Skip to main content
NCBI home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2019 Aug 8;104(11):5393–5405. doi: 10.1210/jc.2018-01880

Lavender Products Associated With Premature Thelarche and Prepubertal Gynecomastia: Case Reports and Endocrine-Disrupting Chemical Activities

J Tyler Ramsey 1,2, Yin Li 1, Yukitomo Arao 1, Ajanta Naidu 3, Laurel A Coons 1, Alejandro Diaz 4, Kenneth S Korach 1,✉
PMCID: PMC6773459  PMID: 31393563

Abstract

Context

Previous case reports associated prepubertal gynecomastia with lavender-containing fragrances, but there appear to be no reports of premature thelarche.

Objective

To add to a case series about lavender-fragranced product use and breast growth in children and to measure endocrine-disrupting chemical activity of essential oil components.

Design, Setting, and Patients

Patients experiencing premature thelarche or prepubertal gynecomastia with continuous exposure to lavender-fragranced products were evaluated in the pediatric endocrinology departments of two institutions. Mechanistic in vitro experiments using eight components of lavender and other essential oils were performed at National Institute of Environmental Health Sciences.

Main Outcome Measures

Case reports and in vitro estrogen and androgen receptor gene expression activities in human cell lines with essential oils.

Results

Three prepubertal girls and one boy with clinical evidence of estrogenic action and a history of continuous exposure to lavender-containing fragrances were studied. Breast growth dissipated in all patients with discontinuation of the fragranced products. Some of the components tested elicited estrogenic and antiandrogenic properties of varying degrees.

Conclusion

We report cases of premature thelarche that resolved upon cessation of lavender-containing fragrance exposure commonly used in Hispanic communities. The precise developmental basis for such conditions could be multifactorial. In vitro demonstration of estrogenic and antiandrogenic properties of essential oil components suggests essential oils in these cases could be considered a possible source and supports a possible link with idiopathic prepubertal breast development. Whether the level of lavender oil estrogenic potency is sufficient to cause these effects is unknown.

Lavender-containing fragrances may cause premature thelarche and prepubertal gynecomastia.

Mammary gland breast growth has been documented as an estrogen receptor α (ERα)-dependent response in experimental models (1) and clinical cases (2). Although breast tissue primarily is thought of as a female-specific end point, men also possess breast tissue (3). Premature thelarche, defined as isolated breast development without other signs of puberty before 8 years of age, is the most common pubertal disorder in girls (4). The causes of this condition remain unclear (5). On the other hand, transient thelarche has been studied longitudinally no earlier than from 8 years of age (6). Gynecomastia is a common clinical condition that involves benign development of breast tissue in men (3), but prepubertal gynecomastia is a relatively rare condition (7, 8) and most cases are of unknown cause (3).

Gynecomastia is suspected to have many etiologies. Selected drugs and environmental exposures such as alcohol, heroin, marijuana, amphetamines, antiulcer medications, antibiotics (e.g., isoniazid, ketoconazole, metronidazole), cancer agents (e.g., alkylating drugs), cardiovascular drugs (e.g., amiodarone, captoptil, digitoxin), and psychoactive drugs (e.g., diazepam, haloperidol, phenothiazines, tricyclic antidepressants) have been identified as possible hormonal mimics for the estrogen and androgen receptors (3). The mechanism by which those drugs disrupt the endocrine system are less well defined but could also involve altering steroidogenesis, with a resultant change in the balance between testosterone and estradiol (E2) levels, increasing proliferation of breast tissue and leading to the onset of gynecomastia (3, 8). Previous clinical cases have established a link between hygiene-product use of lavender oil (LO) and tea tree oil (TTO) and prepubertal gynecomastia in boys (8). However, to our knowledge, there are no confirmed cases of such a link in prepubertal girls nor identification of the active components.

LO is the most widely used essential oil by both men and women (9). Some of the most important constituents in LO are linalyl acetate (LinA) and linalool (Lin) (10). TTO is the third most popular essential oil used today (9). Some of the most important constituents in TTO are α-terpinene (α-Terp) and γ-terpinene (γ-Terp) (11). Eucalyptol (Euc), 4-terpineol (4-Terp), dipentene/limonene (Di/Lim), and α-terpineol (α-Terpl) are common to both LO and TTO. All eight chemicals are mandated by the International Organization for Standardization (ISO) to be included in either LO or TTO (11, 12).

ERα plays a crucial role in mammary gland development. This finding was confirmed from the phenotypes of aromatase knockout mice that lack endogenous estrogen production and ERα knockout mice that lack functional ERα. Both models exhibit impaired mammary gland development (13), which supports the view that estrogen-dependent, ERα-mediated actions are critical for mammary gland development (14–17).

The lack of breast development observed in mice can be extended to patients with clinical cases of aromatase deficiency or ERα mutations (2, 18). In another example, an aromatase transgenic ERα knockout mouse had significantly impaired mammary gland growth even when high levels of endogenous estrogen were present in the tissue (17), thus indicating that ERα-mediated mechanisms are important in mammary gland development as well as stimulation of mammary hyperplasia in aromatase transgenic mice (14–17). Even more interesting, in breast tissue, estrogens including the endogenous hormone E2 regulate growth, cellular differentiation, and physiological functions through ERα (19–21). The receptor belongs to the nuclear receptor (NR) superfamily of ligand-inducible transcription factors (22). The receptor exhibits distinct tissue-specific expression patterns and biological roles (23). The androgen receptor (AR) is also present in normal breast tissue. The AR is a ligand-dependent nuclear transcription factor and a member of the steroid nuclear receptor family. Transcriptional coregulators are the principal factors influencing gene expression in which they directly interact with and modulate the activity of almost all NRs and transcription factors (24, 25). Steroid receptor coactivators (SRCs) were the first of the gene families to be classified as coregulators for NRs. All members can effectively enhance transcriptional activity of NRs by acting as bridging molecules and assist with chromatin modifications involving protein-protein interactions between NRs (24, 25).

An endocrine-disrupting chemical (EDC) is an exogenous chemical, or mixture of chemicals, that can interfere with any aspect of hormone action (26). Some EDCs act through nuclear hormone receptors (27), and LO and TTO have been reported to potentially act as an EDC in prepubertal boys (8). However, the mechanisms by which these essential oil components act remain unclear. In this study, we report three cases of premature thelarche associated with exposure to essential oils and a case of prepubertal gynecomastia. We studied the mechanism of action of LO, TTO, and eight components that are mandated to be included in essential oil products, to determine their abilities to mimic estrogen and androgen.

Materials and Methods

Study design

Case studies were performed at Nicklaus Children's Hospital and the Pediatric Endocrinology Clinic at the University of California, Irvine. All participants were selected on the basis of eligibility. To be eligible for this study, patients had to be prepubertal and continuously exposed to an essential-oil product. No institutional review board approval was needed to report these case studies. All participants were informed that information regarding their case potentially would be compiled for publication. Consent of the patient and their parents is recorded in each patient’s medical chart. All mechanistic in vitro experiments were performed at the National Institute of Environmental Health Sciences.

In vitro studies

We used a variety of human cell models to characterize the estrogenic and antiandrogenic activities of the essential oil components. All models were performed at least three times and all data reported from the experiments are averages of at least three data sets with the exception of the reporter assays, which are a representation of one data set from three others performed.

Plasmids

The expression vector pcDNA3 was purchased from Invitrogen. An internal control plasmid for transfection efficiency, Renilla luciferase, was purchased from Promega. The luciferase reporter 3xERE Luc (synthetic vitellogenin ERE-TATA fused to a luciferase reporter gene) have been described previously (28). The pcDNA/ERα and pcDNA/SRC-2 plasmid have been reported previously (25, 28).

Cell culture

The human breast cancer MCF-7 cell line (ERα positive) and human hepatocellular cancer HepG2 cell line (ERα negative) were purchased from American Type Culture Collection. The human breast cancer MDA-kb2 cell line (AR positive), stably expressing the MMTV promoter, has been described previously (29). MCF-7 cells were maintained in phenol red-free DMEM:F12 medium, HepG2 cells were maintained in MEM medium, and MDA-kb2 cells were maintained in Leibovitz’s medium (Invitrogen). All media were supplemented with 10% fetal bovine serum (FBS; Gemini Bio Products) and 4 mM l-glutamine (Invitrogen). Charcoal/Dextran stripped fetal bovine serum (sFBS, HyClone; Gemini Bio Products) was substituted for FBS during treatment.

Chemicals and components

The E2, flutamide (Flut), LO (CAS no. 8000-28-0; product no. 61718), TTO (CAS no. 68647-73-4; product no. W390208), Euc, 4-terpineol (4-Terp), dipentene/limonene (Di/Lim), α-Terpl, LinA, Lin, α-Terp, γ-Terp, corn oil, and soybean oil were purchased from Sigma-Aldrich. Testosterone was purchased from Steraloids and fulvestrant/ICI 182,780 (ICI) was purchased from Tocris Bioscience. All essential oils and their components were naturally extracted from their source according to Sigma-Aldrich protocols. A mixture of the components was used in similar percentages mandated by the ISO to be included in LO and TTO. These are represented as LO components (LO-Cs) and TTO components (TTO-Cs) and were compared with LO and TTO for estrogenic activity. Details concerning the selection of the eight components and the mixture percentages are listed in Table 1. All components and oils used in this experiment were diluted in dimethyl sulfoxide (DMSO) and the vehicle control contained a final concentration of DMSO not exceeding 0.1%.

Table 1.

Selected Components From LO and TTO

Component CAS RN % Purity Essential Oil % LO-C % TTO-C Sigma-Aldrich Catalog No.
Euc 470-82-6 ≥99 Both 2.8 13.2 W246506- Sample-K
4-Terp 562-74-3 ≥95 Both 7.5 42.3 W224847-Sample
Di/Lim 138-86-3 Mix of isomers Both 1 1.3 W524905-Sample-K
α-Terpl 10482-56-1 ≥96 Both 1.9 7.1 W304522- Sample-K
LinA 115-95-7 ≥97 LO 44.3 — W263605- Sample-K
Lin 78-70-6 ≥97 LO 42.5 — W263508-Sample-K
α-Terp 99-86-5 ≥89 TTO — 11.4 W355801-Sample-K
γ-Terp 99-85-4 ≥95 TTO — 24.7 W355909-Sample-K
Open in a new tab

Abbreviation: CAS RN, Chemical Abstracts Service registry number.

Selection of components

LO and TTO each contains many diverse chemicals that vary in structure and composition (10–12). Predicting whether a substance could potentially be an EDC is difficult, because these components are diverse and may not appear to share any structural similarity other than having small molecular masses (27). Stereochemistry was taken into consideration to observe if any biological changes occurred between isomers. A variety of compounds were chosen on the basis of chemical structures. These parameters included if the compounds contained rings or no rings, single bonds or double bonds, or if isomers of the same compound made a difference in biological activity . Knowing information about stereochemistry may be useful in predicting how other chemicals in essential oils may behave that were not tested.

The ISO mandates that 13 and 14 components must be in certain percentage ranges to be classified as LO or TTO, respectively. Common chemicals observed between LO and TTO are Euc, 4-Terp, Di/Lim, and α-Terpl (11, 12, 30–32). Chemicals specific to either oil were also chosen. LO by concentration can contain ≤92% LinA and Lin by volume (33), and these two components are most likely what gives LO its specific smell. To further justify choosing these two chemicals, literature shows that experiments have been performed previously with the chemical Lin (34). The two chemicals selected specific to TTO were α-Terp and γ-Terp, which are also found in relatively high percentage in TTO (11).

Luciferase reporter assays with ERα

HepG2 cells were seeded in MEM:F12 plus 10% FBS at 1.5 × 105 to 2.0 × 105 cells per well in 24-well plates overnight. The medium was changed to MEM plus 10% sFBS the next day before transient transfection using the Effectene transfection reagent (Qiagen). A total of 0.5 μg/well (3xERE luc, human ERα, and pRL-TK plasmids) or 0.7 μg/well (SRC-2, vector pcDNA3, 3xERE luc, human ERα, and pRL-TK plasmids) were transfected into the cells. After 6 hours, the cells were changed to fresh MEM:F12 plus 10% sFBS medium overnight and were treated with vehicle control (DMSO); 1 nM E2; 0.005%, 0.01%, or 0.025% (v/v) LO, LO-C, TTO, or TTO-C; or 100, 500, or 1000 µM Euc, 4-Terp, Di/Lim, α-Terpl, LinA, Lin, α-Terp, or γ-Terp for 18 hours.

Luciferase reporter assays were performed using the Dual Luciferase Reporter Activity System (Promega) and an LMAX II luminometer (Molecular Devices). Transfection efficiency was normalized using the Renilla luciferase pRL-TK plasmid and fold changes were calculated relative to the vehicle control. Calculations for appropriate concentrations to treat the cells were extrapolated from our previous report (8) and from the data repository (35). Relative percentages of the individual components in reference to LO and TTO were provided by the ISO. These percentages were back calculated to find a range of molarity in which they could be hormonally active.

Two-hybrid mammalian ERα ligand binding affinity assay with SRC2

To evaluate the binding activity of the oil components to the ERα ligand-binding domain (LBD), we performed a mammalian two-hybrid assay, which analyzes the ligand dependency of hERα LBD and SRC-NR element interactions (36).We selected SRC-2-NR for this assay because overexpression of SRC-2 in HepG2 cells enhanced ERα-mediated transcription activity for all chemicals, which is shown in Fig. 1B. A total of 0.3 μg/well [Gal4-DBD-fused SRC-NR-box (pM-SRC-NR), VP16-AD (pACT), or VP16-AD-fused-ERα WT-LBD (pACT-LBD/WT), and pRL-TK plasmids] were transfected into the cells (37). After 6 hours, the cells were changed to fresh MEM plus 10% sFBS medium overnight and were treated with vehicle control (DMSO); 1 nM E2; 0.025% (v/v) LO, LO-C, TTO, or TTO-C; or 500 µM Euc, 4-Terp, Di/Lim, α-Terpl, LinA, Lin, α-Terp, or γ-Terp for 18 hours. Luciferase assays were performed using the Dual Luciferase Reporter Assay System. Transfection efficiency was normalized using the Renilla luciferase pRL-TK plasmid and fold changes were calculated relative to the vehicle control. Experimental transfection conditions can be found in the data repository (35).

Figure 1.

Figure 1.

Open in a new tab

ERα ERE-mediated transcriptional activity, coactivator SRC-2 recruitment, binding to the ERα LBD, and ERα-regulated gene expression. (A) Dose-dependent estrogenic responses of HepG2 cells. The cells were transfected with ERE-luciferase and human ERα expression plasmid [plasmid cloning DNA (pcDNA)/ERα]. A Dunnett multiple comparison test was performed in comparison with the vehicle control. All experiments were performed at least three times. Results are a representation of one experiment and are presented as mean ± SEM. (B) SRC-2 transcriptional enhancement of HepG2 cells. The cells were transfected with human pcDNA/ERα plasmids, with or without pcDNA/SRC-2. Data were normalized against each vehicle group. A Tukey multiple comparison test was performed in comparison with ERα and SRC-2 plus ERα treatment groups. All experiments were performed at least three times. Results are an average of the experiments and are presented as mean ± SEM. (C) Binding to the ERα-LBD in HepG2 cells. The cells were cotransfected with pG5-luciferase reporter plasmid and the expression plasmids for the Gal4-DBD-fused SRC-2-NR-box (pM-SRC-2-NR) in the presence of VP16-AD (pACT) or VP16-AD-fused-ERα WT-LBD. Data were normalized against each vehicle group and its respected pACT treatment group. A Dunnett multiple comparison test was performed in comparison with vehicle control ERα-LBD. All experiments were performed at least three times. Results are an average of the experiments and are presented as mean ± SEM. (D) Induction of ERα endogenous genes and inhibition of induction when cotreated with the antagonist ICI in MCF-7 cells. Total RNA was extracted and mRNA levels of GREB1, PGR, and CTSD genes were quantified by quantitative PCR analysis. All gene fold changes were normalized against ACTB. Results are presented as mean ± SEM. “a” Denotes statistical significance in comparison with vehicle control, P < 0.05; and “b” denotes statistical significance in comparison against each treatment group, with or without the addition of ICI, P < 0.05. (A–C) *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Luc, luciferase.

Treatment, RNA isolation, and quantitative PCR analysis

MCF-7 (ERα-positive) and MDA-kb2 cells (AR-positive) were selected to measure gene changes. For ERα target genes, MCF-7 cells were seeded in six-well plates at 500,000 cells per well in DMEM:F12 plus 10% FBS medium overnight. The cells were then starved in phenol red-free DMEM:F12 plus 5% sFBS medium for 36 hours and treated with vehicle control (DMSO), 1 nM E2, 0.025% (v/v) LO or TTO, or 1000 μM Euc, 4-Terp, Di/Lim, α-Terpl, LinA, Lin, α-Terp, or γ-Terp for 18 hours, with or without 5 μM ICI. For AR target genes, MDA-kb2 cells were seeded in six-well plates at 750,00 cells per well in phenol red-free Leibovitz’s medium plus 10% FBS medium. The cells were then starved in phenol red-free Leibovitz’s medium plus 5% sFBS medium for 24 hours and treated with vehicle control (DMSO), 1 nM T, and 1 nM testosterone plus 1 μM Flut for controls. Cells also were cotreated with 1nM testosterone and 0.025% (v/v) LO or TTO, or 100 μM Euc, 4-Terp, Di/Lim, α-Terpl, LinA, Lin, α-Terp, or γ-Terp for 18 hours.

Total RNA was extracted from MCF-7 and MDA-kb2 cells using the RNeasy Mini Kit (Qiagen). First-strand cDNA synthesis was performed using Superscript reverse transcription according to the manufacturer’s protocol (Invitrogen). The mRNA levels of ERα target genes [Growth regulation by estrogen in breast cancer 1 (GREB1), progesterone receptor (PGR), and cathepsin D (CTSD)] and AR target genes [cytochrome P450 family 4 subfamily F member 8 (CYP4F8), UDP glucuronosyltransferase family 2 member B28 (UGT2B28), and SEC14-like lipid binding 2 (SEC14L2) were measured using SYBR green assays (Applied Biosystems). Cycle threshold values were obtained using the ABI PRISM 7900 Sequence Detection System and analysis software (Applied Biosystems). Each sample was normalized to human β-Actin expression and fold changes were calculated relative to the vehicle control cycle control values. The sequences of quantitative PCR primers used in this study are shown in the data repository (35).

Luciferase reporter assays with AR

To examine transcriptional changes of AR, we investigated promoter activation using the mouse mammary tumor virus (MMTV), which contains an androgen response element, stably transformed in the AR-positive MDA-kb2 cell line. Calculations for appropriate dosing were extrapolated from our previous report (8) and from calculations in the data repository (35). Multiple dose curves were performed to find a range in which these components activated the promotor MMTV. MDA-kb2 cells were seeded in 48-well plates at 100,000 cells per well in phenol red-free Leibovitz’s medium plus 10% FBS overnight. The medium was changed to Leibovitz’s medium 5% sFBS after 6 hours and starved for a total of 24 hours. Cells were treated with the vehicle, 1 nM T, and 1 nM testosterone plus 1 μM Flut for controls. To test the essential oils and their components, cells were cotreated with 1 nM testosterone and 0.01% (v/v) LO or TTO, or 0.1,1, 10, 100, and 1000 μM Euc, 4-Terp, Di/Lim, α-Terpl, LinA, Lin, α-Terp, or γ-Terp for 18 hours.

Statistical analysis

The data were analyzed using Tukey and Dunnett multiple comparison tests in GraphPad prism, version 7.0 (GraphPad Software, San Diego, CA). All statistical tests were considered significant at P < 0.05. Any outliers that were identified were excluded using a Grubb test in GraphPad, version 7.0.

Results

Case reports

A total of 24 patients with prepubertal gynecomastia or premature thelarche have been seen at the Division of Endocrinology at Nicklaus Children’s Hospital over the past 7 years. Five girls with transient thelarche were exposed to lavender-containing products, and breast development resolved after removing lavender exposure. Eleven of the 19 boys with prepubertal gynecomastia were exposed to products (e.g., colognes, shampoos, or soaps) containing lavender fragrance. Of the 16 patients exposed to lavender, 14 were Hispanic, one was Haitian, and one was white. The etiology of prepubertal gynecomastia in the eight other boys was as follows: The causes in three cases was undetermined (idiopathic), three boys had aromatase excess determined by elevated serum estrogen levels, and two were fed with soy milk for several years. Two of the patients with aromatase excess were white and one was of Arabic decent.

Four female patients have been seen at the Pediatric Endocrinology Division at the University of California, Irvine, and Children’s Hospital of Orange County’s clinic with premature breast development. Of the four patients, three were white and one was Middle Eastern. All patients had used lavender-fragrance products (e.g., soaps, lotions, essential oils), but then stopped after visitation to the clinic. Two patients’ conditions resolved since stopping use and the other two patients have developed symptoms of precocious puberty.

Patient 1

A girl 7 years and 6 months old presented for additional evaluation after initial observation of a left breast bud at age 6 years. Her past medical history was unremarkable, and she was taking no medication. Her family history was noncontributory. She was of normal height and weight for her age. Breast tissue growth was observed by physical examination on the left side measuring 4 cm × 3 cm in size and no breast tissue growth on the right side. Pubic hair was Tanner stage 1 and her clitoris was normal. Laboratory tests showed the following prepubertal gonadotropin levels: LH, 0.01 mIU/mL; FSH, 0.319 mIU/mL; E2, <1 pg/mL. Thyroid function tests were normal. Bone age corresponded to her chronological age. On further questioning, the family noted the patient’s frequent exposure to lavender oil in the form of a cologne named “Mi Tesoro Agua de Violetas” since early childhood. Breast development resolved 6 months after discontinuing use of the cologne with no recurrence upon follow-up.

Patient 2

A girl 3 years and 11 months old had right-side breast development at age 1 year that was observed by her mother. The patient complained of 2 months of breast tenderness at 3 years of age and was evaluated by a pediatrician. Her mother reported that she had been bathed with a soap containing lavender oil named “Baby Magic Calming Baby Bath Lavender and Chamomile” since infancy. Her past medical history was unremarkable, and she was taking no medications. Upon physical examination, she had breast tissue growth on the right side measuring 3 cm × 3 cm. Her pubic hair was Tanner stage 1 and her clitoris was normal. A breast ultrasound confirmed the presence of breast ductal development measuring 3.28 cm × 2.8 cm in size and 1.12 cm in depth. Her bone age was within the normal range for her chronological age. Total resolution of her breast tissue was noted 6 months after discontinuation of the soap containing lavender oil.

Patient 3

A girl 7 years and 9 months old had a left breast bud without other signs of puberty. She informed the physician that she sat near a teacher’s table, which had a lavender-oil diffuser running all day and was exposed to lavender oil for 1 year before noticing breast development. Her height and weight were normal for her age. Her physical examination was positive for left subareolar breast tissue, Tanner stage 2; and pubic hair of Tanner stage 1. Pelvic ultrasound showed a uterus that was enlarged to the peripubertal size with a thin endometrial stripe of 0.07 cm and prepubertal ovaries. Her bone age corresponded to her chronological age. Exposure to lavender oil was discontinued and her breast tissue completely regressed upon evaluation 3 months later.

Patient 4

A boy 7 years and 11 months old had breast enlargement at 4 years of age. He was taking no medication. Since infancy, he was exposed daily to lavender oil named “Crusellas Violet Water cologne.” His past medical history was positive for attention deficit and speech delay, but he required no medication. His height and weight were above the percentiles that corresponded to his age; however, his body mass index was at the 95th percentile for age and sex. Upon physical examination, he demonstrated bilateral breast development measuring 4 × 4 cm in size and 3 cm in depth, and bilateral testicular volume of 1 mL. He had a previous chromosomal microarray, which was normal. His gonadotropin levels were prepubertal (LH, 0.01 mIU/mL; FSH 0.67 mIU/mL) and testosterone was 4.8 ng/dL. His level of estrogens (E2, 1.1 pg/mL; estrone, 20 pg/mL), 17-hydroxyprogesterone (77 ng/dL), and prolactin (12 ng/mL) were all normal for his age and sex. His bone age was within the normal range for his chronological age. Resolution of his breast tissue was noted 6 months after discontinuation of the cologne and there was no further recurrence. Of note, these children were not genetically related to each other.

ERα estrogen response element–mediated activity

To determine if there were direct dose-dependent estrogenic responses of LO and TTO components, we transfected HepG2 cells with human ERα and three copies of estrogen response elements (EREs) to investigate ERα ERE-mediated promotor activity. When the cells were treated with 1 nM E2, almost a 50-fold ERα ERE-mediated promotor activity was observed when compared with nontreated cells (35). LO and TTO showed significant activation at all concentrations (0.005%, 0.01%, and 0.025%), with a maximum at ∼20-fold ERα ERE-mediated promotor activity increase for both oils (35). To test the activities of LO and TTO components for comparison with natural LO and TTO, components were reconstituted and mixed (designated as LO-C and TTO-C). Significant activation of LO-C and TTO-C was observed at all concentrations, except 0.005% TTO-C. The highest fold changes observed for these mixtures were approximately 10-fold, with TTO-C being the greatest (35).

When comparing the individual oil components that are common between LO and TTO, α-Terpl had the highest ERα ERE-mediated promotor activity increase in activation at all concentrations, with approximately sixfold increase at 500 μM and greater than an eightfold increase at 1000 μM (Fig. 1A). The second most potent compound was 4-Terp, which had a twofold increase at 100 μM, a fourfold increase at 500 μM, and an almost eightfold increase at 1000 μM (Fig. 1A). Di/Lim showed no activation at any concentrations except with almost a twofold increase at 1000 μM (Fig. 1A). Euc was inactive and showed no significant activation at any concentration (Fig. 1A). Of the components specific to LO, Lin was the most potent at all concentrations, with more than a twofold increase at 100 μM, a sixfold increase at 500 μM, and an almost eightfold increase at 1000 μM. LinA showed a modest twofold increase at 500 μM and fourfold increase at 1000 μM, but LinA showed no activation at 100 μM (Fig. 1A). TTO components, α-Terp and γ-Terp, showed no significant induction at any concentrations except for γ-Terp , which was increased threefold at 1000 μM (Fig. 1A). These findings are summarized in Table 2 and indicate that some, but not all, individual components in LO and TTO showed a dose-dependent transcriptional enhancement of ERα activity ≥100 μM with maximum fold change of ≥20% of E2.

Table 2.

Summary of Estrogenic Activity of LO and TTO Components

Reporter Assaya ERα-Regulated Genesb
Treatment ERα ERE-Mediated Activityc GREB1c PGRc CTSDc
E2d 100 100.0 100.0 100.0
LO 17.0 17.9 8.3 101.3
TTO 17.9 22.6 12.7 90.2
Euc 3.0 10.4 8.1 87.6
4-Terp 15.4 13.8 10.6 94.2
Di/Lim 4.0 11.0 13.3 82.7
α-Terpl 17.6 8.0 5.5 82.2
LinA 8.0 12.9 9.2 86.7
Lin 15.7 14.1 9.6 88.0
α-Terp 3.1 10.9 8.7 76.4
γ-Terp 5.1 8.1 9.4 80.9
Open in a new tab

Data are reported as %.

a

Data summarized from Fig. 1A and data repository (35).

b

Data summarized from Fig. 1D and data repository (35).

c

Higher percentages indicate stronger estrogenicity and smaller percentages indicate weak estrogenicity.

d

E2 activation was set to 100% and percent change in activity was calculated relative to twofold change for LO, TTO, and the eight components.

ERα ERE-mediated activity with SRC-2

To determine if the ERα ERE-mediated transcriptional regulation by the components and oils were modulated by specific coactivators, HepG2 cells were cotransfected with SRC-2 and ERα expression plasmid. Significant induction of ERα ERE-mediated activity occurred with 1 nM E2 and 0.01% (v/v) LO and TTO (35). However, when components of LO and TTO were tested, there was specific SRC-2 selectivity. 4-Terp, α-Terpl, LinA, and γ-Terp were the only components that significantly activated ERα ERE-mediated activity with the addition of SRC-2 (Fig. 1B). These findings suggest that certain components in LO and TTO can effectively and selectively recruit SRC-2 and increase the transcriptional activation in the ERα-ligand complex.

Coactivator recruitment and functional ER α ligand binding

HepG2 cells were cotransfected with pG5 luciferase and pM-SRC-2-NR (SRC-2 NR box) in the presence of VP16-AD pACT or pACT-LBD/WT to determine the ability of ligands to bind to the receptor and initiate recruitment of SRC-2 to the ERα LBD (Fig. 1C). The positive control, 1 nM E2, recruited SRC-2 NR box and demonstrated binding to the LBD of ERα (35). In addition, 0.01% LO, LO-C, and TTO showed recruitment of SRC-2 NR box and demonstrated binding to the LBD of ERα. Studies with components common and specific between LO and TTO demonstrated selective ability to recruit SRC-2 to the ERα LBD. Only three of the components (α-Terpl, LinA, and Lin) showed significant activity dependent on binding to the LBD of ERα and recruitment of the SRC-2 NR box (Fig. 1C). These findings indicate that some of the selected components in LO and TTO can bind to the LBD of ERα and effectively recruit SRC-2.

ERα-regulated gene expression

To determine the activity of the components in regulating endogenous genes, we characterized the ERα-dependent responses of LO and TTO components by examining three well-known ERα regulated genes: GREB1, PGR, and CTSD (28). MCF-7 breast cancer cells were treated with 1 nM E2, 0.025% (v/v) LO or TTO, and 1000 μM of the eight components, with or without the ERα antagonist ICI, and quantitative PCR was performed. All three genes showed significant induction when treated with E2, LO, and TTO, and were blocked when cotreated with ICI, the ERα antagonist. The only exception was the lack of LO activity for PGR stimulation (35). GREB1 showed significant induction with all eight components (Fig. 1D). In addition, significant GREB1 gene induction was inhibited by ICI with all treatments (Fig. 1D). Induction of the PGR gene was more selective than GREB1 induction. Treatment with 4-Terp, Di/Lim, LinA, Lin, and γ-Terp showed significant PGR gene upregulation, and that induction was significantly blocked when cotreated with ICI (Fig. 1D). As for CTSD, all components induced significant upregulation. ICI reduced CTSD induction for all components except α-Terp and γ-Terp (Fig. 1D). These findings indicate that differential gene responses are seen from stimulation by select components in LO and TTO. This group of ERα-regulated genes was blocked by ICI, but the inhibition was individualized and not common across all components and genes (Fig. 1A-D).

AR MMTV-mediated activity

Previously, we reported a unique property of LO and TTO that showed antiandrogenic activities in vitro (8). To test this further, the AR-positive MDA-kb2 cell line stably expressing an MMTV promotor reporter construct was used to investigate AR MMTV-mediated activity by the oils and their components. Cells had a positive response to 1 nM testosterone with a three- to fourfold increase and that response was significantly blocked when cotreated with the AR antagonist Flut (35). Increasing dose of LO and TTO significantly inhibited AR MMTV-mediated activity at all concentrations ≥0.0005% (v/v) when cotreated with testosterone (35).

We then investigated the single oil components that may be responsible for that activity. Three of the four oil components common between LO and TTO significantly decreased fold activation compared with 1 nM testosterone, to the level comparable to 1 μM Flut, a known antiandrogen (Fig. 2A). 4-Terp and α-Terpl showed a significant decrease in AR MMTV-mediated activity at concentrations ≥1 μM. Di/Lim was not as potent and significantly reduced AR-MMTV mediated activity at ≥10 μM. Euc showed no activity or significant reduction in AR MMTV-mediated activity at any of the concentrations (Fig. 2A).

Figure 2.

Figure 2.

Open in a new tab

AR MMTV-mediated transcriptional activity and AR-regulated gene expression. (A) Dose-dependent antiandrogenic responses of MDA-kb2 cells when cotreated with testosterone (T). A Dunnett multiple comparison test was performed in comparison with 1 nM T. All experiments were performed at least three times. Results are a representation of one experiment and are presented as mean ± SEM. (B) Antiandrogenic responses of endogenous genes in MDA-kb2 cells when cotreated with T. Total RNA was extracted and mRNA levels of the genes SEC14L2, CYP4F8, and UGT2B28 were quantified by quantitative PCR analysis. All gene fold changes were normalized against ACTB. A Dunnett multiple comparison test was performed in comparison with 1 nM T. All experiments were performed at least three times. Results are an average of the experiments and are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Luc, luciferase.

The components specific to LO, Lin and LinA, showed inhibition of AR MMTV-mediated activity at ≥1 μM or ≥100 μM, with Lin having the more complete efficacy of ≥1 μM (Fig. 2A). The TTO components reduced AR MMTV-mediated activity at 100 and 1000 μM only, with γ-Terp being more highly effective than α-Terp (Fig. 2A). These findings are summarized in Table 3 and indicate that the selected individual components in LO and TTO can inhibit transcriptional activity of the AR.

Table 3.

Summary of Antiandrogenic Activity of LO and TTO Components

Reporter Assaya,b AR Regulated Genesc
Treatment AR MMTV-Mediated Activity CYP4F8 UGT2B28 SEC14L2
Td 100.0 100.0 100.0 100.0
T + Flut 54.4 10.3 18.8 14.3
T + LO 37.0 4.5 17.5 13.7
T + TTO 35.9 5.5 20.5 14.0
T + Euc 90.0 143.7 80.0 68.7
T + 4-Terp 26.4 64.4 54.1 43.3
T + Di/Lim 75.7 38.9 98.7 57.2
T + α-Terpl 12.3 6.0 28.8 16.8
T + LinA 72.7 36.3 46.7 29.6
T + Lin 46.1 20.1 34.1 24.2
T + α-Terp 74.1 104.8 100.1 79.6
T + γ-Terp 39.1 51.0 102.7 78.8
Open in a new tab

Data are reported as %.

Abbreviation: T, testosterone.

a

Data summarized from Fig. 2A and data repository (35).

b

Smaller percentages indicate stronger antiandrogenicity and larger percentages indicate weaker antiandrogenicity.

c

Data summarized from Fig. 2D and data repository (35).

d

Testosterone activation was set to 100% and percent change in activity was calculated relative to testosterone fold-change for LO, TTO, and the eight components.

AR-regulated gene expression

To test if AR-regulated endogenous gene responses were also inhibited by LO and TTO components, we examined three well-known AR-target genes:SEC14L2, CYP4F8, and UGT2B28. Gene expression was measured using quantitative PCR in treated MDA-kb2 cells (8). All data were normalized against ACTB and are presented in Fig. 2B. The controls, testosterone plus Flut, testosterone plus LO, and testosterone plus TTO, all significantly blocked SEC14L2, CYP4F8, and UGT2B28 (35). All eight components showed significant inhibition of testosterone-stimulated SEC14L2 expression (Fig. 2B). However, inhibition of CYP4F8 was more selective than SEC14L2. 4-Terp, Di/Lim, α-Terpl, LinA, Lin, and γ-Terp significantly blocked CYP4F8 induction, but Euc and α-Terp showed no antiandrogenic effects on CYP4F8. UGT2B28 was the most selective of the three genes examined. 4-Terp, α-Terpl, LinA, and Lin were the only components that showed significant inhibition of induction of UGT2B28, whereas Euc, Di/Lim, α-Terp, and γ-Terp showed no antiandrogenic effects on UGT2B28 (Fig. 2B). These findings demonstrate that some of the essential-oil components demonstrate differential antiandrogenic properties of AR-regulated genes.

Discussion

Premature thelarche and prepubertal gynecomastia are secondary to precocious puberty or to abnormal production of sex hormones before the normal age. Thelarche and pubarche are considered normal after age 8 years. Menarche is considered normal after age 9 years in girls and gonadarche is considered normal after age 9 years in boys (38). Approximately 40% to 60% of boys develop pubertal gynecomastia after age 9 years, which is considered normal (7). However, over the past decade, the onset of puberty has been reduced to younger ages by 12 to 18 months (39). EDCs have been well accepted for their ability to modulate the endocrine system (39). Therefore, it has been thought that environmental exposures could be considered as a possible cause of these conditions (39). Topical applications of LO and TTO have been linked previously to prepubertal gynecomastia in boys, but such a condition in girls has not been reported, to our knowledge (7, 8).

In this study, we describe the development of premature thelarche in three girls and a case of prepubertal gynecomastia in one boy. The continuous use of LO-fragrance products was common across these clinical cases, and in all cases, breast tissue regressed when oils were discontinued. Other possible considerations should be mentioned for these clinical conditions. Apart from LO and TTO components, commercial products containing essential oils could possess ingredients that have EDC effects. Consideration of other active ingredients is difficult, because most commercial over-the-counter products have minimal ingredient listings. Therefore, we could only assume that products labeled as containing LO followed ISO regulations and did contain the components that were relevant to the components we tested. Transient thelarche, a clinically reported condition that can subside on its own, might be another explanation for the breast tissue resolving. Thus, the resolving of the patient’s condition could have been coincidental in timing to the discontinuation of the suspected essential-oil products.

Fragranced products are used at higher rates in Hispanic communities than other ethnicities (e.g., the product agua) (40). Diaz et al. (7) reported HPLC–mass spectrometry analysis of the aqua de violetas product, which detected the presence of Lin and LinA, as well as unidentified substances. These two components of LO were analyzed in our study and both showed antiandrogenic and weak estrogenic activity. The association between LO-containing products and premature thelarche and prepubertal gynecomastia appear to be more prevalent in Hispanic populations, based on observations in our clinic. Premature thelarche may not have been reported previously because of relatively high prevalence of idiopathic breast development in girls compared with boys and the high rate of regression (4, 5). We speculate that another possible explanation for the high prevalence in the Hispanic population could be due to a genetic polymorphism resulting in sensitivity to the essential oils or other components. Additional studies would be required to determine if such a genetic susceptibility exists associated with exposure to these products.

Multiple studies have focused on exposure rates of these essential oils from many personal care products and cosmetics (9, 41). A study examined skin exposures of essential oils between men and women. There were differences between the areas of the body to which the essential oils were most commonly applied between men and women. The chest/breast area was the most commonly applied area on the body of men (64%), whereas the face/neck area was highest in women (71%), and were both statistically significant from the other sex. When comparing exposures, LO was the most used essential oil by men and women (9). Such clinical findings could be another explanation of why topical exposure to essential oils and the development of prepubertal breast growth is seen at higher rates in boys than in girls.

Previous studies have demonstrated that dermal exposure to essential oils results in circulating concentrations that are comparable to the concentrations used in vitro (42, 43). In contrast, an industry-supported experimental study reported percutaneous exposure of LO showed no uterotropic estrogenic activity (44). An additional report indicated no estrogenic activity in a yeast assay using Lin (45). The discrepancies in such findings are likely directly related to differences in experimental approaches used between the studies. An assumption that percutaneous dorsal exposure would produce an uterotropic action revolves around the assumption that sufficient absorption of the LO fragrance would become systemic to produce a distant tissue estrogen effect. The primary method to test uterine estrogenic activity is principally done either by ingestion or injection. In one study, the comparative positive control used gavage of 17α ethynyl estradiol (44). The treatment schemes were quite different between the two groups, although it would have been a directly testable exposure with a fully accepted target end point if found positive (44). On the other hand, in our cases, the patients’ suspected area of exposure, which was the breast area, was in direct contact with the essential oil and not at a distant site. We were unable to determine the circulating concentrations of components of essential oils in these children. However, it is important to note that exposure to directly affected areas was continuous over month-long periods in each case. Moreover, the specific compounds are lipophilic and even low-dose exposures would be expected to result in some accumulation in the breast and associated adipose tissue sites (27). Continuous early-age exposures with other substances have been reported with EDC effects (46, 47).

Essential-oil components have not been previously classified as EDCs, but our current data provide evidence that these components have properties that match the definition of an EDC (26, 27). We confirmed that both LO and TTO contain components that antagonize AR transcriptional activity and weakly induce ERα gene responses, although we found that not all components are hormonally active. Lower transcriptional activation was observed in the composed mixtures vs the natural LO and TTO product mixtures. These differences may indicate that there could be other components or substances present in the essential oils that also possess estrogenic activity or act synergistically with other EDCs, to explain the observed differences (48).

The estrogenic and antiandrogenic effects of the essential oil components are summarized in Tables 2 and 3. Overall, the oils and their components appear experimentally to have greater potency as antiandrogens than as estrogens. Interestingly, components that had stronger estrogenicity also demonstrated the strongest antiandrogenic properties. Antiandrogenic activity is not directly linked to breast stimulation, although the antiandrogenic action of the essential oils would be more relevant to blocking the natural action in boys than girls, in whom androgens inhibit breast development (49). As shown previously (7), the activity would be most effective in the presence of low endogenous androgen levels (e.g., prepubertal boys) than those of adolescents or adults with higher postpubertal androgen levels. Moreover, the same relationship between the strengths of estrogenicity and antiandrogenicity are also seen with other well-known EDCs, such as bisphenol A (BPA) and bisphenol AF (27, 48, 50). Most detrimental actions of hormonally active agents have been linked to very potent substances (e.g., diethylstilbestrol). Whether the lavender-fragrance components have sufficient potency to elicit the effects is impossible to test because intentional dosing with essential-oil components is untenable. The level of activity of the more active components, as reported from the experimental data in Table 2, is similar to that of estriol, the natural weak estrogen (51, 52). Experimentally, it has been shown that persistent exposure to estriol results in comparable activity and responsiveness as to estradiol (53, 54). One possible explanation is that although weaker in potency, the localized and persistent exposure or application to the essential-oil fragrance may result in developing the condition, especially in individuals of high sensitivity. Additional studies will have to be conducted to test this possibility.

Coactivators, including SRC-2, are a critical component needed to execute the many sub steps of transcription involving NRs and modulate the magnitude of transcriptional responses (25, 55). We showed that LO, TTO, and certain components had enhanced ERα transcription with the addition of SRC-2, suggesting that the selected components actions in vivo could be more potent hormonal mimics than demonstrated in vitro, because SRC-2 is present in humans (24, 55). More interestingly, in the functional two-hybrid assay, LO, TTO, and select components bound to the LBD of ERα showed recruitment of SRC-2. Some of the components that displayed stimulation of ERα transcription with addition of SRC-2 showed no recruitment of SRC-2 and binding to the ERα-LBD. It is unclear why this discrepancy occurs, but we previously reported that EDCs, such as BPA, can selectively bind to certain SRC binding motifs, but not other motifs (28). The current two-hybrid binding system used in the current has certain limitations because it is composed of only two binding motifs compared with multiple other motifs present on the full-length SRC-2 (37). Therefore, our assay used only two motifs of the SRC and it is possible that the biological activity was a result of recruitment of another motif. Thus, this may explain why only a few components bound to the ERα-LBD but still showed significant induction when tested with full-length SRC-2 overexpression. A larger and broader coactivator screen, as used previously (28), would need to be done to test this possibility.

Earlier speculation raised questions regarding an earlier report (8) as to whether the properties of an oil in general could dissolve BPA analogs from the plastic assay plates and, therefore, elicit the detected hormonal activity (56). We addressed this concern by conducting additional experiments with corn and soybean oil alongside LO and TTO. Shown in the data repository (35), LO and TTO, as well as some of their individual components, had estrogenic activity, but corn and soybean oil had no detectable activity.

In summary, we have shown regression of premature thelarche in addition to pubertal gynecomastia after withdrawal of lavender-fragranced products. We have also measured the hormonal activity of select essential-oil components and their spectrum of ER and AR activities. Taken together, it is important that physicians are aware that LO and TTO possess EDC activities that should be considered in the evaluation of premature breast development in girls and gynecomastia in boys and adult men. We are not recommending any avoidance of these products; rather, we are suggesting that essential-oil products may be considered for discontinuance if suspected to be a possible cause of idiopathic premature thelarche or prepubertal gynecomastia.

Acknowledgments

We thank Sylvia C. Hewitt and Suzanne E. Fenton, National Institute of Environmental Health Sciences (NIEHS), for critical review of this manuscript. The authors are especially appreciative to Janet Hall, NIEHS clinical director, for her comments, insights, and suggestions regarding the study and to Ricardo Restrepo for his careful evaluation and verification of the breast ultrasound analysis.

Financial Support: This work was supported by the Division of Intramural Research of the National Institute of Environmental Health Sciences (1ZIAES070065 to K.S.K.).

Author Contributions: J.T.R. and Y.L. conducted the experiments. A.D. and A.N. performed the clinical assessments. J.T.R., Y.L., Y.A., L.A.C., and K.S.K. participated in the design of the experiments. J.T.R., Y.L., Y.A., A.D., A.N., and K.S.K. interpreted the data. J.T.R., Y.L., and K.S.K. wrote and edited the manuscript.

Additional Information

Disclosure Summary: The authors have nothing to disclose.

Data Availability: All data generated or analyzed during this study are included in this published article or in the data repositories listed in References.

Glossary

Abbreviations:

4-Terp

4-terpineol

AR

androgen receptor

BPA

bisphenol A

CTSD

cathepsin D

CYP4F8

cytochrome P450 family 4 subfamily F member 8

Di/Lim

dipentene/limonene

DMSO

dimethyl sulfoxide

E2

estradiol

EDC

endocrine-disrupting chemical

ERα

estrogen receptor α

ERE

estrogen response element

EUC

eucalyptol

FBS

fetal bovine serum

Flut

flutamide

GREB1

growth regulation by estrogen in breast cancer 1

ICI

fulvestrant/ICI 182,780

ISO

International Organization for Standardization

LBD

ligand-binding domain

Lin

linalool

LinA

linalyl acetate

LO

lavender oil

LO-C

lavender oil component

MMTV

mouse mammary tumor virus

NR

nuclear receptor

PGR

progesterone receptor

SEC14L2

SEC14-like lipid binding 2

sFBS

stripped fetal bovine serum

SRC

steroid receptor coactivator

TTO

tea tree oil

TTO-C

tea tree oil component

α-Terp

α-terpinene

α-Terpl

α-terpineol

γ-Terp

γ-terpinene

UGT2B28

UDP glucuronosyltransferase family 2 member B28

References and Notes

  • 1. Hamilton KJ, Arao Y, Korach KS. Estrogen hormone physiology: reproductive findings from estrogen receptor mutant mice. Reprod Biol. 2014;14(1):3–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Quaynor SD, Stradtman EW Jr, Kim HG, Shen Y, Chorich LP, Schreihofer DA, Layman LC. Delayed puberty and estrogen resistance in a woman with estrogen receptor α variant. N Engl J Med. 2013;369(2):164–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Braunstein GD. Gynecomastia. N Engl J Med. 1993;328(7):490–495. [DOI] [PubMed] [Google Scholar]
  • 4. Çiçek D, Savas-Erdeve S, Cetinkaya S, Aycan Z. Clinical follow-up data and the rate of development of precocious and rapidly progressive puberty in patients with premature thelarche. J Pediatr Endocrinol Metab. 2018;31(3):305–312. [DOI] [PubMed] [Google Scholar]
  • 5. Çatlı G, Erdem P, Anık A, Abacı A, Böber E. Clinical and laboratory findings in the differential diagnosis of central precocious puberty and premature thelarche. Turk Pediatri Ars. 2015;50(1):20–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Lindhardt Johansen M, Hagen CP, Mieritz MG, Wolthers OD, Heuck C, Petersen JH, Juul A. Pubertal progression and reproductive hormones in healthy girls with transient thelarche. J Clin Endocrinol Metab. 2017;102(3):1001–1008. [DOI] [PubMed] [Google Scholar]
  • 7. Diaz A, Luque L, Badar Z, Kornic S, Danon M. Prepubertal gynecomastia and chronic lavender exposure: report of three cases. J Pediatr Endocrinol Metab. 2016;29(1):103–107. [DOI] [PubMed] [Google Scholar]
  • 8. Henley DV, Lipson N, Korach KS, Bloch CA. Prepubertal gynecomastia linked to lavender and tea tree oils. N Engl J Med. 2007;356(5):479–485. [DOI] [PubMed] [Google Scholar]
  • 9. Dornic N, Ficheux AS, Roudot AC, Saboureau D, Ezzedine K. Usage patterns of aromatherapy among the French general population: a descriptive study focusing on dermal exposure. Regul Toxicol Pharmacol. 2016;76:87–93. [DOI] [PubMed] [Google Scholar]
  • 10. de Groot AC, Schmidt E. Essential oils, part III: chemical composition. Dermatitis. 2016;27(4):161–169. [DOI] [PubMed] [Google Scholar]
  • 11. de Groot AC, Schmidt E. Tea tree oil: contact allergy and chemical composition. Contact Dermat. 2016;75(3):129–143. [DOI] [PubMed] [Google Scholar]
  • 12. de Groot A, Schmidt E. Essential oils, part V: peppermint oil, lavender oil, and lemongrass oil. Dermatitis. 2016;27(6):325–332. [DOI] [PubMed] [Google Scholar]
  • 13. Tanos T, Rojo L, Echeverria P, Brisken C. ER and PR signaling nodes during mammary gland development. Breast Cancer Res. 2012;14(4):210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Arendt LM, Kuperwasser C. Form and function: how estrogen and progesterone regulate the mammary epithelial hierarchy. J Mammary Gland Biol Neoplasia. 2015;20(1-2):9–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Bocchinfuso WP, Lindzey JK, Hewitt SC, Clark JA, Myers PH, Cooper R, Korach KS. Induction of mammary gland development in estrogen receptor-alpha knockout mice. Endocrinology. 2000;141(8):2982–2994. [DOI] [PubMed] [Google Scholar]
  • 16. Chauchereau A, Savouret JF, Milgrom E. Control of biosynthesis and post-transcriptional modification of the progesterone receptor. Biol Reprod. 1992;46(2):174–177. [DOI] [PubMed] [Google Scholar]
  • 17. Tekmal RR, Liu YG, Nair HB, Jones J, Perla RP, Lubahn DB, Korach KS, Kirma N. Estrogen receptor alpha is required for mammary development and the induction of mammary hyperplasia and epigenetic alterations in the aromatase transgenic mice. J Steroid Biochem Mol Biol. 2005;95(1-5):9–15. [DOI] [PubMed] [Google Scholar]
  • 18. Conte FA, Grumbach MM, Ito Y, Fisher CR, Simpson ER. A syndrome of female pseudohermaphrodism, hypergonadotropic hypogonadism, and multicystic ovaries associated with missense mutations in the gene encoding aromatase (P450arom). J Clin Endocrinol Metab. 1994;78(6):1287–1292. [DOI] [PubMed] [Google Scholar]
  • 19. Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA. 1993;90(23):11162–11166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. McDonnell DP, Norris JD. Connections and regulation of the human estrogen receptor. Science. 2002;296(5573):1642–1644. [DOI] [PubMed] [Google Scholar]
  • 21. Nilsson S, Mäkelä S, Treuter E, Tujague M, Thomsen J, Andersson G, Enmark E, Pettersson K, Warner M, Gustafsson JA. Mechanisms of estrogen action. Physiol Rev. 2001;81(4):1535–1565. [DOI] [PubMed] [Google Scholar]
  • 22. Hall JM, McDonnell DP. Coregulators in nuclear estrogen receptor action: from concept to therapeutic targeting. Mol Interv. 2005;5(6):343–357. [DOI] [PubMed] [Google Scholar]
  • 23. Deroo BJ, Korach KS. Estrogen receptors and human disease. J Clin Invest. 2006;116(3):561–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Dasgupta S, Lonard DM, O’Malley BW. Nuclear receptor coactivators: master regulators of human health and disease. Annu Rev Med. 2014;65(1):279–292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Donoghue LJ, Neufeld TI, Li Y, Arao Y, Coons LA, Korach KS. Differential activation of a mouse estrogen receptor β isoform (mERβ2) with endocrine-disrupting chemicals (EDCs). Environ Health Perspect. 2017;125(4):634–642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Zoeller RT, Brown TR, Doan LL, Gore AC, Skakkebaek NE, Soto AM, Woodruff TJ, Vom Saal FS. Endocrine-disrupting chemicals and public health protection: a statement of principles from The Endocrine Society. Endocrinology. 2012;153(9):4097–4110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Diamanti-Kandarakis E, Bourguignon JP, Giudice LC, Hauser R, Prins GS, Soto AM, Zoeller RT, Gore AC. Endocrine-disrupting chemicals: an Endocrine Society scientific statement. Endocr Rev. 2009;30(4):293–342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Li Y, Perera L, Coons LA, Burns KA, Tyler Ramsey J, Pelch KE, Houtman R, van Beuningen R, Teng CT, Korach KS. Differential in vitro biological action, coregulator interactions, and molecular dynamic analysis of bisphenol A (BPA), BPAF, and BPS ligand-ERα complexes. Environ Health Perspect. 2018;126(1):017012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Wilson VS, Bobseine K, Lambright CR, Gray LE Jr. A novel cell line, MDA-kb2, that stably expresses an androgen- and glucocorticoid-responsive reporter for the detection of hormone receptor agonists and antagonists. Toxicol Sci. 2002;66(1):69–81. [DOI] [PubMed] [Google Scholar]
  • 30. Sanderson JT. The steroid hormone biosynthesis pathway as a target for endocrine-disrupting chemicals. Toxicol Sci. 2006;94(1):3–21. [DOI] [PubMed] [Google Scholar]
  • 31. Lu H, Li H, Lu H, Li XL, Zhou AG. Chemical composition of lavender essential oil and its antioxidant activity and inhibition against rhinitis-related bacteria. Afr J Microbiol Res. 2010;4:309–313. [Google Scholar]
  • 32. Shellie R, Marriott P, Cornwell C. Characterization and comparison of tea tree and lavender oils by using comprehensive gas chromatography. J High Resolut Chromatogr. 2000;23(9):554–560. [Google Scholar]
  • 33. Shellie R, Mondello L, Marriott P, Dugo G. Characterisation of lavender essential oils by using gas chromatography-mass spectrometry with correlation of linear retention indices and comparison with comprehensive two-dimensional gas chromatography. J Chromatogr A. 2002;970(1-2):225–234. [DOI] [PubMed] [Google Scholar]
  • 34. Nosková K, Dovrtělová G, Zendulka O, Řemínek R, Juřica J. The effect of (-)-linalool on the metabolic activity of liver CYP enzymes in rats. Physiol Res. 2016;65(Suppl 4):S499–S504. [DOI] [PubMed] [Google Scholar]
  • 35. Ramsey JT, Li Y, Arao Y, Naidu A, Coons LA, Diaz A, Korach KS. Data from: Lavender products associated with premature thelarche and prepubertal gynecomastia: case reports and endocrine-disrupting chemical activities. figshare 2018. Deposited 15 September 2019. 10.6084/m9.figshare.7045109.v1. [DOI] [PMC free article] [PubMed]
  • 36. Arao Y, Korach KS. Detecting the ligand-binding domain dimerization activity of estrogen receptor alpha using the mammalian two-hybrid assay. J Vis Exp. 2018;(142):e58758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Arao Y, Hamilton KJ, Coons LA, Korach KS. Estrogen receptor α L543A,L544A mutation changes antagonists to agonists, correlating with the ligand binding domain dimerization associated with DNA binding activity. J Biol Chem. 2013;288(29):21105–21116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Fuqua JS. Treatment and outcomes of precocious puberty: an update. J Clin Endocrinol Metab. 2013;98(6):2198–2207. [DOI] [PubMed] [Google Scholar]
  • 39. Sultan C, Gaspari L, Maimoun L, Kalfa N, Paris F. Disorders of puberty. Best Pract Res Clin Obstet Gynaecol. 2018;48:62–89. [DOI] [PubMed] [Google Scholar]
  • 40. Dole EJ, Rhyne RL, Zeilmann CA, Skipper BJ, McCabe ML, Low Dog T. The influence of ethnicity on use of herbal remedies in elderly Hispanics and non-Hispanic whites. J Am Pharm Assoc (Wash). 2000;40(3):359–365. [DOI] [PubMed] [Google Scholar]
  • 41. Comiskey D, Api AM, Barratt C, Daly EJ, Ellis G, McNamara C, O’Mahony C, Robison SH, Safford B, Smith B, Tozer S. Novel database for exposure to fragrance ingredients in cosmetics and personal care products. Regul Toxicol Pharmacol. 2015;72(3):660–672. [DOI] [PubMed] [Google Scholar]
  • 42. Jäger W, Buchbauer G, Jirovetz L, Fritzer M. Percutaneous absorption of lavender oil from a massage oil. J Soc Cosmet Chem. 1991;43:49–54. [Google Scholar]
  • 43. Rotha Z, Shcherbinab Y, Nussinovitchb A. Transdermal delivery of Lavandula angustifolia and valencia orange essential oils using gum-karaya patches. Med Aromat Plants (Los Angeles) 2017;6:3. [Google Scholar]
  • 44. Politano VT, McGinty D, Lewis EM, Hoberman AM, Christian MS, Diener RM, Api AM. Uterotrophic assay of percutaneous lavender oil in immature female rats. Int J Toxicol. 2013;32(2):123–129. [DOI] [PubMed] [Google Scholar]
  • 45. Howes MJ, Houghton PJ, Barlow DJ, Pocock VJ, Milligan SR. Assessment of estrogenic activity in some common essential oil constituents. J Pharm Pharmacol. 2002;54(11):1521–1528. [DOI] [PubMed] [Google Scholar]
  • 46. Braun JM. Early-life exposure to EDCs: role in childhood obesity and neurodevelopment. Nat Rev Endocrinol. 2017;13(3):161–173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Kodavanti PR, Coburn CG, Moser VC, MacPhail RC, Fenton SE, Stoker TE, Rayner JL, Kannan K, Birnbaum LS. Developmental exposure to a commercial PBDE mixture, DE-71: neurobehavioral, hormonal, and reproductive effects. Toxicol Sci. 2010;116(1):297–312. [DOI] [PubMed] [Google Scholar]
  • 48. Li Y, Luh CJ, Burns KA, Arao Y, Jiang Z, Teng CT, Tice RR, Korach KS. Endocrine-disrupting chemicals (EDCs): in vitro mechanism of estrogenic activation and differential effects on ER target genes. Environ Health Perspect. 2013;121(4):459–466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Dimitrakakis C, Bondy C. Androgens and the breast. Breast Cancer Res. 2009;11(5):212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Rasier G, Parent AS, Gérard A, Lebrethon MC, Bourguignon JP. Early maturation of gonadotropin-releasing hormone secretion and sexual precocity after exposure of infant female rats to estradiol or dichlorodiphenyltrichloroethane. Biol Reprod. 2007;77(4):734–742. [DOI] [PubMed] [Google Scholar]
  • 51. Hisaw FL, Velardo JT, Goolsby CM. Interaction of estrogens on uterine growth. J Clin Endocrinol Metab. 1954;14(10):1134–1143. [DOI] [PubMed] [Google Scholar]
  • 52. Kuhl H. Pharmacology of estrogens and progestogens: influence of different routes of administration. Climacteric. 2005;8(Suppl 1):3–63. [DOI] [PubMed] [Google Scholar]
  • 53. Lan NC, Katzenellenbogen BS. Temporal relationships between hormone receptor binding and biological responses in the uterus: studies with short- and long-acting derivatives of estriol. Endocrinology. 1976;98(1):220–227. [DOI] [PubMed] [Google Scholar]
  • 54. Anderson JN, Peck EJ Jr, Clark JH. Estrogen-induced uterine responses and growth: relationship to receptor estrogen binding by uterine nuclei. Endocrinology. 1975;96(1):160–167. [DOI] [PubMed] [Google Scholar]
  • 55. Lonard DM, O’malley BW. Nuclear receptor coregulators: judges, juries, and executioners of cellular regulation. Mol Cell. 2007;27(5):691–700. [DOI] [PubMed] [Google Scholar]
  • 56. Carson CF, Tisserand R, Larkman T. Lack of evidence that essential oils affect puberty. Reprod Toxicol. 2014;44:50–51. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES