MOLECULAR AND CELLULAR MECHANISMS OF DISEASE

TRPM2 contributes to antigen-stimulated Ca2+ influxin mucosal mast cells

Satoshi Oda & Kunitoshi Uchida & Xiaoyu Wang &

Jaemin Lee & Yutaka Shimada & Makoto Tominaga &

Makoto Kadowaki

Received: 22 September 2012 /Revised: 10 January 2013 /Accepted: 10 January 2013 /Published online: 31 January 2013# Springer-Verlag Berlin Heidelberg 2013

Abstract Food allergy (FA) is a common allergic diseasewithout any currently available effective drug therapies.Mucosal mast cells (MMCs) play a particularly importantrole in FA, and the increase in their cytosolic Ca2+ concen-tration ([Ca2+]cyt) is considered to be a principal componentof the degranulation process. However, the mechanismsgoverning Ca2+ influx remain poorly understood inMMCs. Recent reports have highlighted the functions ofthe transient receptor potential melastatin 2 (TRPM2) chan-nel in immunocytes, including its role in monocyte chemo-kine production and macrophage phagocytic activity.Although TRPM2 gene expression has been demonstratedin mast cells, the significance of such expression remainsvirtually unknown. In this study, we found that antigen-stimulated degranulation was significantly reduced inmucosal-type bone marrow-derived mast cells (mBMMCs)prepared from TRPM2-knockout (TRPM2-KO) mice(TRPM2-KO mBMMCs) and was suppressed followingthe administration of three TRPM2 inhibitors with differentchemical structures, including econazole, flufenamic acid(FFA), and 2-aminoethoxydiphenyl borate. Furthermore,the antigen-stimulated increase in [Ca2+]cyt was significantly

decreased in TRPM2-KO mBMMCs and was also sup-pressed by the TRPM2 inhibitors econazole and FFA. Inaddition, thapsigargin-induced increase in [Ca2+]cyt was sig-nificantly decreased in TRPM2-KO mBMMCs. Theseresults suggest that TRPM2 may participate in antigen-induced extracellular Ca2+ influx and subsequent degranu-lation. In addition, TRPM2 inhibitors were shown to im-prove food allergic reactions in a mouse model. Together,these results suggest that TRPM2 inhibitors suppress MMCdegranulation via regulation of the increase in [Ca2+]cyt.Thus, TRPM2 may play a key role in degranulation bymodulating intracellular Ca2+ in MMCs.

Keywords Mucosal mast cell . mBMMC . TRPM2 . Foodallergy . Degranulation

Introduction

Food allergy (FA) is a common and serious allergic disease,although no effective drug therapies are yet available. Theonly way to avoid allergic reactions is to avoid the foods thattrigger these responses [24]. Mucosal mast cells (MMCs)play crucial roles in FA pathogenesis via the release ofinflammatory mediators [34].

Mast cells are usually classified as MMCs or connectivetissue mast cells (CTMCs). The differences between these celltypes have been thoroughly described in the literature [19].CTMCs reside in connective tissues such as the skin, whereasMMCs depend on Th2 cells for activation and migrate fol-lowing antigen exposure to intestinal or bronchial mucosaltissues, where they mature. CTMCs contain high concentra-tions of histamine in their granules and mainly contribute toallergic diseases such as urticaria, whereasMMCs contain lowconcentrations of histamine. Consequently, antihistamine

S. Oda :X. Wang : J. Lee :M. Kadowaki (*)Division of Gastrointestinal Pathophysiology, Institute of NaturalMedicine, University of Toyama, 2630 Sugitani,Toyama 930-0194, Japane-mail: makotok@inm.u-toyama.ac.jp

S. Oda :Y. ShimadaDepartment of Japanese Oriental Medicine, Faculty of Medicine,University of Toyama, Toyama, Japan

K. Uchida :M. TominagaDivision of Cell Signaling, Okazaki Institute for IntegrativeBioscience (National Institute for Physiological Sciences),National Institutes of Natural Sciences, Okazaki, Japan

Pflugers Arch - Eur J Physiol (2013) 465:1023–1030DOI 10.1007/s00424-013-1219-y

agents have no effects on FA. Furthermore, CTMC stabilizers(tranilast, ketotifen, cromolyn, etc.) are frequently used for thetreatment of various allergic disorders besides FA, but theseagents have no effect on MMCs [12]. In both types of mastcells, the increase in cytosolic Ca2+ concentration ([Ca2+]cyt) isa principal pathway in antigen-stimulated degranulation [4,23]. In detail, antigen-induced FcεRI aggregation leads to Ca2+

release from endoplasmic reticulum (ER) stores via inositol1,4,5-trisphosphate (IP3) production, and the depletion of theseCa2+ stores causes ERmembrane-resident STIM1 aggregation.Subsequently, this STIM1 aggregation mainly activates theCRAC channel Orai1. Extracellular Ca2+ influx then leads toan increase in [Ca2+]cyt and degranulation. However, the mech-anisms governing Ca2+ influx remain poorly understood, par-ticularly in MMCs. Indeed, other Ca2+ influx channels havebeen reported in addition to Orai1 [20].

Virtually all transient receptor potential (TRP) channelsare Ca2+-permeable nonselective cation channels. TRP mel-astatin 2 (TRPM2) channels are expressed predominantly inthe brain but are also detected in the bone marrow, spleen,heart, liver, lung, and pancreatic β-cells [11, 17, 31, 32].These channels are thermosensitive and can be activated bynicotinamide adenine dinucleotide (NAD), adenosine 5-diphosphoribose (ADPR), hydrogen peroxide (H2O2), andintracellular Ca2+ [21, 25]. The functions of TRPM2 inimmunocytes have recently been described in detail [3, 6,16, 26]. TRPM2-mediated Ca2+ influx induces chemokineproduction in monocytes, which aggravates inflammatoryneutrophil infiltration [37], and TRPM2 is also involved inmacrophage phagocytic activity [15, 28]. Although TRPM2gene expression has been documented in mast cells, theprecise role of such expression remains virtually unknown[1]. As MMCs accumulate in the intestinal mucosal tissuesof animals with experimental FA [2], activated MMCs arethought to contribute to allergic inflammation [19, 35]. Thisstudy sought to elucidate the physiological and pathophys-iological functions of TRPM2 in the activation of MMCsand their pathophysiological role in a murine model of FA.

Materials and methods

Animals

Male C57BL/6 mice and male BALB/c mice (age 5–6 weeks) were obtained from Japan SLC, Inc. (Shizuoka,Japan). TRPM2-deficient (TRPM2-KO) mice, which weregenerously provided by Yasuo Mori (Kyoto University,Japan), were backcrossed to the C57BL/6 strain over eightgenerations. Mice were maintained under a 12-h light/darkcycle at 23±1 °C and were provided ad libitum access towater and food. All experiments were approved by theCommittee of Animal Experiments at the University of

Toyama and the National Institutes for PhysiologicalScience and were in accordance with NIH guidelines forthe care and use of laboratory animals (NIH publication no.85-23; revised 1985).

Cell culture and handling

Mucosal-type bone marrow-derived mast cells (mBMMCs)were prepared from male C57BL/6 (WT and TRPM2-KO)mice or BALB/c mice (age 5–6 weeks) as previously de-scribed [13]. In brief, bone marrow cells were collected byflushing mouse femurs, and the cells were cultured for 4–6 weeks with 20 ng/mL recombinant murine interleukin-3(IL-3; Miltenyi, Germany), 40 ng/mL recombinant murineSCF (Miltenyi), 5 ng/mL recombinant murine IL-9 (R&DSystems, Minneapolis, MN, USA), and 1 ng/mL TGF-β1(Sigma, St. Louis, MO, USA). Mast cell purity was validat-ed by assessing the expression of FcεRI and c-kit, and theseexpression levels were equivalent between WT andTRPM2-KO mBMMCs (data not shown). BMMCs culturedonly with SCF and IL-3 (S3-BMMCs) were regarded asanalogs of CTMCs [29].

Degranulation assay

Chemical mediator release was assessed by measuring theenzymatic activity of β-hexosaminidase as previously de-scribed [12]. In brief, mBMMCs were sensitized with1.5 μg/mL mouse monoclonal anti-dinitrophenyl (DNP)IgE (Yamasa, Japan) for 6 h at 37 °C. The cells wereincubated for 15 min at 37 °C in Tyrode's buffer containing40 ng/mL SCF (Miltenyi) and various concentrations ofDNP-bovine serum albumin (DNP-BSA) with or withoutTRPM2 inhibitors. Released β-hexosaminidase activity wasmeasured enzymatically, and the extent of degranulationwas expressed as a proportion of the enzymatic activitiesin the supernatants and cell lysates.

Reverse transcriptase-polymerase chain reaction

Total RNA was purified from mBMMCs using the RNeasysystem (QIAGEN, Germany) and reverse-transcribed usingthe Superscript III first-strand synthesis system for reversetranscriptase-polymerase chain reaction (RT-PCR)(Invitrogen, Carlsbad, CA, USA). PCR was performed usingrTaq DNA polymerase (TaKaRa, Japan) in a DICE PCRThermal Cycler (TaKaRa). The PCR cycling conditions foreach of the primer sets were set as previously described [9, 36].

Electrophysiology

Cells were plated on Cell-Tak-coated (CollaborativeBiomedical Products, Bedford, MA, USA) glass coverslips,

1024 Pflugers Arch - Eur J Physiol (2013) 465:1023–1030

and whole-cell patch-clamp recordings were performed 1 hafter plating. The standard bath solution contained 140 mMNaCl, 5 mM KCl, 2 mM MgCl2, 2 mM CaCl2, 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), and10 mM glucose (pH 7.4, adjusted with NaOH). CaBuf (http://www.kuleuven.be/fysio/trp/cabuf) was used to calculate thefree Ca2+ concentration in the solution to pipette. The pipettedsolution with 100 nM free Ca2+ contained 120 mM aspartate,10 mM KCl, 3.032 mM CaCl2, 1 mM MgCl2, 5 mM EGTA,17 mM mannitol, and 10 mM HEPES (pH 7.4, adjusted withKOH). Prior to the experiment, 300 μM ADP-ribose (ADPR;Sigma) was dissolved in the pipetted solution. Data fromwhole-cell voltage-clamp recordings were acquired at 10 kHzand filtered at 5 kHz for analysis (Axon 200B amplifier withpCLAMP software, Axon Instruments, Foster City, CA,USA). Membrane potential was clamped at −60 mV. Voltageramps of 50 ms spanning the voltage range of −100 to +40 mVwere delivered at a rate of 0.5 Hz to generate a current–voltage(I–V) curve. Heat stimulation was induced by increasing thebath temperature using a preheated solution warmed in aninline heater or heating stage (1 °C/s, with a maximum of47 °C). Temperature was monitored using a thermocouple(TA-29; Warner Instruments, Hamden, CT, USA) placed with-in 100 μm of the patch-clamped cell.

Fluorescence measurements

To determine the fluorescence intensity, cells plated on Cell-Tak-coated glass coverslips were loaded with 5 μM Fura-2-AM (Invitrogen) for 1 h in media at 37 °C. The standard bathsolution was the same as that used for the electrophysiologyexperiments. The Ca2+-free bath solution was prepared byremoving 2 mM CaCl2 and adding 5 mM EGTA to thestandard bath solution. Cells loaded with Fura-2 were excitedat 340 and 380 nm wavelengths, and the emitted light wasmonitored at 510 nm with a CCD camera (CoolSnap ES;Roper Scientific/Photometrics, Tucson, AZ, USA). Data wereacquired and analyzed using the IPlab software (Scanalytics,Fairfax, VA, USA) and ImageJ (http://rsbweb.nih.gov/ij/).Ionomycin (5 μM) was added at the end of each experimentto confirm cell viability. Cells were selected by reactivity toionomycin (net ratio increase>2.0). All experiments wereperformed at room temperature (25 °C). All data are presentedas the 340/380-nm fluorescence ratio and the incremental areaunder the curve (ΔAUC) over the following durations: Fig. 4b,30–150 s (DNP treatment), 330–360 s (ionomycin treatment);Fig. 4d, 55–100 s (Ca2+(−)), 160–270 s (Ca2+(+)); Fig. 4f,270–360 s (Ca2+(+)); and Fig. 5b, 120–210 s (DNP treatment).

Induction of food allergy in mice

The mouse model of food allergy was prepared as previous-ly described [38]. Male BALB/c mice (5 weeks old) were

sensitized twice at 2-week intervals via intraperitoneal injec-tions of 50 μg of ovalbumin (OVA; fraction V; Sigma)adsorbed on 1.3 mg/100 μL of aluminum hydroxide geladjuvant (Sigma). Two weeks after systemic priming, micewere repeatedly given 50 mg/0.3 mL of OVA orally everyother day. Food allergic diarrhea was assessed using theoriginal stool visual scale [38] (Table 1) at 1 h after oralOVA challenge. Econazole (30 μg/g bw; Wako) and flufe-namic acid (100 μg/g bw; Wako) were administered orally1 h before OVA challenges.

Statistical analysis

Parametric data are expressed as the mean±SE. The statis-tical analyses were performed using Student’s t test orANOVA followed by the Tukey–Kramer multiple compar-isons test. Nonparametric data were compared using theSteel–Dwass test. Probability values (P) of <0.05 wereconsidered statistically significant.

Results

TRPM2 is expressed in mBMMCs

We examined the expression of thermosensitive TRP genesin mBMMCs and S3-BMMCs obtained from C57BL/6mice. Among these genes, TRPV2, TRPA1, TRPM2, andTRPM4 were expressed in mBMMCs (Fig. 1a) .Furthermore, the TRPM2 gene was also detected inmBMMCs from BALB/c mice (Fig. 1b). Then, the func-tional expression of TRPM2 was examined using whole-cellpatch clamp recordings. As shown in Fig. 2a, an inwardcurrent was evoked by heat stimulation in WT mBMMCs.The linear I–V relationship indicated the activation ofTRPM2 [21] (Fig. 2b(p, q)). On the other hand, this currentwas not observed in TRPM2-KO mBMMCs (Fig. 2a, b(r,s)). Heat-evoked current (>100 pA) was observed in 90.9 %(n=20) of WT mBMMCs (n=22), while this current was

Table 1 Stool visual scale

Score Stool appearance

0 Normal Normal

1 Soft stool Squash easily

2 Loose stool Squash easily and sticky

3 Middle diarrhea Retain the shape but too soft to be picked up

4 Severe diarrhea Out of shape

5 Fluid diarrhea Completely watery stool

This scoring system was developed by our laboratory to evaluatediarrhea severity

Pflugers Arch - Eur J Physiol (2013) 465:1023–1030 1025

observed in only 6.7 % (n=1) of TRPM2-KO mBMMCs(n=15). In addition, the amplitude of the current was re-duced by three TRPM2 inhibitors with different chemicalstructures econazole (10 μM) [8], FFA (200 μM) [7], and 2-aminoethoxydiphenyl borate (2APB) (30 μM) [30](Fig. 2c).

TRPM2 is involved in mast cell degranulation

We next examined the involvement of TRPM2 in the de-granulation of MMCs. The total β-hexosaminidase contentand ionomycin-induced degranulation were virtually equiv-alent between TRPM2-KO and WT mBMMCs (Fig. 3a–c).Thus, the ability to form granules and degranulate wasconsidered to be intact in the knockout cells. Although thebasal level of release did not differ between TRPM2-KOand WT mBMMCs, antigen-stimulated degranulation (1 to10 ng/mL DNP-BSA) was significantly reduced in TRPM2-KO mBMMCs (Fig. 3b). In addition, three TRPM2 inhib-itors with different chemical structures suppressed degranu-lation in mBMMCs from C57BL/6 and BALB/c mice,including econazole (10 μM) [8], FFA (200 μM) [7], and2APB (30 μM) [30] (Fig. 3d, e). To elucidate the mechanisms

underlying TRPM2-mediated degranulation, we examined thedegranulation induced by treatment with thapsigargin, whichcauses depletion of Ca2+ stores. As shown in Fig. 3c, thisdegranulation was significantly decreased in TRPM2-KOmBMMCs.

TRPM2 is involved in calcium mobilization

Next, we investigated changes in [Ca2+]cyt during antigenstimulation using Ca2+-imaging. The antigen-induced in-crease in [Ca2+]cyt was significantly lower in TRPM2-KOmBMMCs than in WT mBMMCs (Fig. 4a, b). At a dose of10 ng/mL DNP, the ΔAUC was 51.0±3.0 ratios in WTmBMMCs and 36.4±2.1 ratios in TRPM2-KO mBMMCs(P<0.01). When mBMMCs were stimulated under extracel-lular Ca2+-free conditions, a small increase in [Ca2+]cyt wasobserved in both TRPM2-KO and WT mBMMCs. On theother hand, the increase in [Ca2+]cyt after switching to theCa2+-containing buffer was impaired in TRPM2-KO

S3-BMMC mBMMCCont

RT (+) (-)(+)(-)

TRPV1

TRPV2

TRPV3

TRPV4

TRPA1

TRPM2

TRPM4

TRPM5

TRPM8

GAPDH

C57BL/6

ContRT (-)(+)

TRPM2

mBMMCBALB/c

GAPDH

a

b

Fig. 1 Expression of thermosensitive TRP genes in mBMMCs. a RT-PCR of RNA extracted from S3-BMMCs and mBMMCs from C57BL/6 mice. Reverse transcript (RT) negatives are displayed in each row ofthe right. Positive controls, i.e., pre-purified PCR products, are indi-cated by Cont. b TRPM2 expression in mBMMCs from BALB/c mice

Cur

rent

0.2

nA

10 sec

304050

20Tem

p(˚C

)

p qKOWT

r s

pq

mV

pA-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

-100 -80 -60 -40 -20 0 20 40mV

pA-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

-100 -80 -60 -40 -20 0 20 40

rs

a

bWT KO

2APB 30µM

Cur

rent

0.0

5 nA

10 sec

FFA 200µM

Cur

rent

0.0

5 nA

10 sec

eco 10µM

Cur

rent

0.0

5 nA

10 sec

30

40

50

20Tem

p(˚C

)

c

Cur

rent

0.2

nA

10 sec

Fig. 2 Heat-evoked currents in wild-type (WT) and TRPM2-KO (KO)mBMMCs. a The representative trace of whole-cell patch-clamp at aholding potential of −60 mV. Voltage ramps from −100 to +40 mVwere applied. Temperature curves of heat stimulation are displayedbelow. b Current–voltage (I–V) curves for each ramp are indicated as pto s in a. c Inhibitory effects of the TRPM2 inhibitors econazole (eco;10 μM), FFA (200 μM), and 2APB (30 μM) on the heart-activatedcurrents in WT mBMMCs

1026 Pflugers Arch - Eur J Physiol (2013) 465:1023–1030

mBMMCs (Fig. 4c, d), as the ΔAUC for these cells was17.2±2.8 ratios compared with 26.8±2.4 ratios in WTmBMMCs (P<0.01). Moreover, the increase in [Ca2+]cytinduced by thapsigargin was significantly smaller inTRPM2-KO mBMMCs than in WT mBMMCs (Fig. 4e,f); the ΔAUC was 74.8±4.1 ratios in WT mBMMCs and51.0±4.1 ratios in TRPM2-KO mBMMCs (P<0.01). Inaddition, the expression of Orai1 mRNA remained un-changed in TRPM2-KO mice (Fig. 4g).

Next, we examined whether TRPM2 inhibitors wouldaffect the antigen-stimulated increase in [Ca2+]cyt. As shownin Fig. 5a, b, econazole (10 μM) and FFA (200 μM) reducedthe increase in [Ca2+]cyt in mBMMCs from BALB/c mice(P<0.01).

TRPM2 inhibition improves FA symptoms

To elucidate the pathophysiological role of TRPM2 in a FAmodel, we examined the effects of TRPM2 inhibitors on thedevelopment of FA. OVA-induced allergic diarrhea was ob-served in all mice on the seventh day after the initiation ofOVA challenge. However, administration of the TRPM2inhibitors econazole (30 mg/g bw) and FFA (100 mg/g bw)significantly reduced the OVA-induced diarrhea score (Fig. 6).

Discussion

In this study, we found that TRPM2 was expressed inmBMMCs and was involved in antigen-stimulated increasesin [Ca2+]cyt and degranulation. Furthermore, blockade ofTRPM2 alleviated food allergy in a mouse model.

As described in the introduction, FA is a common andserious disease without any effective drug therapy. Mucosalmast cells (MMCs), but not connective tissue mast cells(CTMCs), play pathogenic roles in FA due to their release ofinflammatory mediators. Because MMCs contain low con-centrations of histamine, antihistamine agents offer no benefitagainst FA, in contrast to their effectiveness at treating allergicdiseases, such as rhinitis, caused by CTMCs. Moreover, con-nective tissue mast cell stabilizers (tranilast, ketotifen, cromo-lyn, etc.) also have no effect on mucosal mast cells. Therefore,it is highly desirable to elucidate the precise mechanisms ofMMC activation and develop useful therapeutic drugs for FA[19]. Although the mechanisms underlying antigen-induceddegranulation have not yet been fully understood, Ca2+ influxis indispensable for mast cell activation and degranulation.Orai1 is a major Ca2+ channel responsible for the influx ofextracellular Ca2+ into mast cells, but other Ca2+ influx chan-nels have also been reported [20].

1

2

3

4

0Tot

al β

-hex

osam

inid

ase

cont

ent

AB

S (

405n

m)

WT KO

a

0

10

20

30

40

50

60

β-he

x re

leas

e(%

of t

otal

cel

l con

tent

)

WT KO

DNP

0ng/mL

WT KO

DNP

1ng/mL

WT KO

DNP

3ng/mL

WT KO

DNP

10ng/mL

WT KO

Iono

5µM

b

*

*

0

10

20

30

40

50

60

β-he

x re

leas

e(%

of t

otal

cel

l con

tent

)

DNPecoFFA

2APB

----

+---

++--

+-+-

+--+

Iono

5µM

*

* *

C57BL/6

0

10

20

30

40

50

60

β-he

x re

leas

e(%

of t

otal

cel

l con

tent

)

DNPecoFFA

2APB

----

+---

++--

+-+-

+--+

Iono

5µM

* * *

BALB/c

d e

0

10

20

30

40

50

60

WT KO

Control

*

β-he

x re

leas

e(%

of t

otal

cel

l con

tent

)

WT KO

Thapsi

1µM

WT KO

Iono

5µM

c

Fig. 3 Effect of TRPM2 on mBMMC degranulation. a Total β-hex-osaminidase content in wild-type (WT) and TRPM2-KO (KO)mBMMCs. b Extent of degranulation induced by various concentra-tions of DNP-BSA and 5 μM ionomycin (Iono). *P<0.01 vs. WT. cExtent of degranulation induced by depletion of intracellular Ca2+

stores with 1 μM thapsigargin. *P<0.01 vs. WT. d, e Inhibitory effectsof the TRPM2 inhibitors econazole (eco; 10 μM), FFA (200 μM), and2APB (30 μM) on degranulation induced by DNP-BSA (10 ng/mL) inmBMMCs from C57BL/6 (d) and BALB/c mice (e). *P<0.01 vs. DNP(+) eco(−) FFA(−) 2APB(−). The data represent the means±SE

Pflugers Arch - Eur J Physiol (2013) 465:1023–1030 1027

In this study, we found that the thermosensitive TRP genesTRPV2, TRPA1, TRPM4, and TRPM2 were expressed in mastcells (Figs. 1 and 2), which is consistent with previous reportsdemonstrating the expression of TRPV2, TRPA1, and TRPM4in mast cells [14, 33, 39]. TRPV2 is involved in mast celldegranulation in response to physical stimuli such as hydro-static pressure, high noxious temperature, and laser light [39].

TRPA1 is localized in intracellular vesicular structures, but itsprecise role remains unclear [22]. These studies were con-ducted using in vitro experiments, while the function ofTRPM4 in mast cells has been assessed in both in vitro andin vivo experiments. TRPM4 regulates the membrane poten-tial and suppresses Ca2+ influx and degranulation in mast cells[33]. In contrast to the inhibitory functions of TRPM4, we

0 30 60 90 120 150 180

2.5

2.0

1.5

1

0

10ng/mL DNP

WTKO

Time(sec)

Rat

io (

340/

380)

10

20

30

40

0WT KO WT KO

DNP

Ca2+ (-)

Δ A

UC

(ra

tio•s

ec)

*

DNP

Ca2+ (+)

10ng/mL DNP 5µM Iono

WTKO

0 60 120 180 240 300 360

Time(sec)

2.0

1.5

1

0

Rat

io (

340/

380)

Ca2+ (-) Ca2+ (+)

10

20

30

40

50

60

0WT KO

DNP

1ng/mLΔ

AU

C (

ratio

•sec

)

*

WT KO

DNP

3ng/mL

WT KO

DNP

10ng/mL

WT KO

Iono

5µM

a b c d

5

e

1µg/mL Thaspi 5µM Iono

0 60 120 180 240 300

Time(sec)

4.0

3.0

1.0

0

Rat

io (

340/

380)

360 420

2.0

Ca2+ (-) Ca2+ (+)

WTKO

60

30

0

Δ A

UC

(ra

tio•s

ec)

WT KO

*

90

f

WTCont

RT (+) (-)(-)(+)

KO WT KO

Orai1

GAPDH

g

Fig. 4 Ca2+-imaging studies of WT and TRPM2-KO mBMMCs. Thedata points represent the means±S.E. a Average time courses of[Ca2+]cyt changes measured according to the fluorescence ratio inresponse to 10 ng/mL DNP-BSA in wild-type (WT) and TRPM2-KO(KO) mBMMCs. b The net area under a [Ca2+]cyt curve (ΔAUC)during DNP application (1 to 10 ng/mL, 30–150 s) (n=12 to 38). Cellviability was confirmed by applying 5 μM of ionomycin (Iono).The ΔAUC in response to Iono was calculated during the last30 s of the experiments, and the case at 10 ng/mL of DNP-BSAis shown here. *P<0.01 vs. WT. c Averaged time course of[Ca2+]cyt responses in WT (n=17) and KO (n=21) mBMMCs

following stimulation with DNP-BSA (10 ng/mL) in Ca2+-freesolution and the subsequent replacement with a 2 mM Ca2+ bathsolution. d ΔAUC during stimulation in a Ca2+-free solution (55–100 s; a magnified chart is presented above) and subsequentsolution exchange (160–270 s). *P<0.01 vs. WT. e Averagedtime course of [Ca2+]cyt responses following thapsigargin (Thaspi;1 μg/mL) treatment in Ca2+-free solution and the subsequentreplacement with a 2 mM Ca2+ bath solution in WT (n=54) andKO (n=80) mBMMCs. f ΔAUC after switching to the Ca2+-containingbuffer (270–360 s). *P<0.01 vs. WT. g Orai1 expression in wild-type(WT) and TRPM2-KO (KO) mBMMCs

6

3

0

Δ A

UC

(ra

tio•s

ec)

Control eco FFA

*

*10 ng/mL DNP

eco or FFA

0 60 120 180 240 300 360

Time(sec)

1.2

1

Rat

io(3

40/3

80)

1.4

0

econazoleFFA

Control

baIono

Fig. 5 Investigation ofmBMMCs from BALB/c miceusing TRPM2 channelinhibitors. a Averaged timecourse of [Ca2+]cyt responsesfollowing exposure to 10 μMeconazole (eco) and 200 μMFFA (n=20 to 40). b ΔAUCduring stimulation with DNP(120–210 s). The data representthe means±SE. *P<0.01 vs.control

1028 Pflugers Arch - Eur J Physiol (2013) 465:1023–1030

found that TRPM2 was involved in the facilitation of degran-ulation in MMCs by increasing [Ca2+]cyt.

In the present study, we found that TRPM2 blockadesuppressed antigen-induced degranulation and [Ca2+]cyt in-crease in mBMMCs using TRPM2-KO mice and TRPM2inhibitors. Experiments without extracellular Ca2+ (Fig. 4c,d) and with thapsigargin-induced activation (Figs. 3c and4e) suggested that the inhibitory effects of TRPM2 blockaderesult from a decrease in extracellular Ca2+ entry induced byCa2+ store depletion. Moreover, the expression of Orai1mRNA remained unchanged in TRPM2-KO mice(Fig. 4g). We think that although the store operated calciumentry (SOCE) is mainly mediated through Orai1, TRPM2also acts as one of SOCE channels independently of Orai1expression. Our results indicate that Orai1 and TRPM2participate in SOCE. Furthermore, in a whole-cell patch-clamp study, TRPM2 currents were observed in the plasmamembrane of mBMMCs (Fig 2a, b), and these currentamplitudes were reduced by all three TRPM2 inhibitors:econazole, FFA, and 2APB (Fig. 2c). Taken together, theseresults suggest that TRPM2 expressed in mBMMCs partic-ipates in antigen-induced extracellular Ca2+ influx andthereby facilitates degranulation.

Finally, we investigated the function of TRPM2 in vivo. Asdescribed above, MMCs accumulate and mature in mucosaltissues following exposure to Th2 cytokines, and they mainlyparticipate in the pathology of allergic diseases such as FA.Thus, the murine model of FA is suitable for investigation ofthe pathology caused byMMCs in vivo [18, 38]. According tothe typical protocols for the development of FA in mice,BALB/c mice develop allergic diarrhea after repeated OVAchallenges, whereas C57Bl/6 mice are resistant to the devel-opment of FA [38]. As TRPM2-KO mice on a BALB/c back-ground were not available, we used TRPM2 inhibitors in

experiments with BALB/c mice. These inhibitors are not spe-cific to TRPM2 and exhibit different chemical structures.However, these TRPM2 inhibitors suppressed antigen-induced degranulation in mBMMCs from both BALB/c andC57BL/6 mice (Fig. 3d, e). Thus, these results suggest that thesuppressive effects of these inhibitors are due to TRPM2blockade. Moreover, these inhibitors alleviated FA diarrheain vivo (Fig. 6). These results indicate that TRPM2 participatesin antigen-stimulated degranulation of MMCs in vivo.

However, it remains unclear how the TRPM2 channel isactivated in the food allergy model. As TRPM2 is known to beactivated by dinucleotides, such as β-NAD and ADPR, H2O2,cytosolic Ca2+, and high temperature [21, 25], we hypothesizethat the release of Ca2+ from ER stores via IP3 productioninduced by antigen stimulation can increase the levels of[Ca2+]cyt, thereby activating TRPM2. Furthermore, mast cellsare capable of generating H2O2 [5, 10, 27], which suggests thatTRPM2 activation may be synergistically facilitated by theincrease in [Ca2+]cyt resulting from antigen stimulation, theincrease in temperature caused by allergic inflammation, andthe release of inflammatory mediators from other mast cells orother immune cells.

Currently, no therapeutic drugs are available for the treat-ment of FA, and the most common solution is the avoidance offood antigens that provoke these allergic responses. MMCshave been reported to participate in rhinitis and asthma inaddition to FA [35]. Thus, our findings suggest that blockadeof TRPM2 may provide a novel strategy for the treatment ofnot only food allergy but also these other allergic diseases.

In conclusion, TRPM2 is closely involved in the activation ofmast cells by facilitating extracellular Ca2+ influx.Pharmacological inhibition of TRPM2 may lead to the devel-opment of useful drugs for patients suffering from food allergies.

Acknowledgments We thank H. Mihara for help with cell prepara-tion and operation of measuring instruments, M. Kashio for animalbreeding and management, and K. Tsuneyama and M. Fujimoto forhelp with immunohistochemical staining. This research was supportedby a Grant-in-Aid for Scientific Research from theMinistry of Education,Culture, Sports, Science and Technology of Japan to M. Kadowaki (nos.21590760 and 24590879) and by the Knowledge Cluster InitiativeProgram of the Ministry of Education, Culture, Sports, Scienceand Technology of Japan to M. Kadowaki.

Conflict of interest The authors have declared that no conflict ofinterest exists.

References

1. Bradding P, Okayama Y, Kambe N, Saito H (2003) Ion channelgene expression in human lung, skin, and cord blood-derived mastcells. J Leukoc Biol 73:614–620

2. Brandt EB (2003) Mast cells are required for experimental oralallergen-induced diarrhea. J Clin Invest 112:1666–1677

Dia

rrhe

a s

core

1

2

3

4

5

0eco FFAControl

**

Fig. 6 Fecal evaluation using the stool visual scale after the fourthOVA challenge (50 mg, on the seventh day). Econazole (eco; 30 μg/gbw) and FFA (100 μg/g bw) were administered prior to the experi-ments. *P<0.01

Pflugers Arch - Eur J Physiol (2013) 465:1023–1030 1029

3. Carter RN, Tolhurst G, Walmsley G, Vizuete-Forster M, Miller N,Mahaut-Smith MP (2006) Molecular and electrophysiologicalcharacterization of transient receptor potential ion channels in theprimary murine megakaryocyte. J Physiol (Lond) 576:151–162

4. Gilfillan AM, Tkaczyk C (2006) Integrated signalling pathwaysfor mast-cell activation. Nat Rev Immunol 6:218–230

5. Hayama K, Suzuki Y, Inoue T, Ochiai T, Terui T, Ra C (2011) Goldactivates mast cells via calcium influx through multiple H2O2-sensitive pathways including L-type calcium channels. Free RadicBiol Med 50:1417–1428

6. Heiner I, Eisfeld J, Lückhoff A (2003) Role and regulation of TRPchannels in neutrophil granulocytes. Cell Calcium 33:533–540

7. Hill K, Benham CD, McNulty S, Randall AD (2004) Flufenamicacid is a pH-dependent antagonist of TRPM2 channels. Neuro-pharmacology 47:450–460

8. Hill K, McNulty S, Randall AD (2004) Inhibition of TRPM2channels by the antifungal agents clotrimazole and econazole.Naunyn Schmiedebergs Arch Pharmacol 370:227–237

9. Inada H, Iida T, Tominaga M (2006) Different expression patternsof TRP genes in murine B and T lymphocytes. Biochem BiophysRes Commun 350:762–767

10. Inoue T, Suzuki Y, Yoshimaru T, Ra C (2008) Reactive oxygenspecies produced up- or downstream of calcium influx regulateproinflammatory mediator release from mast cells: role of NADPHoxidase and mitochondria. Acta Bioenerg 1783:789–802

11. Ishii M, Shimizu S, Hara Y, Hagiwara T, Miyazaki A, Mori Y,Kiuchi Y (2006) Intracellular-produced hydroxyl radical mediatesH2O2-induced Ca2+ influx and cell death in rat beta-cell line RIN-5F. Cell Calcium 39:487–494

12. Kageyama-Yahara N, Suehiro Y, Maeda F, Kageyama S-I,Fukuoka J, Katagiri T, Yamamoto T, and Kadowaki M (2010)Pentagalloylglucose down-regulates mast cell surface FcεRI ex-pression in vitro and in vivo. FEBS Letters 584:111–118

13. Kageyama-Yahara N, Suehiro Y, Yamamoto T, Kadowaki M(2008) IgE-induced degranulation of mucosal mast cells is nega-tively regulated via nicotinic acetylcholine receptors. BiochemBiophys Res Commun 377:321–325

14. Kageyama-Yahara N, Wang X, Katagiri T, Wang P, Yamamoto T,Tominaga M, Kadowaki M (2011) Suppression of phospholipaseCγ1 phosphorylation by cinnamaldehyde inhibits antigen-inducedextracellular calcium influx and degranulation in mucosal mastcells. Biochem Biophys Res Commun 416:283–288

15. Kashio M, Sokabe T, Shintaku K, Uematsu T, Fukuta N, KobayashiN, Mori Y, Tominaga M (2012) Redox signal-mediated sensitizationof transient receptor potential melastatin 2 (TRPM2) to temperatureaffects macrophage functions. Proc Natl Acad Sci USA 109:6745–6750

16. Lange I, Penner R, Fleig A, Beck A (2008) Synergistic regulationof endogenous TRPM2 channels by adenine dinucleotides in pri-mary human neutrophils. Cell Calcium 44:604–615

17. Lange I, Yamamoto S, Partida-Sanchez S, Mori Y, Fleig A, PennerR (2008) TRPM2 functions as a lysosomal Ca2+-release channelin beta cells. Sci Signal 2:ra23–ra23

18. McClain S, Bannon G (2006) Animal models of food allergy:opportunities and barriers. Curr Allergy Asthma Rep 6:141–144

19. Metcalfe DD, Baram D, Mekori YA (1997) Mast cells. PhysiolRev 77:1033–1079

20. Morita T, Tanimura A, Baba Y, Kurosaki T, Tojyo Y (2009) A Stim1-dependent, noncapacitative Ca2+-entry pathway is activated by B-cell-receptor stimulation and depletion of Ca2+. J Cell Sci 122:1220–1228

21. Perraud AL, Fleig A, Dunn CA, Bagley LA, Launay P, Schmitz C,Stokes AJ et al (2001) ADP-ribose gating of the calcium-permeable

LTRPC2 channel revealed by Nudix motif homology. Nature411:595–599

22. Prasad P, Yanagihara AA, Small-Howard AL, Turner H, Stokes AJ(2008) Secretogranin III directs secretory vesicle biogenesis inmast cells in a manner dependent upon interaction with chromog-ranin A. J Immunol 181:5024–5034

23. Rivera J, Gilfillan AM (2006) Molecular regulation of mast cellactivation. J Allergy Clin Immun 117:1214–1226

24. Sicherer SH, Sampson HA (2010) Food allergy. J Allergy ClinImmun 125:S116–S116

25. Starkus J, Beck A, Fleig A, Penner R (2007) Regulation of TRPM2by extra- and intracellular calcium. J Gen Physiol 130:427–440

26. Sumoza-Toledo A, Penner R (2011) TRPM2: a multifunctional ionchannel for calcium signalling. J Physiol (Lond) 589:1515–1525

27. Swindle EJ, Metcalfe DD (2007) The role of reactive oxygenspecies and nitric oxide in mast cell-dependent inflammatory pro-cesses. Immunol Rev 217:186–205

28. Takahashi N, Kozai D, Kobayashi R, Ebert M, Mori Y (2011)Roles of TRPM2 in oxidative stress. Cell Calcium 50:279–287

29. Tegoshi T, Nishida M, Arizono N (2005) Expression and role of E-cadherin and CD103β7 (αEβ7 integrin) on cultured mucosal-typemast cells. APMIS 113:91–98

30. Togashi K, Inada H, Tominaga M (2009) Inhibition of the transientreceptor potential cation channel TRPM2 by 2-aminoethoxydiphenylborate (2-APB). Br J Pharmacol 153:1324–1330

31. Togashi K, Hara Y, Tominaga T, Higashi T, Konishi Y, Mori Y,Tominaga M (2006) TRPM2 activation by cyclic ADP-ribose atbody temperature is involved in insulin secretion. EMBO J25:1804–1815

32. Uchida K, Dezaki K, Damdindorj B, Inada H, Shiuchi T, Mori Y,Yada T, Minokoshi Y, Tominaga M (2011) Lack of TRPM2 im-paired insulin secretion and glucose metabolisms in mice. Diabetes60:119–126

33. Vennekens R, Olausson J, Meissner M, Bloch W, Mathar I, PhilippSE, Schmitz F et al (2007) Increased IgE-dependent mast cellactivation and anaphylactic responses in mice lacking thecalcium-activated nonselective cation channel TRPM4. Nat Immu-nol 8:312–320

34. Wang X, Kageyama-Yahara N, Hayashi S, Yamamoto T, KadowakiM (2012) Sphingosine kinase-1-dependent and -independent inhibi-tory effects of zanthoxyli fructus to attenuate the activation of muco-sal mast cells and ameliorate food allergies in mice. Evid BasedComplement Alternat Med 2012:862743

35. Westerberg CM, Ullerås E, Nilsson G (2012) Differentiation ofmast cell subpopulations from mouse embryonic stem cells. JImmunol Methods 382:160–166

36. Wissenbach U, Philipp SE, Gross SA, Cavalié A, Flockerzi V (2006)Primary structure, chromosomal localization and expression in im-mune cells of the murine ORAI and STIM genes. Cell Calcium42:439–446

37. Yamamoto S, Shimizu S, Kiyonaka S, Takahashi N, Wajima T,Hara Y, Negoro T et al (2008) TRPM2-mediated Ca2+ influxinduces chemokine production in monocytes that aggravates in-flammatory neutrophil infiltration. Nat Med 14:738–747

38. Yamamoto T, Fujiwara K, Yoshida M, Kageyama-Yahara N,Kuramoto H, Shibahara N, Kadowaki M (2009) Therapeutic effect ofKakkonto in a mouse model of food allergy with gastrointestinalsymptoms. Int Arch Allergy Immunol 148:175–185

39. Zhang D, Spielmann A,Wang L, Ding G, Huang F, GuQ, SchwarzW(2012) Mast-cell degranulation induced by physical stimuli involvesthe activation of transient-receptor-potential channel TRPV2. PhysiolRes 61:113–124

1030 Pflugers Arch - Eur J Physiol (2013) 465:1023–1030