Toxicological investigation of lilial

Abstract

Lilial (also called lysmeral) is a fragrance ingredient presented in many everyday cosmetics and household products. The concentrations of lilial in the final products is rather low. Its maximum concentration in cosmetics was limited and recently, its use in cosmetics products was prohibited in the EU due to the classification as reproductive toxicant. Additionally, according to the European Chemicals Agency, it was under assessment as one of the potential endocrine disruptors, i.e. a substance that may alter the function of the endocrine system and, as a result, cause health problems. Its ability to act as an androgen receptor agonist and the estrogenic and androgenic activity of its metabolites, to the best of our knowledge, have not yet been tested. The aim of this work was to determine the intestinal absorption, cytotoxicity, nephrotoxicity, mutagenicity, activation of cellular stress-related signal pathways and, most importantly, to test the ability to disrupt the endocrine system of lilial and its Phase I metabolites. This was tested using set of in vitro assays including resazurin assay, the CHO/HPRT mutation assay, γH2AX biomarker-based genotoxicity assay, qPCR and in vitro reporter assays based on luminescence of luciferase for estrogen, androgen, NF-κB and NRF2 signalling pathway. It was determined that neither lilial nor its metabolites have a negative effect on cell viability in the concentration range from 1 nM to 100 µM. Using human cell lines HeLa9903 and MDA-kb2, it was verified that this substance did not have agonistic activity towards estrogen or androgen receptor, respectively. Lilial metabolites, generated by incubation with the rat liver S9 fraction, did not show the ability to bind to estrogen or androgen receptors. Neither lilial nor its metabolites showed a nephrotoxic effect on human renal tubular cells (RPTEC/TERT1 line) and at the same time they were unable to activate the NF-κB and NRF2 signalling pathway at a concentration of 50 µM (HEK 293/pGL4.32 or pGL4.37). Neither lilial nor its metabolites showed mutagenic activity in the HPRT gene mutation test in CHO-K1 cells, nor were they able to cause double-strand breaks in DNA (γH2AX biomarker) in CHO-K1 and HeLa cells. In our study, no negative effects of lilial or its in vitro metabolites were observed up to 100 µM using different in vitro tests.

Figure 1:

3-(4-(tert-butyl)phenyl)-2-methylpropanal (ChemDraw 16.0).

Figure 2

Retention and mass spectra of GC/MS analysis of lilial (A) and its metabolites (B). (C) spectrum of lilial, (D) spectrum of lysmerylic acid.

Figure 3

Relative viability of the HeLa9903 cells (resazurin assay) after 1 day of incubation with lilial at eight concentrations. The final concentration of DMSO was 0.1%, this vehicle control (VC) was taken as 100% viability. Error bars represents for the standard error deviation of six replicates (wells). Statistically evaluated using t-test (* indicates statistically significant difference from VC at p < 0.05, **p < 0.005, ***p < 0.0005).

Figure 4

Relative viability of the CHO-K1 cells (resazurin assay) after 1 day of incubation with lilial at seven concentrations without or with metabolic activation using the S9 mix. The final concentration of DMSO was 0.1%, this vehicle control (VC) was taken as 100% viability. 0.25% Tween-20 served as a positive control (PC). Error bars represents for the standard error deviation of six replicates (wells). Statistically evaluated using t-test (*indicates statistically significant difference from corresponding VC at p < 0.05, **p < 0.005, ***p < 0.0005).

Figure 5

Intensity of response expressed as fold induction (FI) of HeLa9903 cells after 22 h incubation with E2 and lilial. (NC-MEM with 10% DCC-FBS, VC—0.1% DMSO in MEM with 10% DCC-FBS). Error bars indicate the standard error of the mean. Statistically evaluated using ANOVA (R program). *** indicates statistically significant difference from VC at p < 0.001, ** difference from VC at 0.001 < p < 0.01.

Figure 6

Intensity of response expressed as fold induction (FI) of MDA-kb2 cells after 22 h incubation with DHT and lilial. (NC-RPMI with 10% DCC-FBS, VC—0.1% DMSO in RPMI with 10% DCC-FBS). Error bars indicate the standard error of the mean. Statistically evaluated using ANOVA (R program). *** indicates statistically significant difference from VC at p < 0.001.

Figure 7

Induction of NF-kB pathway expressed as fold induction of transfected HEK293 cells treated with TNFα (positive control), lilial, and lilial activated with the S9 (both 100 µM). (NC—untreated cells, VC—DMSO). Error bars indicate the standard error of the mean. Statistically evaluated using t-test (*indicates statistically significant difference from NC at p < 0.05, **p < 0.005, ***p < 0.0005).

Figure 8

Induction of NRF2-ARE pathway expressed as fold induction of transfected HEK293 cells treated with tert-butylhydroquinone (TBHQ, positive control), lilial, and lilial activated with the S9 (both 100 µM). (NC—untreated cells, VC—DMSO). Error bars indicate the standard error of the mean. Statistically evaluated using t-test (*indicates statistically significant difference from NC at p < 0.05, **p < 0.005, ***p < 0.0005).