The Second-Generation PGI2 Analogue Treprostinil Fails to Chemoprevent Tumors in a Murine Lung Adenocarcinoma Model

Abstract

Prostacyclin (prostaglandin I₂, PGI₂) overproduction in FVB/N mice prevents the formation of carcinogen and tobacco smoke-induced adenomas, and administration of the oral prostacyclin analogue iloprost to wild-type mice also prevented carcinogen-induced mouse lung adenoma formation. Former smokers taking oral iloprost showed improved bronchial dysplasia histology compared with placebo. Next-generation oral prostacyclin analogues, like treprostinil, were developed for the treatment of pulmonary arterial hypertension (PAH). On the basis of our prior studies with iloprost, we performed preclinical studies examining the ability of treprostinil to chemoprevent urethane-induced murine lung adenocarcinoma. We determined the MTD in chow (prior studies had delivered treprostinil by gavage), and this dose produced serum levels in the experimental animals similar to those found in PAH patients treated with treprostinil. We then examined the chemopreventive efficacy of treprostinil exposure initiated both before (1 week) and after (6 weeks) urethane exposure to better model chemoprevention studies conducted in former smokers. Neither of these dosing strategies prevented murine lung cancer; however, we did detect changes in pulmonary inflammatory cell infiltrate and expression of CXCR4 (a chemokine receptor previously shown to increase in response to treprostinil exposure) in tumor-bearing, treprostinil-treated animals, indicating that the drug was bioavailable. One potential explanation stems from iloprost and treprostinil differentially activating cell surface prostaglandin receptors and intracellular peroxisome proliferator-activated receptors. When murine lung tumor cells were treated with treprostinil, their proliferation rate increased; in contrast, iloprost had no effect on proliferation. Future investigations comparing these two agents will provide insight into iloprost's chemopreventive mechanisms. Cancer Prev Res; 10(11); 671-9. ©2017 AACR.

Figure 1. Treprostinil MTD and tumor prevention study

Mice fed increasing doses of treprostinil in chow were weighed daily (A) for 3 weeks, and 28.8 mg/kg was determined to be the maximum tolerated dose (MTD). Treprostinil in chow (28.8 mg/kg chow) was administered 1 week before (early treatment, ET) or after 6 weeks after (late treatment, LT) urethane exposure, and tumor number (B), burden (C), and diameter (D) were measured in mice harvested 16 weeks after urethane exposure. Weight gain was similar between groups (E). Plasma treprostinil levels were measured in each group (F) at harvest and consistent levels were found in each treprostinil treated group

Figure 2. FACs analysis of lymphocytes

Lower right lobe digests from each lung were incubated with fluorescent antibodies against CD3, CD4, and CD8. Using FlowJo software, live cells were gated and doublets excluded on forward scatter (FSC) and side scatter (SSC). CD3⁺ cells were separated into CD4⁺ and CD8⁺ cells (A) for the Urethane and ET groups (LT results were indistinguishable from ET results). Total CD4⁺ and CD8⁺ cells/lobe were quantitated and results graphed in (B). **p < 0.01 vs. Urethane group.

Figure 3. FACS analysis of macrophage and neutrophil populations

Lower right lobe digests from each lung were incubated with fluorescent antibodies against MHCII, Ly6G, CD11c, CD11b, and F4/80. Using FlowJo software, live cells were gated, followed by doublet exclusion on forward scatter (FSC) and side scatter (SSC). Macrophages were identified as a CD11c^hiCD11b^varF4/80^hi subset including resident alveolar macrophages and recruited alveolar and interstitial/tissue macrophages and a CD11c^midCD11b^hiF4/80^loLy6G⁻ subset. Neutrophils were identified as CD11b^hiCD11c⁻Ly6G⁺ cells, (A) Urethane group and (B) Urethane ET group (results from LT mice were indistinguishable from those of ET mice). (C) Average total macrophage subsets and neutrophils/lobe were calculated for each group and results are graphically represented. *p<0.05 vs. Urethane group.

Figure 4. CXCR4 expression increases in lung tissue from tumor-bearing, treprostinil-treated mice

RNA was isolated from lung tissue after tumors had been removed. CXCR4 message was reversed transcribed and measured by qPCR, and relative amounts were compared using the ΔΔCT method. Differences were detected by one-way ANOVA followed by Dunnett post hoc analysis; *p < 0.05.

Figure 5. PGE2 and treprostinil increase proliferation of mouse lung adenocarcinoma cells in vitro but not in vivo

JF32 mouse lung adenocarcinoma cells were serum-starved for 24 h prior to incubation with increasing doses of PGE2 (A) or treprostinil (B) in serum-free media. Cells were harvested by trypsinization and counted; *p < 0.05 vs control. (C) Iloprost pre-treatment prevented PGE2 and treprostinil-mediated increases in JF32 cell proliferation; ***p < 0.01 vs. PGE2 or treprostinil alone. Graphs represent 2 independent experiments with at least 3 samples/group in each experiment. (D) Ki67⁺ staining was used to generate a proliferative index (E). Ki67 nuclear staining (red arrowhead) was counted in each tumor. Area of each tumor was determined using CellSens Entry software, and the number of Ki67⁺ nuclei/mm₂ tumor was determined in each group. No differences were detected between treatment groups.