As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.


Figure S1. Transcriptional activation of PPARs by treatment of transfected reporter cells with lemon balm extract (CFE). (A) PPARα transcriptionally activated by LBE (μg/ml) or GW7647 (nM). (B) PPARβ/δ activated by CFE (μg/ml) or GW7647 (nM). (C) PPARγ activated by CFE (μg/ml) or rosiglitazone (RGZ, nM). Transcriptional activation was determined in a reporter gene assay and is represented relative to the reference compound. Data are expressed as mean ± SD (n=3-4). LBE, lemon balm extract; RGZ, rosiglitazone.

Figure S2. Human primary adipocytes were transfected with siRNA against PPARα, PPARβ/δ and PPARγ (hatched bars) or control siRNA and knockdown efficiencies were evaluated by qPCR. Data are expressed as mean ± SEM (n = 4/group). ***p ≤ 0.001, vs. control siRNA.

Figure S3. Effect of treatment of C57BL/6 mice with high-fat diet (HFD) without or with 200 mg/kg/d lemon balm extract (LBE) on (A) cumulative caloric intake and (B) body weight gain. Data are expressed as mean ± SEM (n = 14/group).

Figure S4. Investigation of the interaction of fractions of ethanolic lemon balm extract with the exemplary PPAR subtype PPARγ. (A) Binding of all 10 subfractions to PPARγ was measured in a competitive time-resolved fluorescence resonance energy transfer assay and observed binding affinity constants Ki (left panel) were plotted against the number of fractions with increasing hydrophobicity (right panel). Point size is related to the weight proportion of that fraction relative to the parental extract. Statistical significance of correlation was calculated assuming nonparametric distribution. r, Spearman correlation coefficient; p, two-tailed p value. (B) Gene expression of 4 fractions of lemon balm extract (LBE) in human adipocytes. Fractions were chosen based on their binding affinity constants and percentage of the parental extract. The concentration applied was 600 μg/ml for the parental extract, and for the resultant fractions we applied 41 μg/ml, 16 μg/ml, 9 μg/ml and 6 μg/ml of fraction numbers 4, 6, 8 and 9, respectively. These concentrations exactly reflect the proportion of the fractions in the parental extract, i.e. 100% for the LBE, and 5.8%, 2.2%, 1.3% and 0.9% for fraction no. 4, 6, 8 and 9, respectively. The concentrations of all subfractions were still clearly above the appropriate thermodynamic binding constants (Ki or KD) for PPARγ, namely 0.6 μg/ml (68-fold), 0.6 μg/ml (27-fold), 0.4 μg/ml (23-fold) and 0.3 μg/ml (20-fold) for fraction number 4, 6, 8 and 9. Primary adipocytes were treated with indicated fractions or vehicle only for 24 h and gene expression was analyzed with qPCR. Data are expressed as mean ± SEM (n=3). n.s. not significant; *p ≤ 0.05; **p ≤ 0.01 vs. vehicle.

Table S1. Primer sequences used for quantitative real-time PCR.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.