the eotaxin-3 promoter in eosinophilic esophagitis cells. PLoS One 2012;

531

7:e50037.

532

23.

Balza E, Piccioli P, Carta S, Lavieri R, Gattorno M, Semino C, et al. Proton pump

533

inhibitors protect mice from acute systemic inflammation and induce long-term cross-

534

tolerance. Cell Death Dis 2016; 7:e2304.

535

24.

Mathias CB, Hobson SA, Garcia-Lloret M, Lawson G, Poddighe D, Freyschmidt EJ, et al.

536

IgE-mediated systemic anaphylaxis and impaired tolerance to food antigens in mice with

537

enhanced IL-4 receptor signaling. J Allergy Clin Immunol 2011; 127:795-805 e1-6.

538

25.

Platzer B, Baker K, Vera MP, Singer K, Panduro M, Lexmond WS, et al. Dendritic cell-

539

bound IgE functions to restrain allergic inflammation at mucosal sites. Mucosal Immunol

540

2015; 8:516-32.

541

26.

542 543

Immunol Rev 2009; 228:149-69. 27.

544 545

Suzuki R, Scheffel J, Rivera J. New insights on the signaling and function of the highaffinity receptor for IgE. Curr Top Microbiol Immunol 2015; 388:63-90.

28.

546 547

Gilfillan AM, Rivera J. The tyrosine kinase network regulating mast cell activation.

Holowka D, Wilkes M, Stefan C, Baird B. Roles for Ca2+ mobilization and its regulation in mast cell functions: recent progress. Biochem Soc Trans 2016; 44:505-9.

29.

Ahrens R, Osterfeld H, Wu D, Chen CY, Arumugam M, Groschwitz K, et al. Intestinal

548

mast cell levels control severity of oral antigen-induced anaphylaxis in mice. Am J Pathol

549

2012; 180:1535-46.

550

30.

Lin RY, Schwartz LB, Curry A, Pesola GR, Knight RJ, Lee HS, et al. Histamine and

551

tryptase levels in patients with acute allergic reactions: An emergency department-based

552

study. J Allergy Clin Immunol 2000; 106:65-71.

24

553

31.

Brandt EB, Strait RT, Hershko D, Wang Q, Muntel EE, Scribner TA, et al. Mast cells are

554

required for experimental oral allergen-induced diarrhea. J Clin Invest 2003; 112:1666-

555

77.

556

32.

Madden KB, Whitman L, Sullivan C, Gause WC, Urban JF, Jr., Katona IM, et al. Role of

557

STAT6 and mast cells in IL-4- and IL-13-induced alterations in murine intestinal epithelial

558

cell function. J Immunol 2002; 169:4417-22.

559

33.

Galli SJ, Tsai M. IgE and mast cells in allergic disease. Nat Med 2012; 18:693-704.

560

34.

Mukai K, Tsai M, Saito H, Galli SJ. Mast cells as sources of cytokines, chemokines, and

561 562

growth factors. Immunol Rev 2018; 282:121-50. 35.

563 564

Bohm M, Richter JE. Treatment of eosinophilic esophagitis: overview, current limitations, and future direction. Am J Gastroenterol 2008; 103:2635-44; quiz 45.

36.

Moriyama Y, Patel V, Ueda I, Futai M. Evidence for a common binding site for

565

omeprazole and N-ethylmaleimide in subunit A of chromaffin granule vacuolar-type H(+)-

566

ATPase. Biochem Biophys Res Commun 1993; 196:699-706.

567

37.

Jordakieva G, Kundi M, Untersmayr E, Pali-Scholl I, Reichardt B, Jensen-Jarolim E.

568

Country-wide medical records infer increased allergy risk of gastric acid inhibition. Nat

569

Commun 2019; 10:3298.

570

38.

571 572

Pali-Scholl I, Jensen-Jarolim E. Anti-acid medication as a risk factor for food allergy. Allergy 2011; 66:469-77.

39.

Handa O, Yoshida N, Fujita N, Tanaka Y, Ueda M, Takagi T, et al. Molecular

573

mechanisms involved in anti-inflammatory effects of proton pump inhibitors. Inflamm Res

574

2006; 55:476-80.

575

40.

Yoshida N, Yoshikawa T, Tanaka Y, Fujita N, Kassai K, Naito Y, et al. A new mechanism

576

for anti-inflammatory actions of proton pump inhibitors--inhibitory effects on neutrophil-

577

endothelial cell interactions. Aliment Pharmacol Ther 2000; 14 Suppl 1:74-81.

578

25

Figure 1

1000

) µM

SO M

(5 0

ul

ep

U

ra

ns

zo

le

tim

D

at ed

) µM (5 0 le

zo ra

O m

O m ep

TNF-α (pg/ml)

200

100

) 0µ

M

M SO zo

le

(5

PTN o

ra

le zo

o

O VA

) 0µ

M

SO

TN

(5

N

ra ep

N

(5 e zo l

o N

****

300

0

PO VA

) 0µ

M

SO M

PTN

D

O VA

) (5 0 e

ra zo l ep O m

TNF-α

H

0

µM

SO M D

PO VA TN o N

M SO

P-

(5

TN

200

0

0

D

O VA

M ) 0µ

SO M

200

400

M

2

400

****

D

4

600

600

ra

IL-6 (pg/ml)

IL-4 (pg/ml)

6

***

800

IL-13

ep

***

8

G

IL-6

O m

F

IL-4

O m

E

o

+

N

+

+

IL-13 (pg/ml)

-

+

le

+

zo

+

ra

-

ep

-

0

O m

TNP-OVA

D

E Ig o N

Omeprazole (50!M)

10

0

0

IgE

20

D

20

2000

TNP-Ova TNP-OVA Omeprazole Omepazole

***

ep

40

30

O m

60

D

Adenosine

**

3000

****

Histamine (nM)

LAMP1+ (% of Live BMMCs)

80

C

Histamine

-

+

-

+

-

-

+

+

IgE lo

2 Nano

Relb Tnfa Ccl7 Il1b Ccl5 Tgfb1 Il13 Il4 Nrf1 Ccl2 Csf2 Mcpt8 Socs2 Il6 Raldh1a1 Cd103 Shfm1 Il25 Fcer1a Cpa3 Fcer1b Ahrr Cxcr5 Ptgs2 ltc4s Il15 Arnt Pdl1 Slc6a4 Ahr Mcpt2 Mcpt1 Tlr4 Nqo1 Ccl3 Clc4 Pla2g12a

3 IL1b

-Log10 (p-value)

B

LAMP1

LAMP1+ (% of Live BMMCs)

A

2

IL13

1

-4

-2

L

Most signifi

Gene Symbo

CPA3

TGFb1 Il1b

Most signif

Gene Symbo Tlr4 Nqo1 -3

-1.5

0

1.5

3

Il15

o an

IL-5

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

60

40

20

E

0

200

0

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

4000

2000

IL-13

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

150

100

50

0

an tiO Ig m E ep D ra O M z m ep ole SO O razo (50 m µM ep le ( ) 1 ra zo 0 0 µ le M ) (2 00 µM )

Histamine

o

6000

PGD2 (pg/mL)

B

N

an tiO Ig m E ep D ra O M z m ep ole SO O razo (50 m µM ep le ( 10 ) ra zo 0µ le M ) (2 00 µM )

o

20

tiO Ig m E ep D ra O M z m ep ole SO O razo (50 m µM ep le ( ) 1 ra zo 00µ le M ) (2 00 µM )

40

N

60

an

80

* **** ****

o

80

Histamine (nM)

LAMP-1

N

o

an tiO Ig m E ep D ra O M z m ep ole SO O razo (50 m µM ep le ( ) 1 ra zo 00µ le M ) (2 00 µM )

N

A

IL-13 (pg/mL)

D

tiO Ig m E ep D ra O M z m ep ole SO O razo (50 m µM ep le ( ) 1 ra zo 00µ le M ) (2 00 µM )

N

IL-5 (pg/mL)

LAMP1+ (% of Live CBMCs)

Figure 2 C PGD2

8000

0

* * **

6000

4000

2000 0

Figure 3

DMSO

A

Minutes

0

2.5

5

10

Omeprazole 20

30

0

2.5

5

10

20

B

30

Percent change [(F-F0)/F0] *100

Total ERK β-actin

DMSO Omeprazole (50uM)

40 20

Fold increase from time 0

p<0.01

60

20

40

DMSO Omeprazole (50uM)

p< 0.0001 30

0 20 10

100 DMSO Ionomycin

Time (mins)

30

20

10

5

5 2.

0

30

20

10

5

5 2.

0

200

300

Time (s)

0

0

0

Fold increase from time 0

40

ERK

SYK

Time (mins)

p<0.0001

P-SYK Total SYK β-actin P-ERK

80

60

Omeprazole (1 µM) Omeprazole (10 µM) Omperazole (50 µM)

Figure 4 B

1

400

MCPT-1

-2

p < 0.0001

-3 PBS Esomeprazole

-4 -5

0

5

10

15

20

25 30 35 Time (min)

40

45

50

55

60

300 200 100 0

8000

**

-1

MCPT-1 (ng/ml)

Temperature (°C)

0

C

Histamine (nM)

A

PBS

Esomeprazole

Histamine

***

6000 4000 2000 0

PBS

Esomeprazole

Ba

fil o

in

ne yc

ui

(2

15000

)

M

)

) M

M 0u 0n

(5

0u

SO

****

m

oq

(5

M

0

or

le

200

MFI PhrodoGreen (Live cKit+FcεRIα+)

400

D

)

uM

)

600

zo

(5 0

M

**

ra

ne

0u

****

hl

ui

(5

800

C

C

oq

le

****

ep

or

SO

FITC Dextran

O m

hl

M

B

C

zo

0

D

20000

MFI (FITC dextran)

40000

ra

)

)

M

nM

0u (2 0

(5

* **

ep

A1

le

M SO

Lysosensor Blue

O m

in

zo

D

60000

yc

m

ra

A

Ba fil o

ep

O m

MFI (Lysosensor Blue)

Figure 5 pHrodoGreen ****

****

10000

5000

0

Figure 6

le zo

Untreated Omeprazole (50uM)

ra

Days

0.0

ep

6 12 18 24 30 36

O m

0

IL 4C

0

0.2

+

20

0.4

C

40

0.6

IL 4

60

*

0.8

ve

80

B

ai

p<0.0001

N

100

% of Lin-cKit+FcεRIα+ of CD45+ cells

% of cKit+FcεRIα+ of live cells

A

Figure 7 A

S1

S2

S3

C1 C2 PPI

DAY 0

7

14

21 23 H

G

*

4000

Total IgE (ng/ml)

100

Total IgE

80 60 40 20 0

3000

2000 1000

1000

0

**

30 20 10 0 PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

K

IL-5 600

L

**

400

200

0 PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

IL-13 3000

**

2000

0

IL-4 40

OVA-IgE

**

3000

PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

J

I OVA-IgE (ng/ml)

Mast cell count

M

**

2000

1000

0

TNF-α 20

Relative expression

E OVA-OM-OVA

F

Relative expression

OVA-UT-OVA

C PBS-OM-OVA

Relative expression

D

PBS-UT-OVA

Relative expression

B

Average number of MC/HPF

HARVEST

**

15 10 5 0

PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

PBS OVA PBS OVA UT UT OM OM OVA OVA OVA OVA

Figure E1 A

Cells

Single cells

250K

Live cells

250K

cKIT+ FceRIa+ cells

250K

200K

150K

150K

150K

100K

100K

100K

50K

50K

50K

0

0

0

Comp-PE-A :: FceRIa

200K

FSC-H

200K

FSC-H

SSC-A

10

10

10

5

4

3

0

-10 0

50K

100K

150K

200K

250K

0

50K

100K

FSC-A

B

** ****

% Live cells of single cells

LAMP-1+ (% of Live BMMCs)

10

5

1

5

10

50

PO V D A M SO TN o

250K

-10

Omeprazole (µM)

0

10

3

10

4

10

5

3

-10

3

0

10

3

10

4

10

5

Comp-APC-A :: c-Kit

Comp-APC-Cy7-A :: Viability

C

0

N

200K

FSC-A

LAMP-1 15

150K

3

Viability 100 98

DMSO Omeprazole (50µM)

96 94 92 90 2

6

24

48

Hours of incubation

72

Figure E2 DMSO

Bafilomycin

Omeprazole

Figure E3

1

ONLINE REPOSITORY

2 3

Omeprazole inhibits IgE-mediated mast cell activation and allergic inflammation induced by

4

ingested allergen in mice

5 6

Cynthia Kanagaratham, PhD1,2, Yasmeen S. El Ansari, MMSc1,2, Benjamin F. Sallis, BSc1,2,

7

Brianna-Marie A. Hollister, BA1, Owen Lewis, BS1, Samantha C. Minnicozzi, MD1,2, Michiko K.

8

Oyoshi, PhD1,2, Rachel Rosen1,2, MD, MPH, Samuel Nurko1,2, MD, Edda Fiebiger, PhD1,2,* and

9

Hans C. Oettgen, MD, PhD1,2,*

10 11

1

12

2

Department of Pediatrics, Boston Children’s Hospital, Boston, MA, 02115, USA. Department of Pediatrics, Harvard Medical School, Boston, MA, 02115, USA.

13 14

*These authors contributed equally to this work.

15 16

Corresponding author:

17

[email protected]

18

1 Blackfan Circle

19

Boston, MA 02115

20

USA

21

Phone: 1 (617) 919-6842

22

Fax: 1 (617) 730-0528

23 24

25 26 27

Methods

28

incubated with anti-TNP IgE (500ng/ml). Unbound IgE was washed away and cells were treated

29

with omeprazole at 50 µM for two hours. Following treatment, cells were stimulated with TNP-

30

OVA (100ng/ml) for 10 minutes at 37ºC while also being stained for cell surface marker c-Kit

31

(APC), degranulation marker LAMP-1 (PE), and viability (Viability-APCCy7). The reaction was

32

stopped with cold FACS buffer. Cells were pelleted and resuspended in FACS buffer prior to

33

acquisition on an LSR Fortessa (BD Biosciences, Franklin Lakes, NJ). Mast cells were identified

34

as live cells that are cKit+. The expression of LAMP-1 on the cell surface was used as a

35

measure of mast cell degranulation. For omeprazole treatment, cells were incubated with the

36

drug (50µM) for two hours, then stimulated with TNP-OVA.

37

For adenosine-mediated degranulation, cells were not incubated with IgE prior to treatment.

38

Cells were stimulated with 100µM adenosine (Sigma) in the presence of staining antibodies.

39

Human mast cell LAMP-1 assay. 105 cord blood mast cells (CBMCs) in 100µl were incubated

40

with human myeloma IgE (100ng/ml; Chemicon International, Tecaluma, CA). Unbound IgE was

41

washed away and cells were treated with omeprazole at various doses (50, 100, and 200µM) for

42

two hours. Following treatment, cells were stimulated with rabbit anti-human IgE (100ng/ml) for

43

10 minutes in the presence of viability dye (Viability-APCCy7), anti-LAMP-1 (PE), and anti-cKit

44

(APC). Cells were pelleted and resuspended in FACS buffer prior to acquisition on an LSR

45

Fortessa (BD). Mast cells were identified as live cells that are cKit+. The expression of LAMP-1

46

on the cell surface was used as a measure of mast cell degranulation. For omeprazole

47

treatment, cells were incubated with the drug (50µM) for two hours, then stimulated with anti-

48

IgE.

49

Histamine ELISA. Histamine concentration in plasma or cell culture supernatants was

50

measured using the histamine ELISA kit from SPI bio (Cayman Chemical, Ann Arbor, Michigan)

Murine mast cell LAMP-1 assay. 105 bone marrow-derived mast cells (BMMCs) in 100µl were

51

following the manufacturer’s instructions. For measurement of histamine concentration in

52

plasma, blood was collected 5 minutes after antigen challenge from mice. For measurement of

53

histamine concentration in cell culture, supernatants were collected 5 minutes after stimulation.

54

Cytokine release from BMMCs. BMMCs were sensitized with 500ng/mL anti-TNP IgE and

55

treated with omeprazole (50µM) for two hours. Following treatment, cells were stimulated with

56

100ng/mL of TNP-OVA for six hours. Supernatants were collected and immediately assayed for

57

to measure cytokine concentration.

58

Human mast cell cytokine secretion. CBMCs were sensitized with human myeloma IgE

59

(100ng/mL). Unbound IgE was washed away and cells were treated with omeprazole at various

60

doses (50, 100, and 200µM) for two hours. Following treatment, cells were stimulated with rabbit

61

anti-human IgE (100ng/ml) for 6 hours then supernatants were collected. Human mast cell IL-5

62

and IL-13 secretion was measured using a Cytometric Bead Array from BD Biosciences (San

63

Jose, California), according to the manufacturer’s instructions.

64

Transcriptional profiling of BMMCs by NanoString. BMMCs (106 cells in 1mL) were

65

sensitized overnight with 500ng/ml anti-TNP IgE. Unbound IgE was removed and cells were

66

treated with 50µM Omeprazole for 2 hours. Cells were then stimulated with 100ng/ml TNP-OVA

67

for 2h. Stimulation was stopped by washing with ice cold PBS, prior to lysis and homogenization

68

in RTL buffer (Qiagen) with β-mercaptoethanol (Sigma). RNA was isolated using the Qiagen

69

RNeasy mini kit according to the manufacturer’s instructions. Purified RNA (100 ng) was

70

hybridized with customized nCounter gene expression code sets as previously established for

71

unbiased analysis of Th2 inflammation1. Direct mRNA counts were determined by the

72

nCounter® Digital Analyzer System (NanoString Technologies, Seattle, WA;

73

www.nanostring.com) according to the manufacturer's protocol. Direct mRNA counts were

74

normalized to internal positive and negative controls and 5 house-keeping genes according to

75

manufacturer’s recommendations. Data is presented as standard deviation from the row mean.

76

PGD2 assay. PGD2 concentration in cell culture supernatants was measured using the

77

commercially available ELISA kit from Cayman Chemical following the manufacturer’s

78

instructions (Cayman). Samples to be assayed were prepared in the same manner as for

79

human mast cell cytokine secretion. Phosphoprotein detection. For each time point, 106

80

BMMCs were sensitized overnight with anti-TNP IgE (500ng/ml). Following sensitization, cells

81

treated with control or omeprazole for two hours then stimulated with 100ng/ml of TNP-OVA.

82

Stimulation was stopped with cold FACS buffer and cells were pelleted and lysed with

83

radioimmunoprecipitation assay (RIPA) buffer (Sigma) containing a cocktail of protease and

84

phosphatase inhibitors (Roche). Lysed cells were spun to collect protein supernatant. 10µl of

85

lysate was resolved on a 12% MiniProtean TGX precast gel (Biorad) and electrophoretically

86

transferred to a PVDF membrane (Millipore). Membranes were blocked with 5% BSA in TBST

87

for at least one hour. Membranes were probed with primary antibody overnight, washed with

88

TBST then probed with appropriate peroxidase conjugated secondary antibodies for one hour.

89

Membranes were developed with Supersignal chemiluminescent substrate reagent and imaged

90

on an iBright Western Blot scanner.

91

Calcium assay. BMMCs sensitized with anti-TNP IgE were treated with omeprazole (50µM) or

92

control. Intracellular calcium was stained using the Fluo 4NW calcium assay kit following the

93

manufacturer’s instructions. Change in the fluorescence of the calcium indicator was measured

94

using a Tecan Spark plate reader. Baseline fluorescence was measured for 30 seconds and

95

cells were then stimulated with 100ng/mL of TNP-OVA and changes in fluorescence were

96

measured for five minutes.

97

pH Assays. PPI-treated BMMCs (106 cells/mL) were incubated with FITC-Dextran overnight

98

(100µg/mL). Cells were washed and stained for cell surface markers c-Kit and FcεRIα, and for

99

viability. Cells were acquired by flow cytometry, and the intensity of the FITC signal of live c-Kit

100

and FcεRIα double-positive cells was assessed. Lysosensor Blue DND 167 (ThermoFisher) was

101

used to measure the changes in pH of acidic organelles following PPI treatment. Briefly, PPI-

102

treated BMMCs were washed and incubated with 1µM Lysosensor Blue for one hour. Cells were

103

washed and resuspended in media and the fluorescence intensity of the Lysosensor Blue was

104

measured using a Tecan Spark plate reader. To measure changes in intracellular pH following

105

treatment of BMMCs with PPI, pHRodoGreen AM intracellular pH indicator was used following

106

the manufacturer’s instructions. Briefly, PPI-treated cells were washed with live cell imaging

107

solution (ThermoFisher) then incubated with a mixture of pHRodoGreen and PowerLoad for 30

108

minutes. Cells were washed and resuspended in FACS buffer then stained for viability and cell

109

surface markers (c-Kit and FcεRIα). BMMCs were acquired on an LSR Fortessa and the

110

intensity of the intracellular pH indicator was measured in the population of live c-Kit and FcεRIα

111

double-positive cells.

112

Enzyme linked immunosorbent assay (ELISA). MCPT-1, OVA-IgE, and total IgE were

113

measured in serum isolated from mouse blood collected by intracardiac puncture. MCPT-1 was

114

measured using the MCPT-1 ELISA (ThermoFisher) following the manufacturer’s instructions.

115

Total IgE was measured using a direct sandwich ELISA using mouse IgE capture antibody

116

(Southern Biotech, #11101-01) and HRP-conjugated anti-mouse IgE (Southern Biotech, #1110-

117

05). OVA-IgE was measured using a direct sandwich ELISA using mouse IgE capture antibody

118

(Southern Biotech, #11101-01), and detected using biotinylated-OVA (US Biologicals, #O8075-

119

01) and Streptavidin-HRP. All ELISAs were developed with TMB and reactions were stopped

120

with 2N H2SO4. Cytokines in cell culture supernatants were measured using IL-6, IL-13, and

121

TNF-a Ready SET-Go! kits (ThermoFisher) following the manufacturer’s instructions. IL-4 in cell

122

culture supernatants was measured using the ELISA MAX Standard Set Mouse IL-4 kit

123

following the manufacturer’s instructions (Biolegend).

124

RT-qPCR. Jejunum sections were collected from mice and stored in RNAlater (ThermoFisher)

125

solution at 4oC overnight before freezing tissue at -80oC. RNA was extracted from 20 to 30mg of

126

tissue using the Qiagen RNeasy mini kit according to the manufacturer’s instructions. RNA

127

concentration was measured using a Nanodrop 2000. One µg of RNA was reverse transcribed

128

into cDNA using iScript cDNA Synthesis Kit (Biorad) following the manufacturer’s instructions.

129

Gene expression was measured using a QuantStudio 3 real time qPCR instrument

130

(ThermoFisher), Taqman Fast Advanced Mastermix (Applied Biosystems) and predesigned

131

probes (Life Technologies, Applied Biosystems). Expression of all genes was normalized to

132

Gapdh and calculated using the Cq values and the 2-ddCq method.

133

Intestinal cell isolation. Jejunum segments were isolated and flushed with PBS containing

134

10% FBS. The intestine was cut longitudinally and chopped into 0.5cm segments then

135

incubated in buffer containing 10mM EDTA and 1mM DTE for 20 minutes to remove the

136

epithelial layer. Tissue was finely chopped using a razor blade then digested with 971U/ml

137

collagenase VIII and 5µg/mL DNAse I (Applichem) in complete RPMI for 30 minutes. Cell

138

fraction was resuspended in 40% Percoll and overlaid above a 70% Percoll fraction. Cells were

139

centrifuged at 600xg for 20 minutes and leukocytes at the interface were collected.

140

Histology. Jejunums were collected from mice and segments of approximately 1cm were fixed

141

in 4% paraformaldehyde for a minimum of 48 hours. Dehydration, paraffin embedding,

142

sectioning and staining were done by the Harvard Medical Area Rodent Histopathology Core

143

(Harvard, Boston, MA). Images were captured on a Nikon E800 microscope (Nikon, Melville,

144

NY).

145

May Grünwald Giemsa Staining. BMMCs were treated with bafilomycin or omeprazole or drug

146

vehicle. 300,000 cells were adhered onto poly-lysine coated slides by cytospin (Shandon

147

Cytospin 3). Slides were stained with May Grünwald solution (Sigma) for 5 minutes, washed 3

148

times in deionized water, stained with Giemsa solution (Sigma) for 15 minutes then washed 6

149

times in deionized water. Slides were air dried then fixed with Permount (Sigma). Images were

150

captured on a Nikon E800 microscope (Nikon, Melville, NY).

151

Human basophil activation test. Human basophil activation tests were performed as

152

previously described 2 using the FLOW CAST Basophil activation test kit (Bühlmann

153

Laboratories, Schönenbuch, Switzerland). Briefly, 50µL aliquots of whole blood from healthy

154

donors were incubated with omeprazole (200µM) for two hours in a 37oC water bath. Following

155

treatment, samples were stimulated with 200ng/mL anti-IgE (Southern biotech) and

156

simultaneously stained for cell viability (Viability-APCCy7), anti-CCR3 (PE), anti-CD63 (FITC)

157

for 15 minutes at 37 oC. Red blood cells were lysed, and cells were washed and acquired on an

158

LSR Fortessa (BD). Basophils were identified as SSClow and CCR3+. The expression of CD63

159

on the cell surface was used as a measure of basophil activation.

160

161

Figure legends

162

Figure E1. Gating strategy for identification of bone marrow derived mast cells by flow

163

cytometry, dose responsive effect of omeprazole on mast cell degranulation, and viability of

164

cells following incubation with omeprazole. A) Cell population, and subsequently single cells,

165

are first identified based on forward and side scatter. Live cells are then selected from the single

166

cell gate based on negative staining for the viability dye. Live cells expressing cKIT and FcεRIa

167

are defined as bone marrow mast cells (BMMCs). B) LAMP-1 expression of IgE anti-TNP bound

168

BMMCs following incubation with various doses of omeprazole and degranulation with TNP-

169

OVA. C) Viability of BMMCs following incubation with omeprazole. Viability staining is applied to

170

all BMMC culture experiments analyzed by flow cytometry. For panel B, data are shown from

171

one experiment representative of two experiments. Statistical analysis by one-way analysis of

172

variance. **P<0.01, ****P<0.0001

173

Figure E2. Alterations in granule morphology following incubations with bafilomycin and

174

omeprazole. May Grünwald/Giemsa staining of cytospin preparations of BMMCs following

175

incubation with DMSO, bafilomycin (20 nM) or omeprazole (50 µM) for 24 hours.

176

Figure E3. Omeprazole blocks activation of basophils. Activation of basophils in whole blood

177

using anti-IgE following treatment with 200 µM omeprazole or vehicle for 2 hours. Basophil

178

activation was quantified by flow cytometry. Basophils were identified as SSClow, negative

179

staining for viability dye, and CCR3+. The expression of CD63 on the cell surface was used as a

180

measure of basophil activation. Data points represent four different individuals assayed in

181

duplicates for each condition. Statistical analysis between treatment groups was done using a

182

paired t-test. *P<0.05

183

184 185

References 1. Platzer B, Baker K, Vera MP, Singer K, Panduro M, Lexmond WS, et al. Dendritic cell-

186

bound IgE functions to restrain allergic inflammation at mucosal sites. Mucosal Immunol

187

2015; 8:516-32.

188

2.

Burton OT, Logsdon SL, Zhou JS, Medina-Tamayo J, Abdel-Gadir A, Noval Rivas M, et

189

al. Oral immunotherapy induces IgG antibodies that act through FcgammaRIIb to

190

suppress IgE-mediated hypersensitivity. J Allergy Clin Immunol 2014; 134:1310-7 e6.

191