Fig 2: Phenotypic consequences of lentiviral tumor necrosis factor–like cytokine 1A (TL1A) overexpression in young human umbilical vein endothelial cells (HUVEC). (A) Lentiviral overexpression of TL1A was confirmed 7 days after transduction of young HUVEC in passage 7 by Western blot using antibodies specific for TL1A. Lentiviral empty vector (Mock) was used as a control. (B) Six days after transduction, morphology of transduced cells was observed by phase contrast microscopy. (C) Population doubling time was assessed via CASY cell counting. (D) Six days after transduction, the amount of cells in S phase was determined using a bromodeoxyuridine assay. 4′-6-diamidino-2-phenylindole was used to stain cell nuclei. Pictures shown are representative of at least three experiments. (E) Four and six days after lentiviral transduction of young HUVEC with either empty vector (Mock) or TL1A, the cells were stained with Annexin V and monitored for apoptosis by FACS. (F) Seven days after lentiviral transduction of HUVEC with either empty vector (Mock) or TL1A, cells were fixed and stained for senescence-associated β-galactosidase (SA-β-gal). Pictures are representative of at least three experiments. (G) Five days after lentiviral transduction of young HUVEC with either empty vector (Mock) or TL1A, 5 × 104 cells each were seeded on 300 μL Matrigel in a 24-well plate. On the next day, tube formation was documented by phase contrast microscopy. Three experiments were performed and representative pictures are shown.

Fig 3: Upregulation of tumor necrosis factor–like cytokine 1A (TL1A) messenger RNA (mRNA) in replicative senescent circulating endothelial progenitor cells (CEP) as well as in elderly donors. (A) CEP of a young donor (Y2 = 35 years) were serially passaged until they reached senescence. mRNA was prepared and subjected to cDNA microarray analysis. Microarray results were normalized, and the regulation for TNFSF15 probe set (TL1A) is shown. (B) In addition, CEP of young (Y1 = 30 years) and old donors (O1 = 70 years and O2 = 79 years) were isolated and analyzed by cDNA microarray. (C) The mRNA expression observed by cDNA microarray analysis was confirmed by real-time-quantitative-PCR (RTq-PCR) using primers specific for the detection of TL1A. Results are shown as fold change compared with “young” CEP regarding replicative age or donor age, respectively. (D) CEP derived from three young donors (Y3, Y4, and Y5) were transduced with either control lentivirus or three lentiviral knockdown vectors targeting TL1A (pLKO1, pLKO99). RTq-PCR showing the lentiviral knockdown of TL1A in CEP was performed 8 and 10 days after transduction at the indicated passage numbers. Results are shown as fold change compared with control. (E) Quantification of senescence-associated β-galactosidase (SA-β-gal)–positive cells after knockdown of TL1A. CEP of young donors (Y3, Y4 and Y5) were fixed and stained for SA-β-gal 7 or 9 days after transduction. Phase contrast microscopic pictures were taken at 400× magnification. The percentage of SA-β-gal–positive cells was calculated by counting about 1,500 cells per experiment. (F) CEP were transduced with lentiviral vectors expressing green fluorescent protein (GFP) or TL1A, and cell extracts were subjected to Western blot using GFP or TL1A-specific antibodies, respectively. Wild-type CEP were used as a control. (G) Six days after transduction, cells were fixed and stained for SA-β-gal. Pictures are representative of at least three experiments.

Fig 4: Identity and biologic activity of tumor necrosis factor (TNF)–like cytokine 1A (TL1A) fragments. (A) Concentrated cell supernatant of TL1A overexpressing U-2OS cells was deglycolysed with N-glycosidase F and compared with the molecular weight of Escherichia coli recombinant TL1A fragments TL1AL72-L251 and TL1AV84-L251 by Western blot using TL1A-specific antibody. (B) Sites for ADAM-mediated cleavage of TL1A. Cleavage between alanine71 and leucine72 (indicated by an arrow) would lead to the described TL1AL72-L251 (21). The newly described TL1AV84-L251 would be cleaved between glutamine83 and valine84. Putative transmembrane region is marked in dark grey. TNF-homology domain is marked in light grey. Both domains were predicted by the Human Protein Reference Database. (C) About 50,000 young human umbilical vein endothelial cells (HUVEC) in passage 7 were seeded in six-wells treated with 20 μg of TL1AL72-L251 or TL1AV84-L251 for 24 or 36 hours and subjected to Annexin V FACS (left) and cell counting (right), respectively. *p < .05 and **p < .01, phosphate-buffered saline (PBS) versus TL1AV84 (Student’s t test; n = 3). (D) TL1A fragments induced interferon (IFN)-γ production in human peripheral blood mononuclear cells (PBMC). (Left) Percentage of IFN-γ-expressing PBMC. (Right) Percentage of IFN-γ-expressing natural killer cells. *p < .05 and **p < .01, IL-12+IL-18 versus L72, V84 and IL-15 (Student’s t test; n = 3).

Fig 5: Overview of tumor necrosis factor (TNF)–like cytokine 1A (TL1A) fragments. TL1A fragments used in this communication and in the previous literature are shown. On top, the schematic drawing of the TL1A full-length sequence with the putative transmembrane domain and the TNF-homology domain (gray) is shown. Below the recombinant fragments used in different studies are indicated with their N-terminal amino acid and their biologic function.