Dihydroorotate dehydrogenase : new insights into an old target

Abstract

Purine and pyrimidine nucleotides are important for both biosynthetic and regulatory pathways in cells. As constituents of the DNA and RNA molecules, the requirements for nucleotides in actively dividing cells are increased. Therefore, the enzymes from the purine and pyrimidine nucleotide synthesis pathways are considered attractive targets for treatment of diseases associated with deregulated cell proliferation, such as cancer. Dihydroorotate dehydrogenase (DHODH), an enzyme from the de novo pyrimidine ribonucleotide synthesis pathway located on the inner mitochondrial membrane, has been intensively studied in the past few years as a target for anti-cancer therapy. As a result, several inhibitors of DHODH are currently in clinical trials for treatment of myeloid malignancies.

In a screen for p53 activators, we discovered a novel class of DHODH inhibitors with a tetrahydroindazole core structure (Paper I). The hit compound from the screen, the chiral compound named HZ00, increased p53 translation without increasing p53 mRNA levels or the stability of the protein. Furthermore, treatment with HZ00 accumulated cancer cells in the S-phase of the cell cycle while it led to cell cycle arrest in normal fibroblasts. After screening of commercially available HZ analogues, we discovered the more potent compound HZ05. The active enantiomer of the chiral HZ compound was the (R)-enantiomer, confirmed by the co-crystallization of DHODH with HZ05 where only (R)-HZ05 was bound to the quinone tunnel of DHODH (membrane binding domain). Additionally, both compounds acted synergistically with the MDM2 inhibitor nutlin-3 (which protects p53 from degradation) in different tumor cell lines and in a melanoma xenograft model.

Following the discovery of the HZ compounds as inhibitors of DHODH, we investigated the structure and function of DHODH using mass spectrometry and molecular dynamics simulations (Paper II) and afterwards optimized the HZ compounds in a structure activity relationship study (SAR, Paper III). Using non-denaturing mass spectrometry, we established that the interactions between DHODH and its co-factor FMN, the lipids present in the mitochondrial membrane, as well as ligands such as the DHODH inhibitor brequinar, have stabilizing effect on the protein. Furthermore, molecular dynamic stimulations demonstrated the flexibility of the membrane binding domain and the transmembrane helix. We also illustrated that the lipid interactions that stabilize the membrane binding domain may affect the binding of longer ligands, which is important for the design of new inhibitors. The optimization of the HZ-compounds was based on a metabolic soft spot study and the crystal structure of DHODH with (R)-HZ05 obtained in Paper I. The activity of many of the new HZ inhibitors was improved in the in vitro enzymatic assay as well as in cellular assays using a melanoma cell line. Based on in vitro metabolic stability analysis we identified a lead compound in the series, which demonstrated high specificity towards DHODH. The new active HZ-analogues were also able to trigger p53 transcription factor function and affected the cell growth/viability of a melanoma cell line with only a small effect on a normal fibroblast, suggesting a reasonably wide therapeutic window. Supplementation with an excess of uridine ameliorated both, the activation of p53 by the HZ compounds and their effect on cell growth/viability. Under our experimental conditions, the new HZ analogues did not activate DNA damage markers. Additionally, we demonstrated that the ability of DHODH inhibitors to induce apoptosis related g-H2AX was prevented by co-treatment with the pan-caspase inhibitor Z-VAD-FMK.

Due to the re-emerging interest in DHODH as a drug target, we investigated the early events triggered after inhibition of this enzyme (Paper IV). For that purpose, we performed RNA sequencing analysis of cells treated with two DHODH inhibitors for up to 4 h. One of the earliest significant changes was the fast downregulation of PNUTS mRNA. This event was observed at the protein level as well. The PNUTS decrease was accompanied by an increase of p53 protein, although the correlation between the two changes requires further investigation. Nevertheless, due to the early decrease in PNUTS we propose that the downregulation of PNUTS mRNA may be a marker for DHODH inhibition.