Chemicals inducing acute irritant contact dermatitis mobilize intracellular calcium in human keratinocytes

Toxicology in Vitro 27 (2013) 402–408

Contents lists available at SciVerse ScienceDirect

Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit

Chemicals inducing acute irritant contact dermatitis mobilize intracellular calcium in human keratinocytes Matthieu Raoux a,1,2, Nathalie Azorin a,2, Cécile Colomban b, Stéphanie Rivoire b, Thierry Merrot c, Patrick Delmas a, Marcel Crest a,⇑ a b c

Université de la Méditerranée, Centre National de la Recherche Scientifique UMR 6231, Marseille, France Eurofins/ATS, Actimart, Aix-en-Provence, France Université de la Méditerranée, Assistance Publique Hôpitaux de Marseille (APHM), Marseille, France

a r t i c l e

i n f o

Article history: Received 4 July 2011 Accepted 2 August 2012 Available online 10 August 2012 Keywords: Chemical irritant In vitro irritation model Irritant contact dermatitis Skin Keratinocyte Calcium signaling

a b s t r a c t Intracellular Ca2+ increase is a common feature of multiple cellular pathways associated with receptor and channel activation, mediator secretion and gene regulation. We investigated the possibility of using this Ca2+ signal as a biomarker for a reaction to chemical irritants of normal human keratinocytes (NHK) in submerged primary cell culture. We tested 14 referenced chemical compounds classified as strong (seven), weak (four) or non- (three) irritants in acute irritant contact dermatitis. We found that the strong irritant compounds tested at 20–40 mM induced an intracellular Ca2+ increase measurable by spectrofluorimetry in an automated test. Weak and non-irritant compounds however did not increase intracellular Ca2+ concentration. We further investigated the mechanisms by which the amine heptylamine, classified as a R34 corrosive compound, increases intracellular Ca2+. Heptylamine (20 mM) induced an ATP release that persisted in the absence of intra- and extra-cellular Ca2+. In addition, we found that this ATP activates NHK purinergic receptors that subsequently cause the increase in intracellular Ca2+ from sarcoplasmic reticular stores. We conclude that measuring the intracellular Ca2+ concentration in NHK is a suitable and easy way of determining any potential reaction to soluble chemical compounds. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction The epidermis, being the topmost living layer of the skin, is constantly exposed to external injury from such as mechanical stimulation, chemical irritants and noxious temperatures. Contact dermatitis is one of the most frequent skin diseases. It comprises

Abbreviations: ATP, adenosine triphosphate; [Ca2+]i, intracellular calcium concentration; DMEM, Dulbecco’s Modified Eagle’s Medium; DMSO, dimethyl sulfoxide; EGF, epidermal growth factor; EGTA, ethyleneglycol-bis(b-aminoethyl)N,N,N0 ,N0 -tetraacetoxymethyl ester; F/F0, normalized ratio of the Fluo-4 AM fluorescence (F) to the basal fluorescence (F0); HBSS, Hanks’ Balanced Salt Solution; IL1a, interleukine 1a; IL8, interleukine 8; NHK, normal human keratinocytes; NRR, neutral red release; PGE2, prostaglandin 2; PLC, phospholipase C; TNFa, tumor necrosis factor a; DMIPA, dimethylisopropylamine; PPADS, pyridoxal-phosphate6azophenyl-20 -40 -disulfonate; SLS, sodium lauryl sulfate; SEM, standard error of the mean. ⇑ Corresponding author. Address: Université de la Méditerranée, CNRS UMR 6231, Centre de Recherche en Neurobiologie-Neurophysiologie de Marseille, Faculté de Médecine, 51 Bd Pierre Dramard, CS80011, 13344 Marseille, Cedex 20, France. Tel.: +33 491 69 89 75; fax: +33 491 69 89 77. E-mail address: [email protected] (M. Crest). 1 Present address: Institut Européen de Chimie-Biologie, Centre National de la Recherche Scientifique UMR 5248, Université de Bordeaux, Bordeaux, France 2 These authors contributed equally to this work. 0887-2333/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tiv.2012.08.010

a major portion of occupational dermatoses in industrialized societies, resulting in considerable social and economic implications. It can be divided into irritant contact dermatitis and allergic contact dermatitis depending on the production or not of specific antibodies. Irritant contact dermatitis is defined as a localized inflammation of the skin caused by contact with toxic compounds such as metals, cleaning solutions, detergents, cosmetics, industrial chemicals and latex rubber. Depending on the time course of the skin reaction, irritation may be classified as acute, delayed or chronic i.e. developing slowly after exposure (Chew and Maibach, 2003). Historically, the irritation index during acute irritant dermatitis due to a single exposure to a chemical has been determined in vivo using the Draize skin irritation test on rabbits (Draize et al., 1944). As a measure of acute irritancy potential, each chemical is given a score based on erythema and edema grade (Bagley et al., 1996). Despite the universal acceptance of this assay, the correlation between animal and human irritancy has come under question since, for some cases, chemicals have been misclassified using in vivo rabbit data (York et al., 1996). In view of this as well as ethical concerns, the 7th amendment to the European Cosmetics Directive stimulated the development of alternative tests for the assessment of the potential toxicological effects of substances.

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

403

Recently, the scientific advisory committee of the European Centre for the Validation of Alternative Methods (ECVAM) began accepting the use of in vitro tests on reconstructed epidermis to distinguish between skin irritating and non-irritating chemicals (Gerberick et al., 2008). Cellular end points measured in alternative skin irritation assays concern keratinocyte viability and cytokine (IL1a, TNFa) release (see review in Macfarlane et al., 2009; Gibbs, 2009). The viability is determined by measuring cell cytotoxicity (Gay et al., 1992; Triglia et al., 1991), loss of membrane integrity (Faller and Bracher, 2002; Osborne and Perkins, 1994) and mitochondrial metabolism (Chan et al., 2005), and the cytokine release is measured by ELISA. Irritant and corrosive compounds may also induce the secretion by keratinocytes of other inflammatory mediators such as PGE2 (Corsini and Galli, 1998; Coquette et al., 2003) and ATP (Mizumoto et al., 2003). However, none of these assays considers the intracellular Ca2+ concentration ([Ca2+]i) as a putative biomarker of acute irritancy potential. The first advantage is that the intracellular Ca2+ signal is ubiquitous and corresponds well with the diversity of the cellular response expected using a large spectrum of chemicals. Another advantage is that the [Ca2+]i is easy to measure using a fluorescent probe and experiments can be miniaturized in 96 well dishes using an automated spectrofluorimeter. With this in mind, we firstly measured the variations in [Ca2+]i of NHK in response to 14 chemical compounds chosen according to their skin irritancy potential and structural diversity, and found a link between Ca2+ increase and irritancy potential. We then analyzed the cellular mechanisms by which the skin corrosive compound heptylamine induces intracellular Ca2+ increase. Here we demonstrate how chemical stimulations by heptylamine, at concentrations that do not alter cell viability (10–20 mM), cause a Ca2+-independent release of ATP from NHK which then activates metabotropic purinergic receptors, in turn bringing about the increase in [Ca2+]i. We conclude that this Ca2+ mobilization in NHK in response to chemical stimuli may be a relevant biomarker to detect potential keratinocyte reactions in vitro and test compound toxicity in acute irritant contact dermatitis.

Ham’s F-12 (30%; Invitrogen), fetal calf serum (FCS, 10%; Invitrogen), L-glutamine (2 mM; Invitrogen), penicillin–streptomycin (100 U/ml; Sigma–Aldrich, St Louis, M), insulin (5 lg/ml; Sigma), hydrocortisone (0.5 lg/ml; Sigma) and EGF (10 ng/ml; Sigma). To obtain keratinocyte proliferation, cells were plated at 80.103 cells/cm2 and co-cultured with feeder mitomycin C-treated 3T3 fibroblasts (30.103 cells/cm2) in Falcon flasks (37 °C, 95% O2–5% CO2). Keratinocytes were allowed to proliferate and the culture medium was renewed every 2–3 days. Having reached confluence after 8 to 15 days in vitro, the NHK were isolated using a two-step trypsination procedure: co-cultures were treated with trypsin–EDTA in HBSS (0.5 mg/ml) for 1 min to dislodge inactivated 3T3 fibroblasts and then NHK were dislodged by a longer trypsin–EDTA treatment (3 min). The NHK were then centrifuged at 1200 r.p.m. (200 g) for 6 min. Samples were either re-suspended in DMEM supplemented with FCS (20%) and DMSO (10%) for cryopreservation (3.106 cells/ml in cryovials) or were used for experiments. For experiments, the NHK were resuspended in the complete culture medium described above and cultivated at 8.103 cells/cm2 in black 96-well dishes for 5–7 days without 3T3 cells. Experiments were performed on NHK having reached confluence.

2. Materials and methods

2.3. Extracellular ATP measurements

2.1. Normal human keratinocyte (NHK) primary cultures

The luciferin-luciferase detection of ATP was performed in Krebs solution using the Tecan infinite F500 spectrofluorimeterluminometer. NHK were isolated using the two-step trypsination procedure and cultured without 3T3 fibroblasts for 2 to 5 days at 4.103 cells/cm2 in white 96-well Greiner dishes with clear bottoms. Dishes were placed in Krebs solution 10 min before the experiment in the presence of luciferin-luciferase (FLAAM, Sigma-Aldrich) at a final concentration of 0.04%. To determine the amount of ATP released from the cells, a calibration curve was also constructed using known concentrations of ATP in solution (10, 100, 1000, 10000 pM) according to product information (Sigma technical bulletin n° BAAB-1). Controls were performed with each drug solution to eliminate any drug effect on luciferase activity and to check for ATP contamination in stock solutions.

Normal human keratinocyte (NHK) primary cultures were prepared from healthy human foreskin biopsies. Samples were obtained as surgical waste tissues. This study has been performed according to the Declaration of Helsinki. Informed consent of the patients or their legal representative was in agreement with the article L1245-2 of the French Public Health Code relating to the use of surgical waste tissues. Skin explants were conserved at 4 °C for up to 12 h in Hanks’ balanced salt solution (HBSS; Invitrogen, Carlsbad, CA) supplemented with penicillin–streptomycin (200 U/ml; Invitrogen). Underlying mesenchyme was removed and the explants then decontaminated by washing in HBSS supplemented with ethanol (30%, v/v) and in HBSS supplemented with penicillin–streptomycin (200 U/ml). The explants were then cut into 0.5 cm2 pieces and incubated for 16 to 20 h at 4 °C with dispase grade II in PBS (24 U/ml; Roche, Indianapolis, IN) supplemented with penicillin–streptomycin (100 U/ml). Epidermal layers were lifted from the dermis and mechanically dissociated before being incubated in HBSS supplemented with trypsin–EDTA (0.5 mg/ml; Invitrogen) for 20 min at 37 °C. The resulting epidermal cell suspension was filtered through a cellular strainer (pore size 70 lm) and centrifuged at 1200 r.p.m. (200 g) for 6 min. Epidermal cells were re-suspended and cultivated in Dulbecco’s Modified Eagle’s Medium (DMEM; Invitrogen) supplemented with

2.2. Intracellular Ca2+ measurements All experiments were performed at room temperature (22– 25 °C) in an isotonic (296 ± 5 mOsM) Krebs solution consisting of (in mM): 130 NaCl, 3 KCl, 1 MgCl2, 10 Hepes, 2.5 CaCl2 and 10 glucose. Variations in the [Ca2+]i were measured with the Tecan infinite F500 spectrofluorimeter-luminometer (Tecan, Research Triangle Park, NC) on NHK isolated using the two-step trypsination procedure and cultured without 3T3 fibroblasts in black 96-well Greiner (Dutscher, France) dishes with clear bottoms. Cells were loaded with Fluo-4 AM (5 lM; Invitrogen) for 30 min (37 °C, 95% O2–5% CO2) and washed twice with Krebs solution. Fluo-4 fluorescence (F) was recorded by exciting the probe at 485 nm. Variation in the calcium concentration was normalized to the basal fluorescence (F0) as the ratio F/F0.

2.4. Chemicals and solutions Experiments were performed in isotonic Krebs solution. To maintain iso-osmolarity, some solutions were prepared with an appropriate reduction in NaCl, which was replaced by D-mannitol in the corresponding control solutions. A Ca2+-free external solution was prepared by substituting Ca2+ with Na+ in the presence of EGTA (400 lM). All chemicals were purchased from SigmaAldrich. The classification of chemicals given by Bagley et al.

404

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

Table 1 Concentrations of chemicals tested in the fluorescence assay. Chemicals were tested at three or four concentrations from 0.2 to 40 mM. At concentrations of 20 mM, we have given the corresponding concentration as a percentage of chemical diluted in Krebs saline (mass of chemical in g/100 g of Krebs).

Strong irrirtant Heptylamine DMIPA 2-4-Hexadienal Piperazine SLS KOH HCl Weak irritant Sodium bisulphate Sodium bicarbonate Linalool oxide Benzyl alcohol Non irritant Isobutyraldehyde Dipropylène glycol Phenoxy-ethanol

0.2 mM

2 mM

10 mM

20 mM

% m/m

40 mM

+ + + + + + +

+ + + + + + +

+ ND ND ND ND + ND

+ + + + + + +

0.23 0.17 0.20 0.17 0.54 0.11 0.07

ND + + + ND ND ND

+ + + +

+ + + +

ND ND ND ND

+ + + +

0.24 0.16 0.30 0.21

+ + + +

+ + +

+ + +

ND ND ND

+ + +

0.14 0.26 0.27

+ + +

ND: Not determined. +: Tested. % m/m: Percentage in mass/mass.

(1996), based on in vivo rabbit irritation tests, was used to choose tested compounds. We used heptylamine, dimethylisopropylamine (DMIPA), 2-4-hexadienal, piperazine, sodium lauryl sulfate (SLS), potassium hydroxide (KOH) and hydrochloric acid (HCl) as the strong irritant compounds; sodium bisulphate, sodium bicarbonate, linalool oxide and benzyl alcohol as the weak irritant chemicals; and isobutyraldehyde, dipropylene glycol and phenoxyethanol as the non-irritant compounds. All chemicals were solubilized in water. The pH value of each solution was adjusted to 7.35 with NaOH or HCl, except when the effects of HCl and KOH were specifically addressed. Table 1 summarizes the concentrations of the investigated chemicals. In some experiments, cells were incubated for 10–30 min (37 °C, 95% O2–5% CO2) with the intracellular Ca2+ chelator EGTA AM (100 lM; Tebu Bio, France), thapsigargin (1 lM; SigmaAldrich), the phospholipase C (PLC) inhibitor U-73122 (10 lM; Sigma-Aldrich) or its inactive analog U-73343 (10 lM), suramin (300 lM; Sigma-Aldrich), PPADS (100 lM; Sigma-Aldrich) and apyrase (50 U/ml; Sigma-Aldrich). These drugs were solubilized in ethanol or DMSO to final concentrations of 60.1% and 60.01% respectively (verified as having no effect on the NHK [Ca2+]i and ATP release. 2.5. Neutral red release test The neutral red release (NRR) assay was performed as previously described by Korting et al., 1994 with minor modifications. NR dye was added to NHK in culture medium and incubated for 3 h at 37 °C, 95% O25% CO2. The NHK were then washed twice with PBS and exposed to chemical irritant for 10 min. Benchmarks with culture medium and PBS were included in each plate. After treatment, wells were washed twice with PBS to remove the chemical irritant. A solution containing 50% ethanol and 1% acetic acid in distilled water was added to extract the NR dye. After 10 min on a shaker, the NR absorbance was measured at 550 nm in the Tecan Infinite 500 spectrofluorimeter. Mortality percentages were calculated from the equation:

Mortality ð%Þ ¼ ½100  ðmean optic density from treated wells= optic density from control wellsÞ  100:

2.6. Data analysis [Ca2+]i and ATP measurements were analyzed using Magellan 4.0 software (Tecan), Excel software (Microsoft, WA), and PRISM 4.0 software (GraphPad Software Inc., CA). Results are given as mean ± S.E.M. Illustrations in Fig. 1 are the mean ± S.E.M of n = 16 to 36 wells from 4–7 independent dishes (4–6 wells/dish). In the neutral red release (NRR) assay, chemicals were tested in 3 to 4 independent replicates. Cellular mortality was calculated as the mean ± S.EM. of n = 18 to 24 wells from 3 or 4 independent dishes (6 wells/dish). Traces in Figs. 3 and 4 are representative results obtained with n = 4 to 12 wells in one dish. Each experiment was performed 3 to 4 times as independent replicates and the total pooled to calculate the mean ± S.E.M. 3. Results 3.1. Irritant compounds trigger intracellular calcium increase We tested the skin irritancy potential of 14 structurally diverse chemical compounds on NHK. This collection contained 7 strong irritant, 4 weak irritant and 3 non irritant compounds (Table 1). Each of the skin irritants increased [Ca2+]i as shown by the increase in normalized fluorescence (F/F0) in proportion to the concentration of the irritant (Fig. 1a). At 40 mM, piperazine, 2-4-hexadienal and DMIPA caused a [Ca2+]i increase with a maximum F/F0 ((F/F0) max) at 4.47 ± 0.26 (n = 16), 1.93 ± 0.14 (n = 16) and 1.40 ± 0.03 (n = 20), respectively (Fig. 1a). At 20 mM, SLS, KOH, HCl and heptylamine caused a [Ca2+]i increase with a (F/F0)max at 2.00 ± 0.68 (n = 20), 10.16 ± 0.13 (n = 16), 1.44 ± 0.32 (n = 16) and 1.82 ± 0.05 (n = 36), respectively. In contrast, the [Ca2+]i of the NHK did not increase significantly with the Krebs alone or with the weak and non irritant compounds sodium bisulphate, sodium bicarbonate, linalool oxide, benzyl alcohol, isobutyraldehyde, dipropylene glycol or phenoxyethanol (all tested at 40 mM; (F/F0) max < 0.2, n = 16 to 20) (Fig. 1b and c). 3.2. Strong irritant compounds cause weak cellular mortality We tested whether the strong irritant compounds investigated induce NHK membrane disruption using the NRR assay (Fig 2). The surfactant SLS, known to disrupt plasma membranes, was tested as a positive control and induced 78 ± 1.2% of cellular mortality at 40 mM after 10 min exposure. 2-4-hexadienal (40 mM), DMIPA (40 mM) and acidic pH (2.1) induced a cellular mortality of 18 ± 1.3%, 11.3 ± 0.7% and 21.5 ± 8%, respectively. Interestingly, piperazine, heptylamine (40 mM) and alkaline pH (11.4) did not induce membrane disruption indicating that [Ca2+]i can be used to detect the irritant properties of such chemicals at concentrations that do not cause cell mortality. Each chemical was tested in 3 to 4 independent dishes with 6 wells measured in each dish. 3.3. Heptylamine causes the release of Ca2+ from endoplasmic reticulum stores in NHK Figure 3a illustrates that heptylamine (10 mM) caused a transient [Ca2+]i increase with a normalized fluorescence (F/F0) peak at 1.85 ± 0.23 (n = 6) (Fig. 3a). The mean peak ± S.E.M for overall experiments was 1.73 ± 0.19 (n = 36). F/F0 variations were directly related to [Ca2+]i since they were abolished in NHK loaded with the cytosolic Ca2+ chelator EGTA AM (100 lM) (Fig. 3b). Removal of extracellular Ca2+ had no effect on the [Ca2+]i increase caused by heptylamine. The normalized fluorescence (F/F0) peaked at 3.17 ± 0.28 under heptylamine 20 mM (n = 7) and the mean

405

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

F/F0 Fluo4

(a)

2-4 hexadienal

Piperazine 3

3

4

2

2

2

1

1

0 1

10

100

0

8

7.2

1

7.3 7.4

100

0

10

100

0

1

10

DMIPA

2.1

2

2

1

1

100 Heptylamine

6.7

9.4

0 1

10

2

11.4

0

1

HCl

KOH

4

0

0 0

12

SLS

6

0

1

10

0

0

100

0

1

10

100

0

1

10

100

Concentrations (mM)

F/F0 Fluo4

(b)

Na Bisulfate

Benzyl alcohol

Linalool oxide

Na Bicarbonate

2

2

2

2

1

1

1

1

0

0 0

1

10

100

0

0 0

1

10

100

0

1

10

0

100

10

1

100

Concentrations (mM)

F/F0 Fluo4

(c) 2

Dipropylene glycol

Isobutyraldehyde

1 0

Phenoxyethanol

2

2

1

1 0

0 0

1

10

100

0

1

10

100

0

1

10

100

Concentrations (mM) Fig. 1. Seven strong chemical irritants increase intracellular calcium in NHK. Chemicals were injected into wells at the indicated concentrations. F/F0 Fluo4 corresponds to the change induced in the mean (±S.E.M) normalized Fluo-4 fluorescence after application of strong irritant (1a), weak irritant (1b) and non irritant (1c) chemical compounds. Red numbers represent pH. Each point is the mean ± S.EM. of n = 16 to 36 wells from 4–7 independent dishes (4–6 wells/dish). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.4. Heptylamine causes a [Ca2+]i-independent ATP release in NHK Since heptylamine has previously been shown to induce the release of ATP (Mizumoto et al., 2003), we postulated that purinergic

100

Cellular Mortality (%)

peak ± S.E.M for overall experiments was 3.02 ± 0.11 (n = 18) (Fig. 3c). These results suggest that heptylamine triggers a Ca2+ release from intracellular stores. We therefore depleted endoplasmic reticulum stores of Ca2+ by inhibiting the Ca2+-ATPase with thapsigargin (1 lM, 30 min) (inset in Fig. 3c). In response to heptylamine, no [Ca2+]i increase was detected in thapsigargin-pretreated NHK, even in the presence of external Ca2+ (Fig. 3c). Since the release of Ca2+ from the endoplasmic reticulum secondary to the production of inositol triphosphate by phospholipase C has been previously shown in keratinocytes (Zholos et al., 2005), we hypothesized that the [Ca2+]i increase caused by heptylamine resulted from activation of the phospholipase C (PLC) pathway. NHK pretreated with the PLC inhibitor U-73122 (10 lM) were unable to mobilize Ca2+ in response to heptylamine 20 mM (Fig. 3d, n = 6 and overall experiments n = 18, F/F0 peak to 1.09 ± 0.16), in contrast to those loaded with its inactive analog U-73343 (10 lM) (Fig. 3d, n = 6 and overall experiments n = 18, F/F0 peak to 1.77 ± 0.20). We conclude that heptylamine increases [Ca2+]i via the mobilization of reticular stores.

80 60 HCl pH=2.1

40 2-4 hexadienal

20

Piperazine

0 - 20

20

40 SLS

40 40 DMIPA

40

pH=11.4 20

20 Heptylamine

20

KOH

(mM) Fig. 2. Cytoxicity of seven strong chemical irritants on skin. Percentage of cellular mortality (±S.E.M) observed after treatment with 20 and 40 mM of SLS, 40 mM of 24-hexadienal, piperazine, DMIPA and 20 mM of KOH, HCl and heptylamine. SLS was used as a positive control known to induce cellular mortality. Other irritants, acidic and alkaline pH led to low cellular mortality (<22%). Each column is the mean ± S.EM. of n = 18 to 24 wells from 3 or 4 independent dishes (6 wells/dish).

signaling was a possible mechanism behind the activation of the PLC pathway in response to heptylamine. We performed measurements of the extracellular ATP concentration ([ATP]e) based on the luciferin-luciferase bioluminescent reaction. We measured the

406

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

(b)

1 min

F/F0 Fluo-4

F/F0 Fluo-4

(a) 2.0 1.5

10 mM 2 mM

1.0

0.2 mM Hept.

3.0

(d)

3.0 2.0 1.0 Thaps

2.0

1 min

1.0

+ EGTA AM 1.0

5 min

F/F0 Fluo-4

F/F0

F/F0 Fluo-4

0Ca2+

1.5

Hept. (20mM) 4.0

(c) 4.0

1 min

2.0

1 min

2.0 1.5

2.5Ca2+ +U-73343

1.0

2.5Ca2+ +U-73122

2.5Ca2+ +Thaps Hept (20 mM)

Hept (20mM). 2+

[ATP] pM

(a) 400 200

30s

600 400 200 0

(b) 400 pH10.9

1 min

[ATP] pM

[ATP] pM

Fig. 3. Heptylamine triggers intracellular calcium release from reticular stores. (a) [Ca ]i variations in NHK after application of heptylamine (Hept) at 0.2, 2 and 10 mM (n = 6 wells/one dish for each trace). (b) Heptylamine (20 mM) did not induce [Ca2+]i increase in NHK loaded with 100 lM EGTA AM, (n = 12 wells/one dish). (c) [Ca2+]i changes induced by 20 mM heptylamine in a Ca2+ free external solution (0Ca, upper trace) and in cells treated for 30 min with 1 lM thapsigargin (Thaps) in a Ca2+-containing external solution (2.5Ca2+, lower trace) (n = 4 wells/one dish for each trace). Inset: Changes in [Ca2+]i induced by application of thapsigargin 1 lM in a Ca2+-containing external solution (n = 4 wells/one dish). (d) Inhibition of the heptylamine effect by the PLC inhibitor U-73122 (10 lM). The inactive analog U-73343 (10 lM) does not inhibit the [Ca2+]i increase induced by heptylamine (n = 6 wells/one dish).

+ EGTA AM 200

1 min 0

0

Hept.

Hept.

F/F0 Fluo-4

1.5

1 min

1.25 1

F/F0 Fluo-4

(d) 2

(c) 1.75

1 min

1.5

1

suramin

suramin Heptylamine 20 mM

ATP 500 nM

(e) F/F0 Fluo-4

1 min

1.5 1.25 1

PPADS Heptylamine 20 mM

F/F0 Fluo-4

(f)

1.75

1.75 1.5

1 min

1.25 1

apyrase Heptylamine 20 mM

Fig. 4. Heptylamine led to ATP release via a Ca2+-independent pathway. (a) Changes in the mean (±S.E.M) [ATP] induced by 20 mM heptylamine (n = 6 wells/one dish). (b) Extracellular [ATP] changes induced by 20 mM heptylamine in NHK loaded with 100 lM EGTA AM (n = 4 wells/one dish). (c) The increase in [Ca2+]i induced by 500 nM ATP was abolished by the purinergic receptor inhibitor suramin (300 lM) (n = 6 wells/one dish). (d–f) The increase in [Ca2+]i induced by 20 mM heptylamine was abolished by the purinergic receptor inhibitors suramin (300 lM, n = 18 wells/three dishes) (d), and PPADS (100 lM, n = 18 wells/three dishes) (e), and by the ATPase apyrase (50 U/ml, n = 18 wells/three dishes) (f).

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

changes induced on the mean (±S.E.M) [ATP]e in NHK after heptylamine application. As shown in Figure 4a, the [ATP]e measured under heptylamine (20 mm) peaked at 325 ± 149 pM after 21.87 ± 2.13 s (n = 7 wells). The mean peak ± S.E.M for overall experiments was 298 ± 132 pM (n = 18). Figure 4b illustrates that when the NHKs were loaded with the intracellular Ca2+-chelator EGTA AM (100 lM), the same [ATP]e was observed (Fig. 4b peak at 332 ± 43 pM, n = 4, overall experiments 307 ± 55 pM, n = 12), thus indicating that heptylamine causes ATP release from NHK through a [Ca2+]i-independent mechanism. We postulated that this heptylamine-induced ATP release activates the P2Y receptor which in turn triggers the rise in [Ca2+]i from reticular stores. We showed that purinergic receptors are expressed in human keratinocytes in culture, since ATP (500 nM) induces a rise in [Ca2+]i via P2Y activation in NHK (Fig. 4c). This response is abolished in the presence of the unspecific purinergic receptor inhibitor suramin (300 lM). To confirm that the heptylamine-induced ATP release activates purinergic receptors, we showed that both suramin (300 lM) and PPADS (100 lM) reduced the rise in [Ca2+]i due to heptylamine. The F/F0 peak was lowered by 76% (1.83 ± 0.14 to 1.20 ± 0.08, n = 18) with suramin (Fig. 4d) and the F/F0 peak was lowered by 84% (1.75 ± 0.10 to 1.12 ± 0.07, n = 18) with PPADS (Fig. 4e). Reducing the extracellular ATP concentration by hydrolysis with apyrase decreased by 88% (1.73 ± 0.09 to 1.09 ± 0.06, n = 18) the rise in [Ca2+]i due to heptylamine, suggesting that the observed calcium increase is dependent on ATP release. 4. Discussion Our study suggests that monitoring [Ca2+]i in cultured NHK is a reliable method to assess the toxicity of soluble chemical compounds independently of their effects on cell viability. Measurements of cell viability (MTT reduction test) and membrane integrity (neutral red uptake or LDH release) are the most commonly used parameters to evaluate the irritant potential of chemical substances in vitro. The value of measuring cell integrity to assess irritant potential was questioned by the European Center for the Validation of Alternative Methods (ECVAM, Gerberick et al., 2008). The members agreed that although there is a correlation between irritant potential and reduced cell viability, other important conflicting factors exist, including the fact that many irritants do not alter cell viability and membrane integrity (Fentem et al., 2001). It was therefore concluded that additional biomarkers should be incorporated in further studies. 4.1. Intracellular calcium increase is a common step in the keratinocyte response to chemical irritants According to the different classes and chemical structures of substances with irritant potential, different pathways are presumed to be involved in chemical-induced skin irritation. However, all irritants systematically initiate a local inflammatory response and induce the release by keratinocytes of pro-inflammatory cytokines IL1a, IL6 and IL8, or mediators such as ATP and prostaglandins. Both IL1a and IL8 release are calcium dependent (Wilson et al., 1993; Cieszynski et al., 1999) and ATP is known to activate metabotropic purinergic receptors coupled to PLC and calcium signaling in keratinocytes (Burrell et al., 2003; Koizumi et al., 2004). These pathways interconnect in keratinocytes since ATP stimulates IL6 production via P2Y receptors (Yoshida et al., 2006; Inoue et al., 2007; Pastore et al., 2007). IL1a induces the expression of IL-8 (Coquette et al., 2003; Poumay et al., 2004) and activates the calcium dependent enzyme phospholipase A2, a key enzyme in prostaglandin production (Reilly and Green, 1999). Overall, it can be hypothesized that irritants would systematically lead to the

407

inflammatory response-associated mobilization of Ca2+ in keratinocytes. Interestingly, the calcium signal is activated regardless of whether or not the irritants disrupt the cell membrane. Those chemical irritants which do disrupt the cell membrane provoke the release of IL1a and ATP from the cytoplasm (Lee and Maibach, 1995), while those that do not still induce the release of ATP (Mizumoto et al., 2003) and interleukins (Poumay et al., 2004) as well as prostaglandins (Gay et al., 1992). An overview of studies performed with various in vitro skin equivalents found that the biomarkers commonly used are IL1a, IL8 and the prostaglandin PGE2 (Welss et al., 2004). The measurement of [Ca2+]i involves more than one mediator pathway and should therefore increase the probability of observing a reliable signal. Intracellular calcium concentration would appear therefore to be a potent biomarker to differentiate between different classes of substances with regards to their irritant capacity for keratinocytes. 4.2. Intracellular calcium mobilization differentiates skin irritants and non irritants In our study, the strong irritant compounds 2-4-hexadienal, piperazine, DMIPA, heptylamine and SLS induced [Ca2+]i increase in NHK when they were applied at around 20 mM. Acidic and alkaline pH led to the same effect. The chemicals were chosen primarily on the basis of unambiguous skin irritation classifications derived from the data included in the ECETOC database. The chemicals covered a range of chemical structures and were reported in these previous studies to induce severe erythema or slight eschar and edema in the Draize tests on rabbit (ECETOC, 1995; Bagley et al., 1996). The primary irritation index (index of erythema, eschar and edema) of these chemicals is higher than the scale 4 corresponding to severe erythema (beet redness) to slight eschar formation (injuries in depth) and severe edema formation (raised more than 1 mm and extending beyond area of exposure). In addition, inhalation of heptylamine (27 p.p.m.) has been shown to depress respiratory rate by 50% (Nielsen and Vinggaard, 1988), 2-4-hexadienal diluted in corn oil administered to rodents by gavage has shown carcinogenic activity (Natl. Toxicol. Program, 2003), and human exposure to piperazine (44 to 100 mg/kg) provokes chronic bronchitis, asthenia, EEG changes, itch and skin sensitization (E.U. Risk Assessment Report, 2005). The concentrations used in vivo in the Draize tests on rabbit are higher than those used in vitro in our tests. Heptylamine, 2-4hexadienal, piperazine, DMIPA were tested in vivo pure at 99%, SLS diluted at 50%, and KOH and HCl diluted at 5%. In our study the concentrations varied from 0.11% to 0.54% (see Table 1). We measured the minimal concentration required below the stratum corneum to induce a [Ca2+]i increase in keratinocytes. Solubility restrictions meant that the highest concentrations tested (20 or 40 mM) remained low and imposed a limit of our study to soluble compounds. However, our aim was to determine in a first instance whether [Ca2+]i can be used as a reliable biomarker of the keratinocyte reaction to chemical irritants. Further studies are now needed in which the monitoring of [Ca2+]i in reconstructed skin equivalent has been adapted to favor the testing of compounds with low water solubility and of finished formulations. Since the response linked to acute irritant contact dermatitis activates calcium dependent pathways, we hypothesize that in general skin irritants increase [Ca2+]i. Our results presented here seem to agree with this hypothesis since there were no false negatives. Nevertheless, the possibility that a weak or non irritant compound can activate a calcium dependent pathway and the fact that any chemical able to generate a calcium signal would introduce false positive information must be considered. We are unable to estimate the occurrence of false positives since none of the

408

M. Raoux et al. / Toxicology in Vitro 27 (2013) 402–408

seven weak or non irritant chemicals tested in our experiments induced an increase in [Ca2+]i. 4.3. Heptylamine increases intracellular calcium concentration by activating purinergic receptors To investigate further the cellular mechanisms by which a chemical irritant may increase [Ca2+]i, we performed a detailed analysis of the effect of the corrosive compound heptylamine. This amine is an amphiphilic molecule able to penetrate the cellular lipid bilayer without provoking membrane disruption (Mizumoto et al., 2003; Raoux et al., 2007). Based on the evidence that mechanical deformation causes ATP release from keratinocytes (Koizumi et al., 2004; Azorin et al., 2011), we hypothesized that heptylamine alters the membrane fluidity leading to the release of ATP. We observed that this release of ATP did not require a subsequent increase in [Ca2+]i. Our results suggest that the ATP acts in an autocrine manner in keratinocytes and initiates the increase in [Ca2+]i by activating purinergic receptors. Indeed, when the extracellular amount of ATP was reduced by apyrase, the increase in [Ca2+]i provoked by heptylamine was also reduced. Our study thus emphasizes the role of ATP as an important mediator of irritant contact dermatitis. 5. Conflict of interest The authors have no conflict of interest. Acknowledgments The manuscript was revised by AngloScribe, (Nîmes, France) for English language editing. The authors thank F. Maingret and Y. Roudaut for helpful discussions, N. Osorio, and F. Padilla for technical assistance and critical discussions. This work was supported by Grants from CNRS and Region Provence-Alpes-Côte d’Azur. References Azorin, N., Raoux, M., Delmas, P., Crest, M., 2011. ATP signaling is crucial for the human keratinocyte response to hyposmotic stimulation. Exp. Dermatol. 20, 401–417. Bagley, D.M., Gardner, J.R., Holland, G., Lewin, R.W., Reignier, J.F., Stringer, D.A., Walker, A.P., 1996. Skin irritation: reference chemicals data bank. Toxicol. In Vitro 10, 1–6. Burrell, H.E., Bowler, W.B., Gallagher, J.A., Sharpe, G.R., 2003. Human keratinocytes express multiple P2Y-receptors: evidence for functional P2Y1, P2Y2, and P2Y4 receptors. J. Invest. Dermatol. 120, 440–447. Chan, K., Truong, D., Shangari, N., O’Brien, P.J., 2005. Drug-induced mitochondrial toxicity. Expert Opin. Drug. Metab. Toxicol. 1, 655–669. Chew, A.L., Maibach, H.I., 2003. Occupational issues of irritant contact dermatitis. Int. Arch. Occup. Environ. Health 76, 339–346. Coquette, A., Berna, N., Vandenbosch, A., Rosdy, M., De Wever, B., Poumay, Y., 2003. Analysis of interleukin-1alpha (IL-1alpha) and interleukin-8 (IL-8) expression and release in in vitro reconstructed human epidermis for the prediction of in vivo skin irritation and/or sensitization. Toxicol. In Vitro 17, 311–321. Corsini, E., Galli, C.L., 1998. Cytokines and irritant contact dermatitis. Toxicol. Lett. 102–103, 277–282. Cieszynski, J.A., Qureshi, M.A., Taylor Jr., R.L., 1999. Calcium dependency of interleukin-1 secretion by a chicken macrophage cell line. Poult. Sci. 78, 70–74. Draize, J.H., Woodard, G., Calvery, H.O., 1944. Method for the study of irritation and toxicity of substances applied topically to the skin and mucuous membranes. J. Pharmacol. Exp. Ther. 82, 377–390. ECETOC, 1995, Skin irritation and corrosion: reference chemicals data bank. Technical report n 66. European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels, 1–247. E.U. Risk Assessment Report, 2005, Piperazine, summary risk assessment report, CAS No: 110-85-0, Brussels, 1–25.

Faller, C., Bracher, M., 2002. Reconstructed skin kits: reproducibility of cutaneous irritancy testing. Skin Pharmacol. Appl. Skin Physiol. 15 (Suppl. 1), 74–91. Fentem, J.H., Briggs, D., Chesné, C., Elliott, G.R., Harbell, J.W., Heylings, J.R., Portes, P., Roguet, R., van de Sandt, J.J., Botham, P.A., 2001. A prevalidation study on in vitro tests for acute skin irritation: results and evaluation by the management team. Toxicol. In Vitro 15, 57–93. Gay, R., Swiderek, M., Nelson, D., Ernesti, A., 1992. The living skin equivalent as a model in vitro for ranking the toxic potential of dermal irritants. Toxicol. In Vitro 6, 303–315. Gerberick, F., Aleksic, M., Basketter, D., Casati, S., Karlberg, A.T., Kern, P., Kimber, I., Lepoittevin, J.P., Natsch, A., Ovigne, J.M., Rovida, C., Sakaguchi, H., Schultz, T., 2008. Chemical reactivity measurement and the predicitve identification of skin sensitisers. The report and recommendations of ECVAM workshop 64. Altern. Lab. Anim. 36, 215–242. Gibbs, S., 2009. In vitro irritation models and immune reactions. Skin Pharmacol. Physiol. 22, 103–113. Inoue, K., Hosoi, J., Denda, M., 2007. Extracellular ATP has stimulatory effects on the expression and release of IL-6 via purinergic receptors in normal human epidermal keratinocytes. J. Invest. Dermatol. 127, 362–371. Koizumi, S., Fujishita, K., Inoue, K., Shigemoto-Mogami, Y., Tsuda, M., Inoue, K., 2004. Ca2+ waves in keratinocytes are transmitted to sensory neurons: the involvement of extracellular ATP and P2Y2 receptor activation. Biochem. J. 380, 329–338. Korting, H.C., Schindler, S., Hartinger, A., Kerscher, M., Angerpointner, T., Maibach, H.I., 1994. MTT-assay and neutral red release (NRR)-assay: relative role in the prediction of the irritancy potential of surfactants. Life Sci. 55, 533–540. Lee, C.H., Maibach, H.I., 1995. The sodium lauryl sulfate model: an overview. Contact Dermatitis 33, 1–7. Macfarlane, M., Jones, P., Goebel, C., Dufour, E., Rowland, J., Araki, D., CostabelFarkas, M., Hewitt, N.J., Hibatallah, J., Kirst, A., McNamee, P., Schellauf, F., Scheel, J., 2009. A tiered approach to the use of alternatives to animal testing for the safety assessment of cosmetics: skin irritation. Regul. Toxicol. Pharmacol. 54, 188–196. Mizumoto, N., Mummert, M.E., Shalhevet, D., Takashima, A., 2003. Keratinocyte ATP release assay for testing skin-irritating potentials of structurally diverse chemicals. J. Invest. Dermatol. 121, 1066–1072. Natl. Toxicol. Program Tech. Rep. Ser., 2003. Toxicology and carcinogenesis studies of 2,4-hexadienal. CAS n 142-83-6, 1–290. Nielsen, G.D., Vinggaard, A.M., 1988. Sensory irritation and pulmonary irritation of C3–C7 n-alkylamines: mechanisms of receptor activation. Pharmacol. Toxicol. 63, 293–304. Osborne, R., Perkins, M.A., 1994. An approach for development of alternative test methods based on mechanisms of skin irritation. Food Chem. Toxicol. 32, 133– 142. Pastore, S., Mascia, F., Gulinelli, S., Forchap, S., Dattilo, C., Adinolfi, E., Girolomoni, G., Di Virgilio, F., Ferrari, D., 2007. Stimulation of purinergic receptors modulates chemokine expression in human keratinocytes. J. Invest. Dermatol. 127, 660– 667. Poumay, Y., Dupont, F., Marcoux, S., Leclercq-Smekens, M., Hérin, M., Coquette, A., 2004. A simple reconstructed human epidermis: preparation of the culture model and utilization in in vitro studies. Arch. Dermatol. Res. 296, 203–211. Raoux, M., Colomban, C., Delmas, P., Crest, M., 2007. The amine-containing cutaneous irritant heptylamine inhibits the volume-regulated anion channel and mobilizes intracellular calcium in normal human epidermal keratinocytes. Mol. Pharmacol. 71, 1685–1694. Reilly, D.M., Green, M.R., 1999. Eicosanoid and cytokine levels in acute skin irritation in response to tape stripping and capsaicin. Acta Derm. Venereol. 79, 187–190. Triglia, D., Braa, S.S., Yonan, C., Naughton, G.K., 1991. In vitro toxicity of various classes of test agents using the neutral red assay on a human three-dimensional physiologic skin model. In Vitro Cell. Dev. Biol. 27A, 239–244. Welss, T., Basketter, D.A., Schroder, K.R., 2004. In vitro skin irritation: facts and future. State of the art, review of mechanisms and models. Toxicol. In Vitro 18, 231–243. Wilson, L., Butcher, C.J., Kellie, S., 1993. Calcium ionophore A23187 induces interleukin-8 gene expression and protein secretion in human monocytic cells. FEBS Lett. 325, 295–298. York, M., Griffiths, H.A., Whittle, E., Basketter, D.A., 1996. Evaluation of a human patch test for the identification and classification of skin irritation potential. Contact Dermatitis 34, 204–212. Yoshida, H., Kobayashi, D., Ohkubo, S., Nakahata, N., 2006. ATP stimulates interleukin-6 production via P2Y receptors in human HaCaT keratinocytes. Eur. J. Pharmacol. 540, 1–9. Zholos, A., Beck, B., Sydorenko, V., Lemonnier, L., Bordat, P., Prevarskaya, N., Skryma, R., 2005. Ca2+- and volume-sensitive chloride currents are differentially regulated by agonists and store-operated Ca2+ entry. J. Gen. Physiol. 125, 197–211.