Chemical mixtures in human post-mortem tissues assessed by a combination of chemical analysis and in vitro bioassays after extraction with silicone.

Passive equilibrium sampling of chemical mixtures from different human post-mortem tissues (liver, brain (cerebrum and cerebellum), adipose tissue) and blood was combined with instrumental analysis using direct sample introduction (DSI) GC-MS/MS and bioanalytical profiling using in vitro bioassays targeting the activation of the aryl hydrocarbon receptor (AhR-CALUX), the adaptive stress response (AREc32) and cytotoxicity. The tissues stemmed from pathology samples collected in two German cities and covered males and females aged 21 to 100 with a mean age of 67 years. Neutral organic chemicals were extracted using polydimethylsiloxane (PDMS) at mass ratios of tissue to PDMS of approximately 6 for blood, 3 for adipose tissue and 10 for liver and brain. Amounts of chemicals in PDMS were converted to lipid-associated concentrations using previously measured partition constants that were chemical-independent despite covering eight orders of magnitude in hydrophobicity. Up to 35 of 99 targeted chemicals were detected in 6 tissues of 16 individuals (88 samples in total), among them legacy persistent organic pollutants (POP) such as DDT and derivatives and polychlorinated biphenyls (PCB) but also modern pesticides and chemicals present in consumer products. POPs were highest in adipose tissue and lipid-associated concentrations increased with age, while concentrations of fragrance materials such as galaxolide were independent of age. In tissues from the same individual, chemical concentrations mostly increased from similar levels in brain and blood to higher levels in liver and highest in adipose tissue. However, easily degradable chemicals such as phenanthrene were mainly detected in blood and brain, and very hydrophilic chemicals were least abundant in adipose tissue. The passive sampling method allows a direct comparison of chemical burden between different tissues and may have forensic applications, for example to study internal distributions or to use one tissue type as a proxy for others. The sum of concentrations of the detected chemicals was positively correlated with the bioassay responses but mixture modeling showed that the detected chemicals explained less than 2% of the activation of the AhR and less than 0.5% of cytotoxicity. This means that more than 10,000 chemicals would need to be included in an analytical method to capture all the effects with many chemicals potentially being below detection limits but still contributing to mixture effects. Therefore, we propose a smart combination of chemical analysis and bioassays to quantify priority chemicals but use bioassay responses as effect-scaled concentrations to capture the entire exposome in future epidemiological studies.