Keratinocyte-induced costimulation of human T cells through CD6 - but not CD2 - activates mTOR and prevents oxidative stress.

In psoriasis and other inflammatory skin diseases, keratinocytes (KCs) secrete chemokines that attract T cells, which, in turn, cause epidermal hyperplasia by secreting proinflammatory cytokines. To date, it remains unclear whether skin-homing T cells, particularly memory T cells, can also be activated by direct cell contact with KCs. In this study, we demonstrated the ability of primary human KCs to activate human memory T cells directly by transmitting costimulatory signals through the CD6/CD166/CD318 axis. Interestingly, despite being negative for CD80/CD86, KCs initiate a metabolic shift within T cells. Blockade of the CD6/CD166/CD318 axis prevents mammalian target of rapamycin activation and T cell proliferation but promotes oxidative stress and aerobic glycolysis. In addition, it diminishes formation of central memory T cells. Importantly, although KC-mediated costimulation by CD2/CD58 also activates T cells, it cannot compensate for the lack of CD6 costimulation. Therefore, KCs likely differentially regulate T cell functions in the skin through two distinct costimulatory receptors: CD6 and CD2. This may at least in part explain the divergent effects observed when treating inflammatory skin diseases with antibodies to CD6 versus CD2. Moreover, our findings may provide a molecular basis for selective interference with either CD6/CD166/CD318, or CD2/CD58, or both to specifically treat different types of inflammatory skin diseases.

Extracellular Vesicle Subtypes Released From Activated or Apoptotic T-Lymphocytes Carry a Specific and Stimulus-Dependent Protein Cargo.

Extracellular vesicles (EVs) are released from nearly all mammalian cells and different EV populations have been described. Microvesicles represent large EVs (LEVs) released from the cellular surface, while exosomes are small EVs (SEVs) released from an intracellular compartment. As it is likely that different stimuli promote the release of distinct EV populations, we analyzed EVs from human lymphocytes considering the respective release stimuli (activation Vs. apoptosis induction). We could clearly separate two EV populations, namely SEVs (average diameter <200 nm) and LEVs (diameter range between 200 and 1000 nm). Morphology and size were analyzed by electron microscopy and nanoparticle tracking analysis. Apoptosis induction caused a massive release of LEVs, while activated T-cells released SEVs and LEVs in considerably lower amounts. The release of SEVs from apoptotic T-cells was comparable with LEV release from activated ones. LEVs contained signaling proteins and proteins of the actin-myosin cytoskeleton. SEVs carried cytoplasmic/endosomal proteins like the 70-kDa heat shock protein 70 (HSP70) or tumor susceptibility 101 (TSG101), microtubule-associated proteins, and ubiquitinated proteins. The protein expression profile of SEVs and LEVs changed substantially after the induction of apoptosis. After apoptosis induction, HSP70 and TSG101 (often used as exosome markers) were highly expressed within LEVs. Interestingly, in contrast to HSP70 and TSG101, gelsolin and eps15 homology domain-containing protein 3 (EHD3) turned out to be specific for SEVs irrespective of the stimulus causing the EV release. Finally, we detected several subunits of the proteasome (PSMB9, PSMB10) as well as the danger signal HMGB1 exclusively within apoptotic cell-released LEVs. Thus, we were able to identify new marker proteins that can be useful to discriminate between distinct LEV subpopulations. The mass spectrometry proteomics data are available via ProteomeXchange with identifier PXD009074.