Here, we show that Lyn promotes the survival of nutrient-stressed human GBM cell lines and primary human GSCs by promoting autophagy and proliferation as well as inhibiting cell death. The mechanism underlying the pro-autophagy effect of Lyn in the nutrient-deprived U87 GBM tumor cells and in the growth factor-deprived GSCs in vitro
correlated with increased activation of AMPK and inhibition of the mTORC1 complex. The decrease in phosphorylation of S6 kinase in the GBM cells expressing CA-Lyn and the increase in the cells expressing DN-Lyn was consistent with the effects of these constructs on autophagy. Glucose, growth factor and amino acid deprivation activate AMPK thereby relieving mTORC1 inhibition of autophagy 
. Consistent with this, nutrient-deprivation was associated with an increase in pAMPK in the GBM cells and GSCs expressing CA-Lyn. Although in most cell types the PI3K/Akt axis signals to activate mTORC1 inhibition of autophagy, the lack of change in Akt activity in the time frame in which autophagy was detectable (days 5 and 7) suggests that it may not be involved in Lyn regulation of mTORC1 activity. Currently, we cannot rule out the possibility that mTORC2 plays a role as there is evidence that it also can regulate autophagy in glioma cells 
SFK activity was necessary for the prosurvival effect of Lyn in nutrient-deprivation conditions, which suggests that Lyn kinase activity is necessary. However, we have not yet identified its substrate. Although multiple SFK substrates have been identified, none are known to positively regulate autophagy. The tyrosine phosphorylation of the known SFK substrates, FAK(Y397 and Y925) (Fig. S3C
), cortactin or N-WASP (WM Liu and CL Gladson, unpublished observations) in the U87-CA-Lyn or U87-DN-Lyn cells was not affected after 5 days of nutrient deprivation.
The manipulation of Lyn activity also affected apoptosis and proliferation of nutrient-deprived cells. Expression of CA-Lyn resulted in a significant reduction in cell death and significant increase in EdU incorporation whereas expression of DN-Lyn resulted in a significant increase in cell death but no change in EdU incorporation. Although miRNA-205 has been shown to inhibit mRNA and protein expression of Lyn, c-Src, and c-Yes in A498 cells resulting in G0/G1 cell cycle arrest and apoptosis 
, and Lyn specifically reduced expression of miRNA-181b that represses the anti-apoptotic protein Mcl1 
, we did not detect a change in miRNA expression (≥ or ≤1.5-fold) in the U87-CA-Lyn or U87-DN-Lyn cells as compared to U87-LV cells using the Human Cancer miRNA PCR Array (MAH-102A, Qiagen) (WM Liu and CL Gladson, unpublished data). Similarly, although SFKs can promote survival through the FAK/p130CAS or PI3K/Akt pro-survival pathways 
, neither FAK nor Akt activity was altered in the nutrient-deprived GBM cells expressing CA-Lyn or DN-Lyn. It is possible that the cellular localization of CA-Lyn 
, the cell surface binding partner of Lyn 
, or Lyn phosphorylation of the growth arrest and DNA damage protein 34 (GADD34) 
contribute to the anti-death effect in GBM but this remains to be elucidated.
Our in vivo
studies in which GBM cells expressing DN-Lyn were propagated in the nude mouse brain showed a significant reduction in tumor size that was associated with increased tumor cell death, suggesting Lyn is important for tumor survival in vivo
. This is consistent with a prior report that dasatinib inhibits malignant glioma growth in a mouse model 
. However, dasatinib used as a single agent in GBM patients who had failed therapy with Avastin® did not result in either a partial or complete response 
. The reason for this is unclear but may be related to the diverse effects of dasatinib, e.g.
, in addition to blocking the function of all SFKs it also targets the cytoplasmic non-receptor tyrosine kinase c-Abl, which is necessary for apoptosis of brain microvessel endothelial cells induced by an integrin αvβ3/αvβ5 inhibitor and by latrunculin 
. Currently, it is not possible to conclude that nutrient deprivation is the trigger of the effects of Lyn in vivo
, but a focal lack of nutrients does occur in GBM 
Elevation of Lyn is associated with resistance to therapy. In CML, it is associated with the development of resistance to imatinib mesylate, a Bcr-Abl tyrosine kinase inhibitor 
. It also is found in AML refractory to therapy 
, and in non-small cell lung cancer cells resistant to cetuximab (an anti-EGFR antibody) where Lyn promotes EGFR nuclear translocation 
. In addition, Lyn cooperates with a CD44-variant receptor in promoting chemoresistance in colon cancer cells 
. In cancers refractory to therapy, autophagy can be a pro-survival response to the stress of chemotherapeutic agents 
; however, autophagy can lead to cell death under certain circumstances (reviewed in 
). There is an evolving concept that cross-talk between autophagy and apoptosis signaling pathways occurs in cells, and not infrequently these two processes appear to be regulated in an inverse manner 
. Hydroxychloroquine, an inhibitor of autophagy, has been entered into a phase II clinical trial involving patients with multiple types of cancer including GBM (reviewed in 
), and blockade of autophagy has been shown to result in the sensitization of prostate cancer cells, multi-drug resistant v-Ha-ras transformed NIH-3T3 cells and other cells to a SFK inhibitor in nutrient-rich conditions 
In summary, we demonstrate the novel findings that Lyn promotes autophagy and proliferation, as well as inhibits cell death, in nutrient-deprived GBM cells in culture and in vivo
using a mouse model. Lyn promotes the malignant phenotype of GBM and multiple other cancers, including promotion of the epithelial-mesenchymal transition in breast cancer 
, as well as chemoresistance in certain cancers. Moreover, a Lyn-specific peptide that inhibits Lyn-dependent phosphorylation has been shown to decrease prostate cancer growth and induce apoptosis in vivo
. Thus, Lyn could be an important new therapeutic target for multiple cancers.