Collaboration with AstraZeneca provides new clue to predict which cancers will be most sensitive to Akt inhibitors
The majority of human cancers harbour mutations promoting activation of the Akt protein kinase. Because of this there are over 200 clinical trials listed on the NIH clinical trials website that have been initiated or planned to evaluate therapeutic efficacy of Akt inhibitors for the treatment of diverse human cancers.
Ability to predict which tumours will be most responsive to Akt inhibitors is an important question and of relevance to Akt inhibitor clinical trials. To tackle this, MRC-PPU PhD student Eeva Sommer, working in Dario Alessi's lab teamed up with the group of Barry Davies and his other colleagues (Darren Cross, Hannah Dry and Sylvie Guichard) in AstraZeneca.
The AstraZeneca group are involved in developing and clinical evaluation of one of the Akt inhibitors termed AZD5363 that is currently being evaluated in the clinic. They had identified a number of breast cancer cell lines that were sensitive to AZD5363 Akt inhibitor but others that were resistant. Mutational analysis revealed that both the sensitive and resistant cells possessed mutations that would be expected to result in the activation of the Akt kinase but no other patterns emerged that provide clues as to why cells were sensitive or resistant to the inhibitors. Immunoblot analysis also confirmed that in both resistant and sensitive cells the Akt pathway was highly activated and downstream targets of this pathway phosphorylated, therefore also not divulging why certain cells were sensitive and others resistant to Akt inhibitors.
There is a neglected, much less well studied group of protein kinases termed SGK, of which there are three isoforms termed SGK1, SGK2 and SGK3 that are highly related to Akt. Interestingly SGK isoforms are also activated by the same upstream machinery as Akt namely by PI 3-kinase-PDK1-mTORC2 system. SGK and Akt isoforms have overlapping substrate specificities and therefore have the potential to possess analogous functions. Akt inhibitors such as AZD5363 and MK-2206 that are being evaluated in cancer clinical trials do not inhibit SGK isoforms.
Eeva wondered whether elevated levels of one of the SGK isoforms activity might account for the differing sensitivities of breast cancer cells to Akt inhibitors. Expression of SGK isoforms is much more variable between cells and tissues than Akt, which might also explain why only a subset of tumour cells would possess elevated SGK activity.
Eeva's experiments revealed strikingly that the majority (but not all) of Akt inhibitor resistant breast cancer cells possess markedly elevated levels of SGK1 mRNA and protein. In contrast, none of the Akt inhibitor sensitive breast cancer cells displayed high levels of SGK1. Eeva also studied SGK2 and SGK3 but found no major differences in the expression of these enzymes between the sensitive and resistant cells.
Eeva next demonstrated that SGK1 knockdown markedly reduced proliferation and migration of Akt inhibitor resistant cells displaying high SGK1 levels but not sensitive cells. Furthermore, Eeva also found that inhibition of SGK1 activity by treating the Akt inhibitor resistant cells with an mTOR inhibitor markedly suppressed proliferation.
Previous work has suggested that SGK1 specifically phosphorylates the substrate N-myc Downstream-Regulated Gene 1 (NDRG1). Eeva therefore expected to only see elevated NDRG1 phosphorylation in the Akt inhibitor resistant cells that express high levels of SGK1. However, Eeva observed significant phosphorylation of NDRG1 in both the Akt inhibitor resistant and sensitive cells. Further experimentation revealed that in the Akt inhibitor sensitive cells, NDRG1 was in fact phosphorylated by Akt-as NDRG1 phosphorylation was potently suppressed by Akt inhibitors. In contrast, in the Akt inhibitor resistant cells displaying high SGK1 activity phosphorylation of NDRG1 was not impacted by treatment with Akt inhibitors.
Our data suggests that the trick to work out whether a tumour will be sensitive to an Akt inhibitor or not will be to monitor the effect that Akt inhibitors have on NDRG1 phosphorylation. This will be helped by the excellent monoclonal antibodies that can be used to recognise phosphorylated NDRG1 in clinical samples. The prediction that will need to be tested in future work is that if Akt inhibitors are found to reduce NDRG1 phosphorylation-then the tumour will likely be sensitive to Akt inhibitors. In contrast, the observation of high NDRG1 phosphorylation that is unaffected by Akt inhibitors-would suggest that SGK activity is elevated and that this cancer will be resistant to Akt inhibitors and may be better treated with mTOR inhibitors.
Our findings also suggest that development of SGK inhibitors or dual Akt/SGK inhibitors might be worthwhile for treatment of cancers displaying elevated SGK activity. To read a copy of our paper paper click here
Ability to predict which tumours will be most responsive to Akt inhibitors is an important question and of relevance to Akt inhibitor clinical trials. To tackle this, MRC-PPU PhD student Eeva Sommer, working in Dario Alessi's lab teamed up with the group of Barry Davies and his other colleagues (Darren Cross, Hannah Dry and Sylvie Guichard) in AstraZeneca.
The AstraZeneca group are involved in developing and clinical evaluation of one of the Akt inhibitors termed AZD5363 that is currently being evaluated in the clinic. They had identified a number of breast cancer cell lines that were sensitive to AZD5363 Akt inhibitor but others that were resistant. Mutational analysis revealed that both the sensitive and resistant cells possessed mutations that would be expected to result in the activation of the Akt kinase but no other patterns emerged that provide clues as to why cells were sensitive or resistant to the inhibitors. Immunoblot analysis also confirmed that in both resistant and sensitive cells the Akt pathway was highly activated and downstream targets of this pathway phosphorylated, therefore also not divulging why certain cells were sensitive and others resistant to Akt inhibitors.
There is a neglected, much less well studied group of protein kinases termed SGK, of which there are three isoforms termed SGK1, SGK2 and SGK3 that are highly related to Akt. Interestingly SGK isoforms are also activated by the same upstream machinery as Akt namely by PI 3-kinase-PDK1-mTORC2 system. SGK and Akt isoforms have overlapping substrate specificities and therefore have the potential to possess analogous functions. Akt inhibitors such as AZD5363 and MK-2206 that are being evaluated in cancer clinical trials do not inhibit SGK isoforms.
Eeva wondered whether elevated levels of one of the SGK isoforms activity might account for the differing sensitivities of breast cancer cells to Akt inhibitors. Expression of SGK isoforms is much more variable between cells and tissues than Akt, which might also explain why only a subset of tumour cells would possess elevated SGK activity.
Eeva's experiments revealed strikingly that the majority (but not all) of Akt inhibitor resistant breast cancer cells possess markedly elevated levels of SGK1 mRNA and protein. In contrast, none of the Akt inhibitor sensitive breast cancer cells displayed high levels of SGK1. Eeva also studied SGK2 and SGK3 but found no major differences in the expression of these enzymes between the sensitive and resistant cells.
Eeva next demonstrated that SGK1 knockdown markedly reduced proliferation and migration of Akt inhibitor resistant cells displaying high SGK1 levels but not sensitive cells. Furthermore, Eeva also found that inhibition of SGK1 activity by treating the Akt inhibitor resistant cells with an mTOR inhibitor markedly suppressed proliferation.
Previous work has suggested that SGK1 specifically phosphorylates the substrate N-myc Downstream-Regulated Gene 1 (NDRG1). Eeva therefore expected to only see elevated NDRG1 phosphorylation in the Akt inhibitor resistant cells that express high levels of SGK1. However, Eeva observed significant phosphorylation of NDRG1 in both the Akt inhibitor resistant and sensitive cells. Further experimentation revealed that in the Akt inhibitor sensitive cells, NDRG1 was in fact phosphorylated by Akt-as NDRG1 phosphorylation was potently suppressed by Akt inhibitors. In contrast, in the Akt inhibitor resistant cells displaying high SGK1 activity phosphorylation of NDRG1 was not impacted by treatment with Akt inhibitors.
Our data suggests that the trick to work out whether a tumour will be sensitive to an Akt inhibitor or not will be to monitor the effect that Akt inhibitors have on NDRG1 phosphorylation. This will be helped by the excellent monoclonal antibodies that can be used to recognise phosphorylated NDRG1 in clinical samples. The prediction that will need to be tested in future work is that if Akt inhibitors are found to reduce NDRG1 phosphorylation-then the tumour will likely be sensitive to Akt inhibitors. In contrast, the observation of high NDRG1 phosphorylation that is unaffected by Akt inhibitors-would suggest that SGK activity is elevated and that this cancer will be resistant to Akt inhibitors and may be better treated with mTOR inhibitors.
Our findings also suggest that development of SGK inhibitors or dual Akt/SGK inhibitors might be worthwhile for treatment of cancers displaying elevated SGK activity. To read a copy of our paper paper click here