UVB-induced DHODH upregulation, which is driven by STAT3, is a promising target for chemoprevention and combination therapy of photocarcinogenesis

Abstract

The leading cause of cutaneous squamous cell carcinomas (cSCCs) is exposure to ultraviolet radiation (UV). Unlike most other cancers, the incidence rates of cSCCs are still on the rise and the treatment options currently available are limited. We have recently found that dihydroorotate dehydrogenase (DHODH), which is the rate-limiting enzyme in the de novo pyrimidine synthesis pathway, plays a critical role in UVB-induced energy metabolism reprogramming. Using a multistage model of UVB radiation-induced skin cancer, we show that UVB-induced DHODH upregulation is mainly regulated transcriptionally by STAT3. Our results indicate that chronic inhibition of DHODH by leflunomide (LFN) blocks UVB-induced tumor initiation. Human tumor xenograft studies showed that LFN treatment reduces growth of established tumors when used in combination with a genotoxic agent, 5-fluorouracil (5-FU). Our data suggest that DHODH is a promising target for chemoprevention and combination therapy of UVB-induced cSCCs.

Fig. 1. Chronic UVB exposure results in DHODH overactivation SKH-1 mice exposed to chronic UVB irradiation.

a The relative levels of DHODH mRNA were quantified by quantitative reverse transcription PCR. b Total protein extracts of skin biopsies at different stages of tumorigenesis were assessed for expression of DHODH by western blot. β-actin was used as a loading control. Full-length blots are presented in Fig. S1. c DHODH activity was measured in skin biopsies at different stages of UVB-induced carcinogenesis. d DHODH is the only mitochondrial enzyme of the de novo pyrimidine synthesis pathway that converts dihydroorotate to orotate. e, f Cells were incubated for 4 h with [1,4-¹³C] aspartate. e Schematic depicting atoms of orotate that are derived from aspartate and carbamoyl phosphate. Blue color represents labeled carbons from aspartate, red color represents non-labeled carbons and nitrogen from aspartate, and black color represents carbon and nitrogen derived from carbamoyl phosphate. f ¹³C-NMR spectra of perchloric acid cell extract after 4 h of incubation with [1,4-¹³C] aspartate. nIr non-irradiated cell, Ir irradiated cell. 1: Aspartate C4, 2: Aspartate C1, 3: orotate C2, and 4: orotate C5. g Specific DHODH activity was measured by normalization of DHODH activity to its expression level in each sample. N = 10 [(A–C, G)] and 6 [(F)] mice per group *P < 0.05 and **P <0.01 for irradiated versus non-irradiated mice. PRPP phophoribosylpyrophosphate, CAD carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase, CTPS1 CTP synthase, DHODH DHO dehydrogenase, OMP orotidine 5′-monophosphate, UMPS uridine 5′-monophosphate synthase

Fig. 2. UVB-induced DHODH overexpression is regulated via a STAT3-dependent mechanism.

a Representation of the DHODH gene and its promoter in the UCSC genome browser, including location of the eight motifs matching the consensus STAT3-binding site [TTNNNNNAA)] and the location of PCR products for ChIP-qPCR. b Eight nucleotide-sequences matching the consensus STAT3-binding site (GAS-1–GAS-8) in the upstream region of mouse DHODH gene are numbered in relation to the translational start codon, ATG. c The expression of STAT3 was assessed in mouse skin specimens at different stages of tumorigenesis. Upregulation of STAT3 expression persisted at different stages of UVB-induced carcinogenesis. Full-length blots are presented in Fig. S2. d Irradiated and non-irradiated skin were subjected to ChIP assay using an anti-STAT3 antibody. Bands indicate PCR products using primers that span the indicated GASs. The relative levels of corresponding precipitated GAS fragments following ChIP were quantified by qRT-PCR. e Luciferase reporter plasmids containing indicated fragments were transfected into siCtrl-transfected or siSTAT3-transfected keratinocytes isolated from irradiated or non-irradiated skin. N = 6 mice per group. *P < 0.05 and **P < 0.01

Fig. 3. Inhibition of DHODH using LFN treatment results in hypersensitivity to UVB exposure.

One-month-old SKH-1 mice were subjected to chronic UVB irradiation ± intraperitoneal injection of either LFN (20 mg/kg/day) or placebo. a The relative DHODH activity in irradiated and non-irradiated skin was measured at 8 weeks after irradiation. Increased DHODH activity in UVB-irradiated skin samples was blocked following treatment of mice with LFN. b Photographs are representative examples of placebo-treated and LFN-treated mice after 18 and 28 weeks of chronic UVB irradiation. c, d The percentage of tumor-free mice c and desquamative feature-free mice d were assessed at indicated times. e, f the numbers e, and the combined volumes f of tumors per mouse were recorded at different intervals. g The distribution of the mean volume of all single tumors/mouse is presented at week 28 of chronic UVB irradiation. h The specific tumor growth rate SGR was estimated for each mouse according to the following equation: SGR = Ln2/DT, in which DT = doubling time. i Cyclopyrimidine dimer (CPD) levels were quantified in mouse skin at 8 weeks of chronic UVB irradiation by immuno-dot blot analysis. Full-length blots are presented in Fig. S3. N = 12 [(A–I)] mice per group *P < 0.05, **P < 0.01, and ****P < 0.0001

Fig. 4. Uridine supplementation largely restored DHODH inhibition-mediated decreased DNA repair capacity and hypersensitivity to UVB.

One-month-old SKH-1 mice were subjected to chronic UVB irradiation + intraperitoneal injection of either LFN (20 mg/kg/day) or placebo + intraperitoneal uridine. a Time courses of plasma uridine concentrations after an i.p. injection of exogenous uridine at week 8 of chronic UVB exposure. b Photographs are representative examples of mice treated with either placebo + uridine or LFN + uridine after 18 weeks of chronic UVB irradiation. c–e The percentage of tumor-free mice c, the tumor numbers d, and the volume e of tumors per mouse were assessed at indicated times. f The distribution of the mean volume of all single tumors/mouse is presented at week 28 of chronic UVB irradiation. g The specific tumor growth rate SRG was estimated for each mouse. h Immuno-dot blot analysis reveals the same quantity of cyclopyrimidine dimer (CPD) in both LFN-treated and placebo-treated mice supplemented with uridine upon chronic UVB irradiation. Full-length blots are presented in Fig. S3. N = 12[(A–H)] mice per group

Fig. 5. The combination of LFN and 5-FU reduces tumor growth in vivo.

a, b The effect of different concentrations of LFN and 5-FU on the viabilities of A431 a and SCC-15 b was measured at 24 h after treatment. The statistical analysis on the graphs compares the drug combinations to each respective treatment alone. All values are represented as a percentage (%) relative to the placebo. Data is presented as the mean ± SEM of three independent experiments each performed in triplicate

Fig. 6. The combination of LFN and 5-FU reduces tumor growth in vivo.

a A431 and SCC-15 cell (5 × 10⁵) were injected subcutaneously into NSG mice. When tumors were palpable (4-weeks post-xenografting), mice were treated daily with placebo, LFN, 5-FU, or combined 5-FU and LFN for 18 days. Tumor volumes were measured at six time points (T1–T6). b, c The body weight of mice after treatment was measured at the indicated time points. Results represent the mean ± SD. d The combination of LFN and 5-FU reduced the average A431 tumor volume greater than either drug alone. Data is presented as the mean ± SEM of one independent experiment. Statistical analysis (two-way ANOVA with Bonferroni’s post-hoc test) compares the combination therapy versus each drug alone. ****P ≤ 0.0001. °P ≤ 0.05, °°°°P ≤ 0.0001 for combined 5-FU and LFN vs. 5-FU. e Tumor growth curves of individual A431 tumors in mice treated with 5-FU (black lines) and 5-FU plus LFN (gray lines). The effect of each treatment on mean volume of tumors at indicated day are shown at the top of the panel. f The combination of leflunomide and 5-FU reduced tumor weight greater than either drug alone. g–i The mean volume of tumors at indicated time points g, individual tumor growth curves h, and the weight of tumors at day 18 i are shown for transplanted SCC-15 cells. N = 10 mice per group, one injection per mice. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001