Mutant KRAS associated malic enzyme 1 expression is a predictive marker for radiation therapy response in non-small cell lung cancer

Abstract

Background: Advanced non-small cell lung cancer (NSCLC) is an aggressive tumor that is treated with a combination of chemotherapy and radiation if the patient is not a candidate for surgery. Predictive biomarkers for response to radiotherapy are lacking in this patient population, making it a non-tailored therapy regimen with unknown outcome. Twenty to 30 % of NSCLC harbor an activating mutation in KRAS that may confer radioresistance. We hypothesized that mutant KRAS can regulate glutamine metabolism genes in NSCLC and maintain tumor redox balance through transamination reactions that generate cytosolic NADPH via malic enzyme 1 (ME1), which may contribute to radioresistance.

Findings: A doxycycline-inducible mouse model of KRAS (G12D) driven NSCLC and patient data was analyzed from multiple publicly accessible databases including TCGA, CCLE, NCBI GEO and Project Achilles. ME1 expression was found to be mutant KRAS associated in both a NSCLC mouse model and human NSCLC cancer cell lines. Perturbing glutamine metabolism sensitized mutant KRAS, but not wild-type KRAS NSCLC cell lines to radiation treatment. NSCLC survival analysis revealed that patients with elevated ME1 and GOT1 expression had significantly worse outcomes after radiotherapy, but this was not seen after chemotherapy alone.

Conclusions: KRAS driven glutamine metabolism genes, specifically ME1 and GOT1 reactions, may be a predictive marker and potential therapeutic target for radiotherapy in NSCLC.

Fig. 1

Mutant KRAS is associated with ME1 and GOT1 expression in NSCLC. a Model of mutant KRAS-reprogrammed glutamine utilization (red). GLS1 = glutaminase 1; GLUD1 = glutamate dehydrogenase 1; GOT2 = mitochondrial aspartate aminotransferase; ASP = aspartate; GOT1 = cytosolic aspartate aminotransferase; OAA = oxaloacetate; MDH1 = malate dehydrogenase 1; ME1 = malic enzyme 1; GSR = glutathione disulfide reductase. b When fed doxycycline, the mice develop lung tumors that are dependent on constitutive KRAS ^G12D expression [20]. Within 48 h of doxycycline withdrawal, KRAS ^G12D expression was extinguished and whole-genome gene expression analyses of lung tumors were performed. Consistent with mutant KRAS-driven reprogramming of glutamine metabolism, ME1 and GOT1 levels were up-regulated when KRAS ^G12D was induced vs 48 h extinction with doxycycline withdrawl. c KRAS ^G12D induction upregulated ME1 mRNA in mouse doxycycline inducible KRAS ^G12D embryonic fibroblasts derived from the transgenic mice. d mRNA expression of ME1 in mutant KRAS vs wild-type KRAS NSCLC cell lines. Mutant KRAS lines: A549, CALU6, NCIH1155, NCIH1373, NCIH1385, NCIH1573, NCIH2030, NCIH2122, NCIH2347, NCIH460 and NCIH647. Wild-type KRAS lines: CALU3, HCC2108, HCC2279, HCC2935, HCC4006, NCIH322, NCIH520, NCIH522, NCIH596, NCIH661 and NCIH838. e NSCLC cell line dependencies on ME1 based on ATARiS gene phenotype value assessed from Project Achilles. Black bars = mutant KRAS cell. White bars = wild-type KRAS cell. Mutant KRAS lines: A549, CALU1, CORL23, HCC44, NCIH1650, NCIH1792, NCIH2122, NCIH23 and NCIH441. Wild-type KRAS lines: HCC2814, HCC827, NCIH1299, NCIH1437, NCIH1975, NCIH661, NCIH838 and HCC827GR5. f-g Seven day clonogenic survival assay of H522 and HCC44 with RNAi knockdown of ME1. h ME1 western blot in H522; band at 64 kDa. All results were compared using Student’s t-tests as indicated. *p < 0.05; **p < 0.01; ***p < .001

Fig. 2

Targeting glutamine metabolism sensitizes mutant KRAS NSCLC cell lines to radiation treatment. a, b Seven day clonogenic survival of HCC44 or H522 after radiation treatment after growth in either complete media or Gln deprived media for 16 h. c Clonogenic survival screen of mutant KRAS (H2009, H1573 and A549) or wild-type KRAS (H661, H322 and H596) NSCLC cell lines grown in either complete media or Gln deprived media for 16 h followed by treatment with 2 Gy of ionizing radiation. d Clonogenic survival of HCC44 and H522 pre-treated with 1 μM CB-839 for 48 h followed by treatment with various doses of ionizing radiation. All results were compared using Student’s t-tests as indicated. *p < 0.05; **p < 0.01; ***p < .001

Fig. 3

GOT1 and ME1 expression predicts response to radiation therapy in NSCLC patients. a Percent of complete responders to ionizing radiation (IR) in NSCLC patients separated based on KRAS status. Total number of complete responders in TCGA database = 14; wild-type KRAS = 13, mutant KRAS responders = 1. OR = odds ratio. Results compared using Fisher’s exact test. b, c ME1 and GOT1 log2 mRNA expression levels with calculated mean from TCGA NSCLC patients prior to radiation treatment with associated patient outcome after radiation treatment, CR = complete response, disappearance of all target lesions; PD = progressive disease, >20 % increase in the sum of the longest diameter of target lesions. Multiple probes integrated for each gene. d, e Kaplan-Meier overall survival curves in IR-treated NSCLC patients from KMPLOT database separated into high and low GOT1 and ME1 expression. Total number of NSCLC patients analyzed = 73; number of patients with high expression: ME1 = 40, GOT1 = 45; number of patients with low expression: ME1 = 33, GOT1 = 28. All results were compared using Student’s t-tests or a Cox regression analysis unless otherwise stated. *p < 0.05; **p < 0.01; ***p < .001