Effects of Selected Bioactive Food Compounds on Human White Adipocyte Function

Christel Björk; Uta Wilhelm; Susanne Mandrup; Bjørk Ditlev Larsen; Alessandra Bordoni; Per Hedén; Mikael Rydén; Peter Arner; Jurga Laurencikiene


Nutr Metab. 2016;13(4) 

In This Article


Using a systematic assessment of the effects of specific bioactive food compounds and their metabolites in human white fat cells, we found beneficial effects of DHA, AC and PI, either alone or in combination with DHA, on fat cell metabolism and inflammatory phenotype in vitro. The in vitro differentiated non-expanded human primary adipocytes used herein, is the closest model to human fat cells in vivo and enables longer treatments with the compounds. Treatments in in vitro systems always raise a question on the physiological concentrations of the stimuli applied and relevance of the resulting effects for the in vivo situation. We utilized lower concentrations of DHA than that measured in blood.[43,44] However, higher concentrations of DHA induced cytotoxicity and pro-inflammatory response as in previous studies.[20,45] Since the formation of reactive oxygen species (ROS) correlates with the number of double bonds in a polyunsaturated fatty acid, DHA is highly susceptible to spontaneous oxidation. ROS activate surface receptors such as Toll-like receptors which can trigger a downstream signalling cascade, leading to activation of the NF-kappaB pathway, which then turn on the transcription of many pro-inflammatory genes.[46,47] In the support of this hypothesis, ascorbic acid suppressed the pro-inflammatory effect in our primary human adipocyte culture (data not shown). Furthermore, total plasma levels of DHA should not be interpreted as equivalent to local extracellular or intracellular concentrations. To the best of our knowledge, there are no studies that have measured the actual concentration of DHA in adipose tissue. However, we cannot exclude that in the in vivo environment higher concentrations of DHA would be needed for optimal effects. Additionally, none of the other compounds displayed cytotoxicity in our cell cultures and the concentrations used for PI and AC were close to those reported in human blood.[48,49]

All three compounds suppressed basal adipocyte lipolysis, where the effect of PI was slightly more pronounced compared with DHA and AC. This could at least in part be due to a fact that PI also significantly attenuated the secretion of TNFα, a potent inducer of adipocyte lipolysis.[33] Our finding corroborate results from a study where healthy subjects were supplemented for 4 weeks with non-digestible starch and demonstrated decreased adipose tissue glycerol release as well as reduction in the activity of hormone-sensitive lipase, the rate-limiting enzyme responsible for mobilization of stored TAGs.[49] We therefore suggest that PI might exert its effects at least in part by altering adipocyte lipid metabolism via indirect (through TNFα) or direct mechanisms. In fact, although the latter mechanism has not been not studied in humans, in mice PI has been suggested to interact with the G-protein coupled receptor 43,[29] a receptor that is suggested to be involved in regulating inflammatory responses, and is also expressed in human WAT.[28] Regarding ACs, an anti-lipolytic effect has been observed in 3T3-L1 adipocytes during hyperglycemia, where C3G regulated FoxO1-mediated transcription of adipose triglyceride lipase resulting in decreased expression of this lipolytic enzyme.[50] Whether or not this also implies for human adipocytes remains to be investigated. The mechanism of lipolysis reduction by DHA is not known. However, no additive effects between compounds were observed, suggesting that they might act via similar mechanisms.

The synergistic effect between DHA and AC on insulin-induced lipogenesis might partly be explained by the activation of PPARγ by AC, which results in increased expression and translocation of GLUT4.[22] In addition, the effect on lipogenesis could result from decreased TNFα secretion caused by combined DHA and AC treatment. We and others have shown that TNFα addition in cultures of primary human adipocytes causes insulin resistance.[51,52] However, a combination of PI and DHA also lowered TNFα secretion but did not affect lipogenesis, which may suggest different mechanisms of action.

Earlier studies both in mice WAT and human adipocytes have shown suppression of pro-inflammatory cytokines by AC,[24–26] DHA[17] and PI,[27] effects that are generally corroborated in the current study. DHA appears to be the strongest anti-inflammatory compound among the tested ones as this fatty acid decreased expression of three out of four investigated pro-inflammatory cytokines. We also for the first time report effects of bioactive compounds on chemerin – an insulin resistance inducing cytokine,[8] which is largely secreted by adipocytes. DHA was the only compound suppressing chemerin secretion, which was antagonized by AC. Admittedly, at present we do not know how the selected bioactives regulate adipokine secretion, a question which needs further investigations. The outcome of compound interaction could also depend on their concentrations, the effects might not be linear and additional experiments using a concentration-dependent response for each bioactive alone and in combination are warranted to better establish any possible additive/synergistic effects on human adipocytes. However, our data suggest interactions between bioactives on fat cell function, which is relevant for in vivo physiology and nutrition.

Publications from murine 3T3-F442A and -L1 fat cells as well as isolated primary human adipocytes have demonstrated increased adiponectin secretion after DHA treatment.[18,19,53] However, it should be noted that fetal bovine serum, an additive which may protect the cells from ROS exposure, was included in the cell culture medium in all these previous studies. This may explain why higher concentrations of DHA could be used in those experiments. Other possible reasons for the discrepancies may be treatment duration or delivery as in vitro cell systems seem to be sensitive to this compound. The effect of PI on adiponectin secretion was unexpected, but could be due to its effect on PPARγ reported previously.[54] On the other hand we did not observe any effects of the compounds on fat cell differentiation/lipid accumulation, which speaks against their possible role regulating important adipogenic factors at least in in vitro cell system.