Environmental changes feedback to climate through their impact on soil functions such as carbon (C) and nutrient sequestration. Abiotic conditions and the interactions between above- and belowground biota drive soil responses to environmental change but these (a)biotic interactions are challenging to study. Nonetheless, better understanding of these interactions would improve predictions of future soil functioning and the soil-climate feedback and, in this context, permafrost soils are of particular interest due to their vast soil C-stores. We need new tools to isolate abiotic (microclimate, chemistry) and biotic (roots, fauna, microorganisms) components and to identify their respective roles in soil processes. We developed a new experimental setup, in which we mimic thermokarst (permafrost thaw-induced soil subsidence) by fitting thawed permafrost and vegetated active layer sods side by side into mesocosms deployed in a subarctic tundra over two growing seasons. In each mesocosm, the two sods were separated from each other by barriers with different mesh sizes to allow varying degrees of physical connection and, consequently, (a)biotic exchange between active layer and permafrost. We demonstrate that our mesh-approach succeeded in controlling 1) lateral exchange of solutes between the two soil types, 2) colonization of permafrost by microbes but not by soil fauna, and 3) ingrowth of roots into permafrost. In particular, experimental thermokarst induced a ~60% decline in permafrost nitrogen (N) content, a shift in soil bacteria and a rapid buildup of root biomass (+33.2 g roots m^-2 soil). This indicates that cascading plant-soil-microbe linkages are at the heart of biogeochemical cycling in thermokarst events. We propose that this novel setup can be used to explore the effects of (a)biotic ecosystem components on focal biogeochemical processes in permafrost soils and beyond.