Targeting 6-phosphogluconate dehydrogenase in the oxidative PPP sensitizes leukemia cells to antimalarial agent dihydroartemisinin

Abstract

The oxidative pentose phosphate pathway (PPP) is crucial for cancer cell metabolism and tumor growth. We recently reported that targeting a key oxidative PPP enzyme, 6-phosphogluconate dehydrogenase (6PGD), using our novel small-molecule 6PGD inhibitors Physcion and its derivative S3, shows anticancer effects. Notably, humans with genetic deficiency of either 6PGD or another oxidative PPP enzyme, glucose-6-phosphate dehydrogenase, exhibit non-immune hemolytic anemia upon exposure to aspirin and various antimalarial drugs. Inspired by these clinical observations, we examined the anticancer potential of combined treatment with 6PGD inhibitors and antimalarial drugs. We found that stable knockdown of 6PGD sensitizes leukemia cells to antimalarial agent dihydroartemisinin (DHA). Combined treatment with DHA and Physcion activates AMP-activated protein kinase, leading to synergistic inhibition of human leukemia cell viability. Moreover, our combined therapy synergistically attenuates tumor growth in xenograft nude mice injected with human K562 leukemia cells and cell viability of primary leukemia cells from human patients, but shows minimal toxicity to normal hematopoietic cells in mice as well as red blood cells and mononucleocytes from healthy human donors. Our findings reveal the potential for combined therapy using optimized doses of Physcion and DHA as a novel antileukemia treatment without inducing hemolysis.

Figure 1. Knockdown of G6PD or 6PGD sensitizes human K562 leukemia cells to anti-malarial agent DHA

(A) Purified 6PGD (left) and G6PD (right) are assayed for 6PGD and G6PD enzyme activity, respectively, in the presence of increasing concentrations of Physcion. (B) K562 cells were treated with increasing concentrations of Physcion for 36 hours, followed by cell viability assay assessed by cell numbers. (C) K562 cells were treated with increasing concentrations of DHA for 36 hours, followed by cell viability assay assessed by cell numbers. (D) Results of Western blot experiments detecting 6PGD (upper) and G6PD (lower) protein levels in 6PGD (upper) and G6PD (lower) stable knockdown cells, respectively. (E) 6PGD (left) or G6PD (right) stable knockdown K562 cells and control cells harboring an empty vector were tested for enzyme activity and oxidative PPP flux rate. (F) 6PGD stable knockdown K562 cells and control vector cells were treated with increasing concentrations of DHA for 36 hours, followed by cell viability assay. The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (*0.01 < p < 0.05; (**0.001 < p < 0.01; ns, not significant). The results of one representative experiment from at least two independent experiments are shown.

Figure 2. Treatment with 6PGD inhibitor Physcion with DHA results in synergistic inhibition of diverse human leukemia cells

(A) Left: K562 cells were treated with increasing concentrations of DHA in the presence or absence of Physcion for 36 hours, followed by cell viability assay. Right: The combination index (CI) plot of results shows the CI for combined treatment with Physcion + DHA in K562 cells. CI<1 denotes synergy. (B) K562 cells were treated with increasing concentrations of DHA in the presence or absence of Physcion for 24 hours, followed by apoptosis assay. The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (**0.001 < p < 0.01). (C) Upper panels: KG1a (left), Molm14 (middle) and HEL (right) cells were treated with increasing concentrations of DHA in the presence or absence of Physcion for 36 hours, followed by cell viability assay. Lower panels: The combination index plots show the CI for combined treatment with Physcion + DHA in KG1a (left), Molm14 (middle) and HEL (right) cells. CI<1 denotes synergy. The results of one representative experiment from at least two independent experiments are shown.

Figure 3. Combined treatment with Physcion + DHA does not signal through COX2

(A) K562 cells treated with or without DHA in the presence or absence of Physcion were tested for ROS levels. K562 cells treated with or without NAC for 36 hours, followed by ROS assay (B) or cell viability assay (C). (D) The mRNA (upper) and protein expression (lower) levels of COX2 in K562 cells treated with increasing concentrations of DHA in the presence or absence of Physcion were determined by RT-PCR and Western blot assays, respectively. (E) The effect of combined treatment with Physcion and COX2 inhibitor Celecoxib on K562 cell viability was examined. The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (*0.01 < p < 0.05; (**0.001 < p < 0.01; ns, not significant). The results of one representative experiment from at least two independent experiments are shown.

Figure 4. Combined treatment with Physcion + DHA activates AMPK

(A–B) K562 cells were treated with increasing concentrations of DHA in the presence or absence of Physcion for 4 hours, followed by (A) Western blot to detect protein expression and phosphorylation levels of both AMPK (T172) and downstream substrate ACC1 (S79); (B) Lipid biosynthesis assay to detect de novo lipogenesis. (C) K562 cells were treated with increasing concentrations of DHA in the presence or absence of Physcion alone or combined treatment with Physcion and AMPK inhibitor Compound C for 36 hours, followed by cell viability assay. (D) Left: AMPK stable knockdown K562 cells and control vector cells were treated with increasing concentrations of DHA in the presence or absence of Physcion for 36 hours, followed by cell viability assay. Right: Result of Western blot experiment detecting AMPK protein levels in AMPK stable knockdown cells compared to control vector cells. The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (*0.01 < p < 0.05; (**0.001 < p < 0.01; ns, not significant). The results of one representative experiment from at least two independent experiments are shown.

Figure 5. Combined treatment with Physcion and AMPK activator results in synergistic inhibition of diverse leukemia cells

(A) Left: K562 cells were treated with increasing concentrations of AMPK activator A769662 in the presence or absence of Physcion for 48 hours, followed by cell viability assay. The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (*0.01 < p < 0.05; (**0.001 < p < 0.01; ns, not significant). Right: The combination index plot shows the CI for combined treatment with Physcion and A769662 in K562 cells. CI<1 denotes synergy. (B) K562 cells were treated with increasing concentrations of AMPK activator A769662 in the presence or absence of Physcion for 24 hours, followed by apoptosis assay. (C) Upper panels: KG1a (left), Molm14 (middle) and HEL (right) cells were treated with increasing concentrations of A769662 in the presence or absence of Physcion for 48 hours, followed by cell viability assay. Lower panels: The combination index plots show the CI for combined treatment with Physcion + DHA in KG1a (left), Molm14 (middle) and HEL (right) cells. CI<1 denotes synergy.

Figure 6. Combined treatment with S3 and DHA attenuates tumor growth in xenograft nude mice injected with K562 cells

(A) Histological morphology of hematoxylin-eosin stained tissue sections of representative K562 xenograft nude mice in control DMSO, single agents DHA or S3, and S3+DHA treated groups. Mice were treated with vehicle control, S3 alone (5 mg/kg/day), DHA alone (2.5 mg/kg/day), or S3+DHA for 15 days. The vital organs were collected for histopathological analysis. Histopathologic tissue sections (kidney, lung, liver, spleen, and bone marrow) from representative nude mice stained with hematoxylin-eosin did not reveal significant differences among control DMSO, single agents DHA or S3, and S3+DHA treated groups. Images were analyzed and captured using ImageScope software (Aperio Technologies Inc.) without any additional or subsequent image processing (high power images are 20×; low power images are 4×). Scale bars are indicated. (B) Tumor growth curve in xenograft nude mice injected with K562 cells were compared among the group of mice treated with control DMSO, single agents DHA or S3, or S3+DHA. (C) Tumor size in xenograft nude mice injected with K562 cells were compared between groups of mice treated with control DMSO, single agent Physcion derivative S3 or DHA, and combined therapy using S3 and DHA (n=6/group). p values were determined by a two way ANOVA test (**0.001 < p < 0.01; ns, not significant). (D) Dissected tumors in representative nude mice treated with control DMSO, single agents DHA or S3, or S3+DHA. (E) Western blot results detecting protein expression and phosphorylation (T172) levels of AMPK in tumors harvested from K562 xenograft mice with distinct drug treatment is shown.

Figure 7. Combined treatment with Physcion + DHA inhibits primary human leukemia cells with minimal toxicity

(A) Effects of combined treatment with Physcion + DHA for 48 hours on cell viability were examined in human primary leukemia cells isolated from the PB of representative APL, AML and CML patients. (B) Effect of Physcion + DHA treatment on phosphorylation levels of AMPK (T172) was examined by Western blot using primary leukemia cells isolated from the PB of a representative AML patient. (C) Combined treatment with Physcion + DHA shows no toxicity in treatment (48 hours) of isolated mononucleocytes (left panels) and red blood cells (right panels) from peripheral blood samples of two representative healthy human donors (#61 on the left and #62 on the right). The error bars represent mean values ±SD from three technical replicates of each sample. p values were determined by two-tailed Student’s t test (*0.01 < p < 0.05; (**0.001 < p < 0.01; ns, not significant).