A TRAIL-TL1A paracrine network involving adipocytes, macrophages and lymphocytes induces adipose tissue dysfunction downstream of E2F1 in human obesity

Nitzan Maixner ¹ Tal Pecht ¹ Yulia Haim ¹ Nir Goldstein ¹ Tania Tarnovscki ¹ Idit F. Liberty ⁴ Boris Kirshtein ⁵ Vered Chalifa-Caspi ⁶ Rachel Golan ² Nava Bashan ¹ Mathias Bluher ³ Assaf Rudich ^1,6

¹Department of Clinical Biochemistry, Ben Gurion University, Israel
²Department of Epidemiology and Preventive Medicine, Ben Gurion University, Israel
³Department of Medicine, University of Leipzig, Germany
⁴Diabetes research and treatment center, Soroka University medical Center, Israel
⁵Surgical Department A, Soroka University Medical Center, Israel
⁶The National Institute of Biotechnology in the Negev, Ben Gurion University, Israel

Background: In human visceral adipose-tissue, elevated levels of the transcription factor E2F1 in adipocytes were recently associated with cardio-metabolically-at-risk obesity, demonstrated in both in vitro and clinical models (Autophagy, 11:2074,2015). However, given the ubiquitous nature of E2F1 the specific pathways mediating this effect are of particular interest, and these were not fully depicted yet.

Hypothesis: We hypothesized that the E2F1-driven dysfunctional profile is mediated by a complex network of downstream effectors.

Methods: An unbiased approach utilizing RNA-sequencing was used to assess the transcriptome of omental adipose-tissue samples from age-, sex- and BMI–matched patients, whose omental E2F1 protein expression was either high (E2F1^high) or low (E2F1^low). Significant pathways were identified using STRING© analysis. This was completed using in-vitro molecular approaches challenging the functional significance of identified pathways.

Results: Paired analysis revealed over 1300 differentially expressed genes consisting of several functionally significant pathways. Of those, stood out a network of TNF-superfamily factors, especially TRAIL(TNFSF10), TL1A(TNFSF15) and their receptors, later validated in a second cohort. Further ex-vivo experiments unraveled a feed-forward loop in which TL1A overexpression was the product of TRAIL-exposed adipose tissue components, specifically T-lymphocytes. In-vitro, TL1A induced insulin resistance and secretory malfunction in human adipocytes, while facilitating a lipid-accumulating profile in human macrophages, generating foam-cells which further aggravated tissue dysfunction. These phenomenon were also reflected in human cohorts, correlating TL1A with clinical parameters of cardio-metabolic risk determinants.

Conclusions: We propose an intra-adipose tissue paracrine loop, consisting of E2F1-driven TNF-SF activation, ultimately impinging on adipose tissue function and contributing to the generation of a systemic dys-metabolic phenotype.