Targeting a xenobiotic transporter to ameliorate vincristine-induced sensory neuropathy

Abstract

Vincristine is a widely used chemotherapeutic drug for the treatment of multiple malignant diseases that causes a dose-limiting peripheral neurotoxicity. There is no clinically effective preventative treatment for vincristine-induced sensory peripheral neurotoxicity (VIPN), and mechanistic details of this side effect remain poorly understood. We hypothesized that VIPN is dependent on transporter-mediated vincristine accumulation in dorsal root ganglion neurons. Using a xenobiotic transporter screen, we identified OATP1B3 as a neuronal transporter regulating the uptake of vincristine. In addition, genetic or pharmacological inhibition of the murine orthologue transporter OATP1B2 protected mice from various hallmarks of VIPN - including mechanical allodynia, thermal hyperalgesia, and changes in digital maximal action potential amplitudes and neuronal morphology - without negatively affecting plasma levels or antitumor effects of vincristine. Finally, we identified α-tocopherol from an untargeted metabolomics analysis as a circulating endogenous biomarker of neuronal OATP1B2 function, and it could serve as a companion diagnostic to guide dose selection of OATP1B-type transport modulators given in combination with vincristine to prevent VIPN. Collectively, our findings shed light on the fundamental basis of VIPN and provide a rationale for the clinical development of transporter inhibitors to prevent this debilitating side effect.

Keywords: Oncology; Pharmacology; Toxins/drugs/xenobiotics; Transport.

Figure 1. OATP1B2 deficiency attenuates vincristine-induced sensory peripheral neurotoxicity (VIPN).

(A) Intracellular accumulation of vincristine into HeLa cells overexpressing mouse (m), rat (r), or human (h) transporters, except for mOATP1B2, whose model was generated in HEK293 cells. Relative uptake is expressed as percentage change compared with empty vector controls (n = 4–14 per group). (B) Expression of the OATP1B1 and OATP1B3 genes in human whole DRG samples (pooled RNA from 21 male/female subjects), by reverse transcription PCR (RT-PCR) (depicted by the 154 bp product for OATP1B1 and 111 bp product for OATP1B3). Human OATP1B1 and OATP1B3 transporter–overexpressed cells were used as positive controls; empty vector controls were used as the negative control. LD, 100 bp ladder. (C–E) VIPN in WT mice or OATP1B2 deficient mice (OATP1B2^–/–) at baseline and at 2, 3, and 4 weeks following weekly administrations of vincristine at a dose of 1 mg/kg (cumulative dose 4 mg/kg). Mechanical allodynia (C), thermal hyperalgesia (D), and sciatic nerve maximal action potential amplitudes (E) are expressed as percentage change relative to baseline values (n = 6–8 per group). Statistical analysis was performed using a 2-way ANOVA with Bonferroni’s post hoc test. **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with baseline values. (F) Plasma concentration-time profile of vincristine (1 mg/kg) in WT mice or OATP1B2^–/– mice (n = 4 per group). Data are shown as mean ± SEM.

Figure 2. Inhibition of OATP1B-type transporters by nilotinib.

(A and B) Nilotinib-mediated inhibition of vincristine and estradiol-17β-ᴅ-glucuronide (EβG) uptake in HEK293 cells overexpressing vector control, mouse (m) OATP1B2, human (h) OATP1B1, or hOATP1B3. Vincristine and EβG uptake data are expressed as percentage change compared with empty vector controls (n = 3–6 per group). Statistical analysis was performed using 1-way ANOVA with Dunnett’s post hoc test. * P < 0.05. (C) Nilotinib uptake in HEK293 cells overexpressing vector control, mOATP1B2, hOATP1B1, or hOATP1B3. (D) Plasma concentration-time profile of nilotinib (10 mg/kg) in male and female WT mice or OATP1B2^–/– mice (n = 12 per group, 6 for each sex). (E) Levels of nilotinib (10 mg/kg, 6 hours after treatment) in liver samples from male and female WT mice or OATP1B2^–/– mice (n = 12 per group, 6 for each sex). (F) Levels of nilotinib (25 mg/kg or 100 mg/kg, 1.5 hours after treatment) in DRG samples from female WT mice or OATP1B2^–/– mice (n = 3 per group). Statistical analysis was performed using an unpaired 2-tailed Student’s t test with Welch’s correction for liver and DRG drug accumulation. Data are shown as mean ± SEM.

Figure 3. Inhibition of OATP1B2 with nilotinib protects against vincristine-induced peripheral neuropathy.

(A–C) Mechanical allodynia (A), thermal hyperalgesia (B), and sciatic nerve maximal action potential amplitudes (C) at baseline and 4 weeks (n = 6–7 per group) in WT mice or OATP1B2-deficient (OAP1B2^–/–) mice receiving weekly i.p. injections of vincristine at a dose of 1 mg/kg (cumulative dose 4 mg/kg). Mice were pretreated with vehicle (hydroxypropyl methylcellulose) or nilotinib (100 mg/kg; p.o.) 30 minutes before every vincristine injection. Statistical analysis was performed using a 2-way ANOVA with Tukey’s post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (D) Light microscopy analysis of the sciatic nerve of WT mice and OATP1B2^–/– mice receiving weekly vincristine (cumulative dose 4 mg/kg). Mice were pretreated with vehicle (hydroxypropyl methylcellulose) or nilotinib (100 mg/kg; p.o.) 30 minutes before every vincristine injection. After 4 weeks, vincristine-induced axonopathy (arrow) and fiber loss (circle) was evident. Nilotinib pretreatment and OATP1B2 deficiency protected against these morphologic changes. Magnification, 63×. Scale bar:20 μm. (E) Plasma concentration-time profile of vincristine (1 mg/kg) in WT mice or OATP1B2^–/– mice pretreated with vehicle or nilotinib (100 mg/kg) (n = 7–9 per group). (F) DRG to plasma AUC ratio of vincristine (1 mg/kg) in WT mice or OATP1B2^–/– mice pretreated with vehicle or nilotinib (100 mg/kg) (n = 3 per group). Statistical analysis was performed using 1-way ANOVA with Dunnett’s post hoc test. *P < 0.05, compared with WT mice pretreated with vehicle. Data are shown as mean ± SEM.

Figure 4. Nilotinib as an adjunct to vincristine therapy.

(A) Expression of the human OATP1B1 and OATP1B3 genes in leukemia cells as measured by RT-PCR (depicted by the 154 bp product for OATP1B1 and 111 bp product for OATP1B3). HEK293 cells engineered to overexpress OATP1B1 or OATP1B3 were used as positive controls (denoted “Cell”), and cells transfected with an empty vector were used as negative controls (denoted “VC”). (B and C) Low OATP1B1 and OATP1B3 expression in malignant cells was confirmed in samples of 314 pediatric patients with acute myeloid leukemia (AML) (B) and 655 pediatric patients with B-lineage acute lymphoblastic leukemia (ALL) (C). Expression of the reduced folate carrier protein 1 (RFC1) was used as a reference gene in all samples. Median values in AML samples were RFC1 = 2.77, OATP1B1 = 0.00155, OATP1B3 = 0.00172; median values in ALL samples were RFC1 = 9.91, OATP1B1 = 0.00218, OATP1B3 = 0.00205. (D) Uptake of vincristine in leukemia cells in the presence or absence of nilotinib (1 μM) (n = 9 per group). Uptake data were normalized to total protein content. (E) Cytotoxicity of vincristine in leukemia cells in the presence or absence of nilotinib (1 μM). Cytotoxicity was measured by an MTT assay in 2-dimensional culture following continuous 72-hour exposure to vincristine (n = 9 per group). Data are shown as mean ± SEM. Statistical analysis was performed using a Student’s t test with Welch’s correction. (F) Proposed model of vincristine-induced injury to the peripheral nervous system in mice. Vincristine is taken up into cells within the peripheral nervous system by the transporter OATP1B2, ultimately leading to peripheral neuropathy, and these effects can be prevented by the OATP1B2 inhibitor nilotinib without negatively affecting anti-tumor efficacy.

Figure 5. Identification of a DRG-specific endogenous biomarker of OATP1B2.

(A) Schematic overview of the untargeted metabolomics screen to discover an endogenous biomarker of OATP1B2. (B) Volcano plot of differentially endogenous metabolites in untreated plasma of WT mice or OATP1B2-deficient (OATP1B2^–/–) mice. Positive fold change indicates higher plasma concentration in OATP1B2^–/– mice (n = 6 per group). Dotted lines indicate a log₁₀ P value threshold of > 1.3 and a log₂FC (fold change) of ± 0.2. (C) Volcano plot of differentially endogenous metabolites in untreated whole DRG samples of WT mice or OATP1B2^–/– mice. Negative fold change indicates lower DRG levels in OATP1B2^–/– mice (n = 6 per group). (D) Plasma levels and DRG/plasma ratios of α-tocopherol in WT mice or OATP1B2^–/– mice in secondary validation studies. Statistical analysis was performed using an unpaired 2-tailed Student’s t test with Welch’s correction. ***P < 0.001 (n = 6 per group). (E) Relative DRG/plasma ratio of α-tocopherol in WT mice or OATP1B2^–/– mice 2 hours after the administration of vehicle control (hydroxypropyl methylcellulose), rifampin (20 mg/kg; i.p.), or nilotinib (100 mg/kg; p.o.). Data are normalized to baseline levels observed in WT mice (n = 3 per group). Statistical analysis was performed using 1-way ANOVA with Dunnett’s post hoc test. ****P < 0.0001.