Fig. 8

Targeting TAK1 restricts in vivo SASP and minimizes drug resistance acquired from the treatment-damaged TME. a Experimental design of severe combined immunodeficient (SCID) mouse-based in vivo studies. Two weeks after tumour implantation and stable uptake, mice received either single or combinational agents administered as metronomic treatments composed of several cycles. b Comparative statistics of tumour volumes. PC3 cells were implanted alone or with stromal cells subcutaneously to animals, which were then subject to cyclic treatments. Tumour volumes were measured at the end of an 8-week preclinical regimen. c Representative images of tumour-bearing animals in the preclinical trial. Digital signals were proportional to in vivo luciferase activities measured by an IVIS device. d Mice were sacrificed upon presence of advanced bulky diseases. Survival duration was calculated from the time of tissue recombinant injection until the day of death. Data compared with log-rank (Mantel–Cox) test. e Transcript assessment of several canonical SASP factors expressed in stromal cells isolated from PC3 tumours. Animals that had both stromal and cancer cells in the tumour foci were selected for analysis, with stromal cells acquired by LCM. f Graphical summary. DNA damage triggers an acute response of stromal cells, during which Zscan4 is expressed via the ATM-TRAF6-TAK1 axis and translocates to the nucleus. Zscan4 promotes the expression of a subset of ASAP-associated factors through NF-κB signaling, forming the first positive feedback loop. Further, TAK1 collaterally activates p38, a kinase that subsequently engages the PI3K/Akt/mTOR pathway. mTOR subsequently activates the NF-κB machinery both directly via interaction with IKKα and indirectly by promoting the translation of IL-1α, a cytokine that strengthens TAK1 phosphorylation in the chronic SASP. Together, several feedforward mechanisms favor the SASP development until its culmination in stromal cells, which confers substantial resistance on surviving cancer cells in the damaged TME and enables disease recurrence post-therapy. Data in all bar plots are shown as mean ± SD and representative of 3 biological replicates. For animal treatments, n = 10/group. Data in b, e were analyzed by Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ^P > 0.05