PROJECT SUMMARY AND ABSTRACT
Rationale: Despite novel chemotherapies have significantly improved the prognosis of melanoma patients, drug
resistance occurs inevitably and tumor progression becomes inexorable, presenting an unmet and urgent need for
novel therapeutic intervention. Most melanomas harbor oncogenic BRAFV600E and are responsive to the
BRAFV600E-specific inhibitor dabrafenib and the MEK inhibitor trametinib; however, patients eventually acquire
resistance mainly due to MAPK reactivation and/or the counteraction by phosphoinositide 3-kinase (PI3K). While
substantial efforts have been made to mitigate therapy resistance using PI3K inhibitors; there has been a lack
of success as no PI3K inhibitors are currently capable of serving for this purpose. PI3K has four functionally
divergent catalytic kinases PI3Ka/ß/d/¿. Drugs that block all four kinases or individual PI3Ka/d/¿ have shown clinical
benefit, but elicited notable metabolic or immune-related toxicities. Such clinical challenge in drugging PI3Ka/d/¿ has
fueled interest in PI3Kß therapies; however, PI3Kß ATP competitive inhibitors GSK2636771, TGX-221, and AZD6482
are not clinically effective, notwithstanding their promising preclinical anti-cancer activities. This is perhaps because
ATP competitive inhibitors interact with well-conserved residues at motifs pivotal for executing kinase activity in
PI3Ka/ß/d/¿, making them not as PI3Kß-selective as desired. Rising to this challenge, we have recently identified an
18-residue motif dubbed ß18 that occurs only in PI3Kß, but not in PI3Ka/d/¿. Selectide-18 bearing ß18 and a cell-
penetrating peptide distorts PI3Kß complexes, inactivates PI3Kß, slows down the growth of BRAFV600E melanomas
expressing high levels of PI3Kß and low levels of the PI3K-antagonizing phosphatase PTEN (designated as
PI3Kßhyper), and sensitizes therapy-resistant PI3Kßhyper melanoma cells to dabrafenib/trametinib. Our further in-silico
analyses have identified Selectide-9, a short version of Selectide-18 which only contains 9 residues derived from the
S18 motif, but still retains the same cytotoxic activities as Selectide-18. Hypothesis: We hypothesize that Selectide-
9 selectively degrades and inactivates PI3Kß, thereby overcoming therapy resistance in PI3Kßhyper melanoma.
Specific Aims: We will use approaches established in the PI’s laboratory to test the above hypothesis. In Aim 1, we
will test how Selectide-9 degrades and inactivates PI3Kß selectively using biochemical and imaging techniques. In
Aim 2, we will test whether Selectide-9 is superior to ATP competitive inhibitors in overcoming therapy resistance
using two human BRAFV600E patient-derived xenograft lines in immunodeficient mice. The WM4701-6673 line has
low levels of PI3Kß and the recurrent/therapy-resistant WM4701-7329 tumor is derived from WM4701-6673 treated
with a BRAFV600E inhibitor. We will determine if WM4701-6673 is sensitive, whilst WM4701-7329 is resistant, to
dabrafenib and trametinib and if Selectide-9 is more potent than GSK2636771 in restoring the sensitivity to dabrafenib
and trametinib. Moreover, we will identify gene signatures defining therapy-resistant PI3Kßhyper melanomas using
single-cell RNA sequencing. Impact: Our work will lay the groundwork to support future studies to gain deeper
insights into therapy resistance in melanoma and identify an innovative approach to circumventing this resistance.