The current emphasis is to understand first the role of lysosomal membrane protein 1,LAMP1, in cytotoxicity of human NK cells. To address this issue, RNAi was used to disrupt LAMP1 expression and assess its effects on lytic granule exocytosis and NK cell cytotoxic potential. We generated lentiviral constructs carrying shRNA targeting LAMP1 RNA and used them to infect a NK cell tumor line, YTS, creating a NK cell line with a stable knock-down (KD) of LAMP1 (80% decrease at protein level). To assess the role of LAMP1 in NK cell function, we investigated the effect LAMP1 silencing on NK cell cytotoxic activity against a susceptible tumor cell line, 721.221. We established that LAMP1 KD resulted in severe ( 75%) inhibition of NK cell activity. We obtained similar results using ex vivo isolated NK cells in which LAMP1 was knocked-down by introduction of LAMP1 siRNA. Our results demonstrate for the first time that LAMP1 plays critical role in NK cell cytolytic activity. Furthermore, we found that, in contrast to control RNAi-transduced cells, LAMP1 RNAi cells failed to deliver granzyme B to 721.221 target cells. Similarly, siRNA-mediated KD of LAMP1 in ex vivo NK cells also blocked the transfer of granzyme B from NK to tumor cells. Importantly, LAMP1 RNAi cells (both YTS and ex vivo NK cells) had normal level of granzyme B mRNA and protein, and the activity of granzyme B from LAMP1 RNAi cells was undistinguishable from the activity in control cells. LAMP1 RNAi cells had normal conjugation and were able to upregulate CD107b on the cell surface, excluding the possibility that the block of granzyme B delivery and, subsequently, cytotoxicity resulted from improper cell-cell adherence or inhibition of fusion of lytic granules with the plasma membrane. Our results suggested that the defect of granzyme B delivery could be due to problems with granule transport of lytic granules to the cell to cell contact site. Laser scanning confocal microscopy experiments showed that although LAMP1 RNAi cells were able to translocate lytic granules toward the immunological synapse, the granules were more dispersed around the MTOC and did not cluster efficiently at the immunological synapse, suggesting a possible defect in granule transport in LAMP1 RNAi cells. Indeed, we discovered that compared to control RNAi cells, granules in LAMP1 RNAi cells traveled shorter distances, had smaller displacement and moved slower, indicating that LAMP1 is important for the proper granule movement along cytoskeletal tracks. Using cell fractionation and analytical ultracentrifugation, we found that, when compared to control RNAi cells, lytic granules isolated from LAMP1 KD cells had less p150glued, an important component of the microtubule motor protein complex-dynein/dynactin. The decreased recruitment of motor proteins to lytic granules provides an explanation for the defects in the movement of the granules caused by the disruption of LAMP1 expression. More importantly, we also found that compared to control RNAi cells, silencing of LAMP1 resulted in 30-40% decrease of perforin fluorescence, despite normal level of perforin mRNA. Analysis of intracellular perforin level by flow cytometry confirmed the specific decrease of perforin in ex vivo NK and YTS cells with LAMP1 KD. Imaging and lytic granule isolation studies led us to discovery that LAMP1 RNAi cells had less perforin associated with lytic granules. Interestingly, the disruption of LAMP1 expression affected the amount of perforin, but not granzyme B , in the lytic granules, indicating that LAMP1 is involved in trafficking of some components of NK cell lytic machinery, namely perforin, to the lytic granules. We postulate that LAMP1 is not only a convenient marker of NK cell degranulation, but plays a crucial role in NK cell activity and disruption of LAMP1 function has detrimental effects on NK cell lytic activity.
