Target specificity, in vivo pharmacokinetics, and efficacy of the putative STAT3 inhibitor LY5 in osteosarcoma, Ewing's sarcoma, and rhabdomyosarcoma

Abstract

Background: STAT3 is a transcription factor involved in cytokine and receptor kinase signal transduction that is aberrantly activated in a variety of sarcomas, promoting metastasis and chemotherapy resistance. The purpose of this work was to develop and test a novel putative STAT3 inhibitor, LY5.

Methods and findings: An in silico fragment-based drug design strategy was used to create LY5, a small molecule inhibitor that blocks the STAT3 SH2 domain phosphotyrosine binding site, inhibiting homodimerization. LY5 was evaluated in vitro demonstrating good biologic activity against rhabdomyosarcoma, osteosarcoma and Ewing's sarcoma cell lines at high nanomolar/low micromolar concentrations, as well as specific inhibition of STAT3 phosphorylation without effects on other STAT3 family members. LY5 exhibited excellent oral bioavailability in both mice and healthy dogs, and drug absorption was enhanced in the fasted state with tolerable dosing in mice at 40 mg/kg BID. However, RNAi-mediated knockdown of STAT3 did not phenocopy the biologic effects of LY5 in sarcoma cell lines. Moreover, concentrations needed to inhibit ex vivo metastasis growth using the PuMA assay were significantly higher than those needed to inhibit STAT3 phosphorylation in vitro. Lastly, LY5 treatment did not inhibit the growth of sarcoma xenografts or prevent pulmonary metastasis in mice.

Conclusions: LY5 is a novel small molecule inhibitor that effectively inhibits STAT3 phosphorylation and cell proliferation at nanomolar concentrations. LY5 demonstrates good oral bioavailability in mice and dogs. However LY5 did not decrease tumor growth in xenograft mouse models and STAT3 knockdown did not induce concordant biologic effects. These data suggest that the anti-cancer effects of LY5 identified in vitro were not mediated through STAT3 inhibition.

Fig 1. LY5 inhibits pSTAT3 in sarcoma cell lines.

(A) The tumor cell lines RH30, EW8, and RD were treated with increasing concentrations of LY5 for 30 minutes. Cell lysates were collected for Western blotting of p-STAT3, total STAT3, and GAPDH. GAPDH was used as loading control. (B-C) The SJSA cell line was serum-starved overnight and then left untreated or treated with LY5 for two hours followed by stimulation with 50 ng/mL of OSM, IL-6, IFN-γ, IFN-α, or IL-4 for 30 minutes prior to collection of cell lysates for Western blotting of p-STAT3, total STAT3 p-STAT1, total STAT1, p-STAT2, p-STAT4, or p-STAT6. GAPDH was used as loading control.

Fig 2. Biologic activity of LY5 in sarcoma cell lines.

Dose response curves for LY5: RMS cell lines (RH30, RD, JR-1), ES cell lines (ES-3, ES-6, ES-7, ES-8), and an OS cell line (OS-17) were treated with LY5 (concentrations ranging from 10 nM to 10 μM) for 96 hours. Cell viability was determined using Alamar Blue staining. IC₅₀ values are indicated for each cell line. Analyses were performed in triplicate and data are plotted as means ± standard deviation.

Fig 3. Pharmacokinetic analysis of LY5 in mice and normal dogs.

(A) Semi logarithmic plot of LY5 plasma concentration over time in mice administered 5 mg/kg LY5 IV, IP, or PO (10 mice of each, n = 30 total). Animals were euthanized at various time points between 5 minutes and 24 hours to collect plasma samples. (B) Semi logarithmic plot of LY5 plasma concentrations in normal dogs administered a single 1 mg/kg dose of LY5 IV or through oral gavage. (C) Linear plot of concentration-time profile of LY5 in normal fasted dogs administered as powder in an oral capsule (25 and 50 mg/kg).

Fig 4. LY5 inhibits STAT3 phosphorylation in lung metastases but does not inhibit lung metastasis formation.

(A-B) IHC staining of lung metastasis from mice treated with LY5. The metastatic tumor sections were stained with pSTAT3 or hematoxylin and eosin. Western blotting was performed to confirm inhibition of STAT3 phosphorylation in lung tumor sections. (C) Mice were administered OS-17 cells stably expressing luciferase by tail vein injection and subsequently treated with LY5. After 30 days, imaging was performed to evaluate the effects of LY5 on the development of pulmonary metastasis.

Fig 5. LY5 does not inhibit growth of ex vivo pulmonary metastasis.

MG63.3 and K7M2 cell lines were exposed to increasing concentrations of LY5 to determine the IC₅₀ (A). Western blot analysis was performed to determine the dose and duration of LY5 exposure required to inhibit STAT3 phosphorylation (B-C). Serial imaging of fluorescently labeled MG63.3 and K7M2 cells was performed in ex vivo LY5 treated lung sections (D). Metastatic burden was quantified by evaluation of the percent tumor burden (mean normalized fluorescent area). Data are reported as the mean ± standard deviation. In MG 63.3 experiment, p < 0.0001 Kruskal-Wallis test, p < 0.05 10 μM vs control at days 7, 14, and 21, 10 μM vs 1 μM at days 7, 14, and 21, 50 μM vs control at days 7, 14, and 21, and 50 μM vs 10 μM at days 7, 14, and 21 post-test Dunn. In K7M2 experiment, p < 0.0001 Kruskal-Wallis test, p < 0.05 10 μM vs control at days 14 and 21, 10 μM vs 0.1 μM at days 14 and 21, and 10 μM vs 1 μM at days 14 and 21, post-test Dunn.

Fig 6. Effects of LY5 are not mediated via STAT3 inhibition.

(A) Antitumor activity of LY5 against OS-1 PDX model of OS. Left panel, growth of individual control (untreated OS-1 xenografts); Center panel, growth of individual OS-1 xenografts in mice receiving LY5 (40 mg/kg twice daily for 5 days/week for 4 consecutive weeks); Right panel, median relative tumor growth (RTV) from control and treated groups. (B) Sarcoma cell lines (RMS cell line RH30 and ES cell lines ES-3 and ES-7) were transfected with either negative control siRNA (NC siRNA) or STAT3-targeting siRNA #7. After 72 hours of transfection, cells were harvested and analyzed for total (T) STAT3 and GAPDH protein expression by immunoblotting. (C) RH30, ES-7, and EW-8 cell lines were transfected with either NC siRNA or STAT3-targeting siRNA. After 48 hrs, cells were incubated with LY5 (1μM) for an additional 48 hrs. Cell viability was then determined using Alamar Blue staining. Data represent the mean ± standard deviation, n = 3. * denotes p < 0.05 by ANOVA.