Summary /

Cancer cells are known to exist in highly phosphorylated states and certain peptide motifs with phosphorylated tyrosines are known PKM2 inhibitors (1Christofk 2008, Ikeda 2000). Binding of these phosphorylated peptides results in 1,6-FBP extrusion from the allosteric site. Use of a known phosphorylated peptide motif that binds to PKM2 (pM2tide) causes a shift in abundance of the more active tetrameric state to the less active monomeric state compared to the same, non-phosphorylated peptide (M2tide) (Figure 2a). ML265 was able to maintain potent activation of PKM2 in the presence of both M2tide and pM2tide (Figure 2b). Treating cells with pervanadate, a known phosphatase inhibitor, causes increased amounts of phosphorylated proteins and decreased PKM2 activity (1Christofk 2008). To test whether ML265 could activate PKM2 in this cellular context, we determined pyruvate kinase activity in cell lysate of both pervanadate and non-pervanadate treated cells (Figure 2c). Remarkably, at 1 μM, ML265 was able to activate PKM2 regardless of pervanadate treatment. Cantley and co-workers have shown that RNAi mediated replacement of PKM2 with PKM1 results in higher pyruvate kinase activity and reduces cellular proliferation in a hypoxic environment (2Christofk 2008).

Figure 2. Figure 4. ML265 activates PKM2 in the presence of phosphorylated peptides. a) Size exclusion chromatography of recombinant PKM2 after incubation with an M2 binding peptide motif (M2tide) or phosphorylated version of the same peptide motif (pM2tide). b) ML265’s ability to activate PKM2 in the presence of M2tide or pM2tide. c) Pyruvate kinase activity in A549 cell lysates following treatment with or without pervanadate in the presence of either DMSO or ML265.

Summary /

ML265 activates PKM2 via an increase in PEP affinity (Jiang 2010). In the absence of activator, PKM2 has a relatively low affinity for PEP (KM ~ 1.5 mM), but treatment with either 1,6-FBP or ML265 resulted in a significant shift in KM to ~0.08 and ~0.09 mM respectively (Figure 3a). In accordance with our previous studies, this effect was not seen with respect to ADP affinity (Figure 3b). The homotetramer of PKM2 is the most active form and its endogenous activator, 1,6-FBP is thought to induce this subunit association. To investigate this hypothesis, we used size exclusion chromatography to separate the monomers from tetramers and then determined the effect of 1,6-FBP on PKM2 activity (Figure 4a). As expected, the untreated tetramer had considerably higher kinase activity than the untreated monomer. Addition of 100 μM 1,6-FBP had little effect on the tetramer activity, but resulted in ~70% increase in activity for the monomer (Figure 4b). In an effort to understand whether our activators could induce the more active tetrameric state, we incubated ML265 with recombinant PKM2 and then used sucrose gradient ultra-centrifugation followed by SDS-PAGE to separate the protein. As can be seen in Figure 4c, ML265 caused a significant shift to tetrameric species, similar to PKM1 and consistent with the affect seen with 1,6-FBP on PKM2.

Figure 3. PKM2 activation (scaled from initial velocities) as a function of a) PEP and b) ADP concentration in the presence of 10μM 1,6-FBP (gray circles), ML265 (black circles) or DMSO control (white circles).

Figure 4. a) Size exclusion chromatography was used to separate monomeric and tetrameric PKM2 species and the fractions were b) assayed for pyruvate kinase activity with and without 1,6-FBP (100 μM). c) Sucrose gradient ultracentrifugation profiles of recombinant PKM1 or PKM2 and effects of 1,6-FBP or ML265 on PKM2 subunit stoichiometry.

Summary /

A high resolution X-ray structure of ML265 bound to PKM2 (Figure 5) is generated. The structure showed that two activators and four 1,6-FBP molecules bind per tetramer. The two equivalents of ML265 bind at the dimer-dimer interface and are completely buried within this interfacial pocket. ML265 is accommodated through van der Waals interactions and water mediated hydrogen bonds to the pocket lining residues (Figure 5b).

Figure 5. a) X-ray structure of the PKM2 tetramer with two equivalents of ML265 and four equivalents of 1,6-FBP bound. ML265 binds at the dimer-dimer interface, termed A-A’. b) The sulfoxide and carbonyl group of ML265 form water mediated hydrogen bond networks the protein, while the aniline moiety forms a direct hydrogen bond with D354.

Summary /

ML265 gave superior plasma concentrations (vs the prior art PKM2 activator ML202) that persisted at higher levels over the 24 hour study (Figure 6). ML265 also displayed good oral bioavailability, low clearance, a long half-life and good volume of distrubtion (Table 1). These characteristics were deemed appropriate for use in the mouse xenograft efficacy model.

Figure 6. Plasma concentration-time profiles of a) ML202 and b) ML265 in male Balb/c mice following intravenous, intraperitoneal and oral administration.

Table 1. In vitro ADME profile for ML202 and ML265.

Summary /

ML265 was advanced into a 7-week H1299 mouse xenograft study using immunocompromised (nu/nu) mice. The mice were randomly divided into two cohorts with one group receiving the established 50 mg/kg BID of ML265, the other cohort received only vehicle. Blood counts, serum chemistries and histological examination of various tissues were followed to determine if there were overt toxicities associated with ML265 treatment (Table 2). No significant aberrations were observed in these toxicity measures when compared to the vehicle treated mice. ML265 was able to delay tumor formation compared to vehicle-treated mice and the tumors that did form were significantly smaller (Figure 7). Further examination of the tumors revealed that ML265 was detectable signifying that the tumor was being exposed to the activator.

Table 2. Serum chemistries and blood counts of mice treated with ML265. *WBC: White Blood Cells. §PCV: Packed Cell Volume.

Figure 7. H1299 cells were injected subcutaneously into nu/nu mice which were subsequently randomly divided into two cohorts, one given vehicle and the other 50 mg/kg BID ML265 throughout the duration of the experiment. a) Injection sites were monitored for tumor emergence [p value calculated by log rank (Mantel-Cox) test]. b) At 52 days, the tumors were dissected and final tumor weights were measured. Mean tumor weights ± s.e.m. are shown and p value was calculated by unpaired Student’s t-test.