(I-349) Investigating Obesogenic Effects of Common Artificial Food Colors on Adipose Tissue

Introduction: Identifying factors that cause obesity is critical to combating its development as a global pandemic that contributes to common deadly disorders such as coronary heart disease and type 2 diabetes. Investigations into energy balance and behavioral factors have demonstrated how both can influence obesity, contributing to treatment strategies and dialogues on both. However, a novel explanation for the growing prevalence of obesity worldwide is the obesogen hypothesis: certain endocrine disrupting chemicals induce obesogenic effects on adipose tissue, such as hypertrophy or hyperplasia. Several manmade organic compounds have already been linked to obesogenic expansion of adipose tissue, namely tributyltin (TBT) and bisphenol A (BPA). This study compares these known obesogens to five common artificial food colors (AFCs): Allura Red AC, Sunset Yellow FCF, Brilliant Blue FCF, Tartrazine, and Fast Green FCF. As these FDA-approved dyes are currently under examination for several toxic effects such as carcinogenicity, it is worthwhile to consider whether they have contributed to the obesity epidemic as well. This is an especially critical consideration for childhood obesity, as AFCs used in 43% of food products targeted towards young children due to their strong visual appeal. Determining whether direct exposure of AFCs causes obesogenic effects in adipose tissue will establish grounds for a longitudinal study of AFC exposure levels during childhood and connections to the development of obesity.

Materials and Methods: To investigate the interactions between AFCs and adipose tissue, tissue samples acquired from elective surgeries were cultured ex vivo to emulate natural metabolism and growth in an easily observable setting. HFIP-dissolved B. mori silk scaffolds of a 500-600 um pore size were used, as they are biocompatible, porous, highly reproducible microenvironments for adipose tissue. The adipose tissue samples were processed and seeded into these silk scaffold microenvironments, with a 7-day culture period allowing for expansion and integration. Following this, adipose tissue was cultured in food dye and control media for 15-day trial periods (n=3). Four concentrations of each of the five AFCs were used, scaling by ten from 100 uM to 0.1 uM. In addition to negative control media and two positive controls of trace BPA and TBT, a total of 23 experimental media were used for each trial. Upon trial termination, n=3 scaffold samples per media type were chopped and frozen for PicoGreen assays and n=2 scaffolds were fixed for confocal imaging. PicoGreen assays were selected to interpret adipocyte proliferation through the sample’s DNA content. Confocal imaging was performed to identify the ratio of nuclei to lipid droplets to assess cell locularity and potential hyperplasia, as well as measuring lipid diameters to assess hypertrophy.

Results, Conclusions, and Discussions:

PicoGreen assays suggested high levels of adipocyte proliferation when exposed to Allura Red AC (Red 40), Tartrazine (Yellow 5) and Sunset Yellow FCF (Yellow 6), as the DNA concentrations for each of these samples were substantially higher than the controls and the two other AFCs: Brilliant Blue FCF (Blue 1) and Fast Green FCF (Green 3). A chart of all assay results is included as a figure. High DNA content in the samples suggest hyperplasia may have occurred, though this is one possible explanation and does not confirm obesogenic action due to food dye exposure. Additional assays to assess lipid content of the samples may help clarify growth patterns of the tissue, though triglyceride assays have been eliminated from consideration after several attempts due to colorimetric interference from the dyes themselves.

Confocal imaging analysis is currently in progress. Initial BODIPY stains demonstrate multilocularity and dense cell clusters from the food dye scaffolds with high DNA expression, such as those exposed to 1 uM of Allura Red AC. Reference images of a control and one of these samples are provided. Exact cell to lipid droplet ratios and lipid diameters for each of the 276 currently fixed samples will provide a strong basis for interpreting whether there is a significant difference between normal adipocyte growth and growth under exposure to each of the candidate obesogens. Currently, manual measurements are being performed as automated cell counts and measurements are inhibited by the visual noise created by smaller lipid droplets over the boundaries of cells.

For the future, a longitudinal study featuring temporal assessment of food dye levels in patients’ plasma could establish whether observed obesogenic effects correlate with the development of obesity in children and adults. It is possible that ex vivo tissue level interactions are not representative of how adipocytes respond to AFCs in the human body. However, combining these two analyses would be sufficient to consider whether these artificial food colors are obesogens, and whether action is required to mitigate potential contributions to obesity-linked disorders.

Acknowledgements (Optional): Rosalyn Beauregard-Abbott, Elizabeth Johnston, Megan DeBari, Hanchuan Xu

References (Optional):