Pacific Blue Derivatives of Paclitaxel as Fluorescent Probes of OATP1B‐Type Transporters

ABSTRACT

Paclitaxel (Taxol) is a widely used anticancer agent that undergoes extensive hepatic metabolism and that causes a debilitating, dose‐limiting peripheral neurotoxicity. We previously reported that the uptake of paclitaxel in hepatocytes and dorsal root ganglion neurons, the site of injury within the nervous system, is mediated by the organic anion transporting polypeptides OATP1B1 and OATP1B3 (Oatp1b2 in rodents), transporters that are highly sensitive to pharmacological inhibition. To facilitate future screens of chemical libraries to identify modulators and imaging‐based drug distribution studies, we explored the utility of PB‐Gly‐Taxol and PB‐GABA‐Taxol, derivatives of paclitaxel linked to the coumarin‐derived fluorophore Pacific Blue, as in vitro and in vivo substitute biomarker probes of paclitaxel. Transport studies in transfected HEK293 cells revealed efficient uptake of these PB‐taxoids by human and murine OATP1B/Oatp1b‐type transporters, with up to 100‐fold increases in uptake relative to values observed in vector control cells, and inhibition of this transport by known inhibitors. Although cell viability assays demonstrated lower cytotoxicity of both PB‐taxoids (IC₅₀: 13~80 nM) against a panel of breast cancer cell lines, ensuing investigations confirmed their ability to induce peripheral neurotoxicity phenotypes in mice (p < 0.05), in an Oatp1b2‐dependent manner, to the same extent as paclitaxel. These findings imply that PB‐taxoids mimic the transport and toxicokinetic features of paclitaxel, and these agents thus offer potential as fluorescent imaging tools for exploring drug–drug interaction liabilities and paclitaxel‐related toxicity profiles that involve OATP1B/Oatp1b‐type transporters.

Study Highlights

• What is the current knowledge on the topic?

  • ○
    Synthetic analogues of paclitaxel linked at the 7‐position to the fluorophore Pacific Blue (PB‐Gly‐Taxol and PB‐GABA‐Taxol) are known to potently bind microtubules in vitro, retain cytotoxicity toward HeLa cells, and are substrates of efflux transporters.
• What question did this study address?

  • ○
    Whether the fluorescent compounds PB‐Gly‐Taxol and PB‐GABA‐Taxol can be used as fluorescent alternatives to paclitaxel in translationally relevant preclinical models.
• What does this study add to our knowledge?

  • ○
    We found that PB‐Gly‐Taxol and PB‐GABA‐Taxol share requisite similarities with the parent compound, paclitaxel, in terms of OATP1B/Oatp1b transporter‐mediated cellular uptake, cytotoxicity against multiple malignant and normal cell lines, and the ability to induce mechanical allodynia in an Oatp1b2 transporter‐dependent manner.
• How might this change clinical pharmacology or translational science?

  • ○
    The fluorescent analogues PB‐Gly‐Taxol and PB‐GABA‐Taxol closely mimic the transport and toxicokinetic features of paclitaxel, and both compounds have the potential to serve as fluorescent detection and imaging tools for exploring drug–drug interaction liabilities and paclitaxel‐related toxicity profiles that involve OATP1B/Oatp1b‐type transporters.

1. Introduction

Many tubulin poisons induce a dose‐dependent sensory peripheral neurotoxicity that is characterized by tingling, numbness, and increased sensitivity to cold and touch. The incidence of this neurotoxicity is particularly high in the case of paclitaxel, as it occurs in up to 70%–80% in patients with breast cancer [1]. The mechanistic basis of this side effect remains incompletely understood, although prior studies have demonstrated that paclitaxel induces morphological and biochemical alterations in dorsal root ganglion (DRG) satellite glial cells [2], the main site of paclitaxel uptake within the nervous system [3]. We previously reported that paclitaxel accumulates into murine DRG through a process mediated by the organic anion transporting polypeptide Oatp1b2 (OATP1B3 in humans), that paclitaxel‐induced peripheral neurotoxicity in mice is Oatp1b2‐dependent, and that neurotoxicity phenotypes can be prevented by pre‐treatment with pharmacological inhibitors such as nilotinib and rifampin [4]. Since the mechanism by which paclitaxel is taken up into hepatocytes, in advance of metabolism, is also dependent on Oatp1b2 (OATP1B1 and OATP1B3 in humans) [5], careful consideration of potential drug–drug interaction liabilities associated with the use of OATP1B/Oatp1b inhibitors as neuroprotective agents is warranted. Measurements of the affinity and selectivity of the microtubule‐binding chemotherapies are largely based on studies using purified proteins, and few quantitative methods for transmembrane transportation of small molecules with microtubules in living cells have been reported. The availability of fluorescent derivatives of paclitaxel that retain key toxicokinetic features of the parent compound would facilitate such studies, and we hypothesized that two amino acid derivatives of paclitaxel we recently synthesized by conjugating paclitaxel with the coumarin‐derived fluorophore Pacific Blue (PB) could serve this purpose. These compounds, referred to as PB‐Gly‐Taxol and PB‐GABA‐Taxol (collectively, PB‐taxoids) potently bind to microtubules in vitro, retain cytotoxicity toward HeLa cells, and, similar to paclitaxel, are substrates of the efflux transporters ABCB1 (P‐glycoprotein) and ABCC2 (MRP2) [6, 7, 8]. The purpose of this present study was to evaluate (i) sensitivity of these derivatives to OATP1B/Oatp1b‐mediated transport, (ii) antitumor activity against breast cancer cells, and (iii) ability to induce peripheral neurotoxicity in translationally relevant preclinical models.

2. Materials and Methods

2.1. Synthesis of PB‐Taxoids

Paclitaxel derivatives were synthesized as previously described [7]. Briefly, the hydroxyl group at the 7‐position of a TBS‐protected derivative of paclitaxel was esterified with glycine (Gly) or gamma‐aminobutyric acid (GABA); these amino acids were coupled to the fluorophore Pacific Blue, and the TBS protecting group was removed to generate two fluorescent derivatives, PB‐Gly‐Taxol and PB‐GABA‐Taxol.

2.2. Cellular Accumulation

Cellular uptake studies were performed in the presence or absence of nilotinib in HEK293 cells (ATCC, Manassas, VA) engineered to overexpress human OATP1B1, human OATP1B3, or the orthologous murine transporter, Oatp1b2 [4, 5]. All cells are routinely checked to ensure there is no mycoplasma contamination using MycoAlert Detection Kit (Lonza, Pearland, TX). Briefly, cells were grown to confluence (80%–90%) in an incubator supplied with 5% CO₂ and 95% relative humidity at 37°C, and 50,000 cells were seeded per well in poly‐lysine pre‐coated 96‐well plates. After 18–24 h of seeding, cells were washed with pre‐warmed PBS (pH 7.4) and pre‐incubated with either vehicle or the OATP1B/Oatp1b inhibitor, nilotinib, prepared in DMEM without FBS and phenol red. After pre‐incubation for 15 min, cells were further treated with either PB‐Gly‐Taxol or PB‐GABA‐Taxol (0.1 μM) in the presence or absence of 10 μM nilotinib for 10 min. Transport was stopped by washing cells three times with ice‐cold PBS. Intracellular fluorescence originating from the two taxoids was measured by flow cytometry and confocal laser‐scanning fluorescence microscopy (PB‐taxoids were excited with a 405 nm laser and emission was collected from 425 nm to 500 nm as described previously [6, 7]) or using a plate reader (Agilent BioTek Synergy H1 Hybrid with excitation at 405 nm and emission at 460 nm). The concentrations of PB‐taxoids were calculated based on fluorescence‐concentration standard curves. Net uptake rates were calculated as the difference between OATP1B/Oatp1b overexpressing and vector control cells. Kinetic analysis of uptake of PB‐taxoids in OATP1B3 overexpressing cells was performed in a substrate concentration range of 10 to 1000 nM after the linearity of cellular uptake over time was determined and normalized by protein levels. The kinetic parameters K _m (Michaelis constant) and V _max (maximum velocity), as well as the inhibition constant (K _i ), were calculated in GraphPad Prism using the Michaelis–Menten function (Velocity of uptake = V _max*[Substrate]/(K _m  + [Substrate])) and the Dixon plot function.

2.3. Animal Models

Wild‐type mice and Oatp1b2‐deficient mice [Oatp1b2(−/−)] on a DBA background strain were bred in‐house. These mice were provided by Richard B. Kim (Western University, London, Ontario, Canada) and Jeffrey L. Stock (Pfizer Inc., Groton, CT). All animals were kept in a controlled setting with a 12‐h light–dark cycle, given access to food and water as needed, and handled in accordance with The Ohio State University's Animal Care and Use Committee's approved protocol 2015A00000101‐R2. Following random assignment to control and intervention groups, mice were distributed across groups in a manner that maintained balance in terms of group size and baseline traits like weight, sex, and age [9].

2.4. Evaluation of Peripheral Neurotoxicity

A Von Frey Hair (VFH) test of mechanical allodynia in mice was utilized as a readout of drug‐induced peripheral neurotoxicity [4]. In brief, for acute neurotoxicity testing, a single dose of 10 mg/kg of either paclitaxel (Hospira Inc., Lake Forest, IL), PB‐Gly‐Taxol or PB‐GABA‐Taxol, all formulated in a mixture of Cremophor EL and ethanol (1:1, v/v) and diluted with 0.9% saline solution, was administered as a bolus injection in the tail vein. The VFH tests were employed before treatment to establish baseline levels of sensitivity and were repeated 24, 48, 72‐h and 1 week after drug administration. All animals were acclimated for 1 h in a top‐wire mesh network prior to sensitivity testing. Mechanical allodynia readings were expressed as a percentage change from baseline values to account for inter‐day variability of the results. The analysts involved in drug administration and VFH test evaluation were blinded to the treatment and mouse genotypes.

2.5. Stability of PB‐Gly‐Taxol and PB‐GABA‐Taxol

In addition to the four esters present in paclitaxel, PB‐Gly‐Taxol and PB‐GABA‐Taxol possess an additional potentially labile ester functional group that links the fluorophore to the core drug structure. For this reason, the stability of these compounds was tested in mouse plasma ex vivo and in vivo. Fresh plasma from wild‐type mice was pooled and stored on ice until the experiment. PB‐Gly‐Taxol or PB‐GABA‐Taxol (1 μM) were added to pooled plasma, incubated for various time points up to 2 h, and the disappearance of parent compounds was measured by LC–MS/MS. Results were calculated as percentages of stock solution prepared at time zero. The in vivo stability of PB‐Gly‐Taxol and PB‐GABA‐Taxol (10 mg/kg) in mice was examined by measuring the presence of paclitaxel as a breakdown product of PB‐Gly‐Taxol and PB‐GABA‐Taxol by a validated LC–MS/MS method as described in the supplemental materials (Figure S1, Table S1). The metabolic ratios were calculated as a measure of stability.

2.6. Statistical Analysis

All data are reported as the mean and standard deviation, and all experiments were performed at least twice. Statistical assessments for mechanical allodynia were done using one‐way ANOVA with Dunnett's post hoc test, and p < 0.05 was considered statistically significant. Statistical analyzes were performed using Prism 10 (GraphPad).

3. Results and Discussion

We previously identified paclitaxel as a substrate of OATP1B/Oatp1b‐type transporters [4]. Because the tissue distribution and primary toxic side effects of this drug depend on these transporters, we examined the transporter specificity of the fluorescent derivatives PB‐Gly‐Taxol and PB‐GABA‐Taxol (Figure 1A,B) as putative probes that might mimic key pharmacological properties of paclitaxel. In HEK293 cells engineered to overexpress human OATP1B1, human OATP1B3, or mouse Oatp1b2, the intracellular uptake of PB‐Gly‐Taxol and PB‐GABA‐Taxol dramatically increased (up to 100‐fold) compared to cells transfected with an empty vector (Figure 1C,D). The cellular uptake of PB‐Gly‐Taxol and PB‐GABA‐Taxol mediated by OATP1B3 was characterized further and found to be completely blocked by pre‐incubation with the transport inhibitor nilotinib (Figure 1E,F). Furthermore, uptake by OATP1B3 and Oatp1b2 is time‐dependent (Figure S2), with intracellular fluorescence signals increasing linearly with time up to at least 15 min. Using 15‐min incubation periods, the uptake was also found to be concentration‐dependent, with values for the Michaelis–Menten constant (K_m) of both compounds of approximately 100 nM for OATP1B3, and a higher K_m of PB‐Gly‐Taxol for Oatp1b2 (Figures 1G,H and S3). These findings indicate that PB‐Gly‐Taxol and PB‐GABA‐Taxol have OATP1B/Oatp1b type transporter interaction profiles that show higher transporter affinity compared to the parent compound paclitaxel (Table S2) [5]. Furthermore, nilotinib‐mediated inhibition of PB‐Gly‐Taxol and PB‐GABA‐Taxol transport by OATP1B3 was independent of substrate concentration, implying a non‐competitive mechanism, with observed inhibition constants (K_i) ranging from 1 to 3 μM (Figure S4, these Ki values were obtained graphically due to limited concentrations of substrates). These values are in line with Ki values of nilotinib reported previously for inhibition of OATP1B3‐mediated transport of 8‐(2‐[fluoresceinyl]‐aminoethylthio)‐adenosine‐3′,5′‐cyclic‐monophosphate (8‐FcA) [10].

FIGURE 1.

Sensitivity of PB‐taxoids to OATP1B/Oatp1b‐mediated transport in vitro. The chemical structures of PB‐Gly‐Taxol (A) and PB‐GABA‐Taxol (B). The intracellular uptake of PB‐Gly‐Taxol (C) and PB‐GABA‐Taxol (D) in HEK293 cells overexpressing human OATP1B1, OATP1B3, and murine Oatp1b2, measured by flow cytometry. The relative uptake (0.1 μM for 10 min) is expressed as a percentage change compared with empty vector controls. Representative micrographs showing differential interference contrast (DIC) and confocal fluorescence microscopy of inhibition uptake of PB‐Gly‐Taxol (E) and PB‐GABA‐Taxol (F) (0.1 μM) by nilotinib (10 μM) in HEK293 cells overexpressing OATP1B3. Scale bar: 10 μm. The concentration kinetic profiles of PB‐Gly‐Taxol (G) and PB‐GABA‐Taxol (H) in OATP1B3 overexpressing cells were determined at various concentrations with 15‐min incubation. The intracellular fluorescence was detected using a plate reader. The velocity of transporter‐mediated uptake was normalized by protein levels. The kinetic parameters, K _m and V _max, were calculated using the Michaelis–Menten function in GraphPad Prism. Data represent mean ± SD of two independent experiments performed in triplicate.

To further characterize the pharmacological properties of PB‐Gly‐Taxol and PB‐GABA‐Taxol, we confirmed their cytotoxicity toward HeLa cells (Figure S5A) and additionally examined their effects on five breast cancer cell lines (Figure S5B–F) and HEK293 cells (Figure S5G). Compared to paclitaxel, both compounds were less cytotoxic, as noted previously [7], but nonetheless, they both possessed substantial cytotoxic activity, with half‐maximal inhibitory concentrations (IC₅₀) consistently being less than 100 nM at 72 h (Figure S5H), although with lower potency compared to paclitaxel. These observations further confirm that PB‐Gly‐Taxol and PB‐GABA‐Taxol functionally mimic key properties of the parent compound, paclitaxel.

To determine whether PB‐taxoids can cause mechanical allodynia in mice, a commonly used readout of peripheral neurotoxicity [11], we administered equivalent doses of PB‐taxoids and paclitaxel to wild‐type mice and observed statistically significant changes in sensitivity to touch over time for all compounds (Figure 2A). The maximum change occurred between 48 and 72 h after drug administration, and readings returned to baseline values within 1 week. Importantly, no mechanical allodynia was observed in Oatp1b2‐deficient mice with either PB‐taxoids or paclitaxel (Figure 2B), implying that, as with paclitaxel [4], the peripheral neurotoxicity associated with single doses of PB‐Gly‐Taxol and PB‐GABA‐Taxol is both reversible and dependent on Oatp1b2‐mediated transport. Ensuing studies confirmed that both PB‐taxoids were stable ex vivo in mouse plasma and cell culture media for at least 2 h (Figure S6A), with no detectable formation of paclitaxel as a metabolite (Figure S6B). Plasma concentrations in mice of both PB‐Gly‐Taxol and PB‐GABA‐Taxol, administered intravenously at equivalent doses, were higher than those observed after administration of paclitaxel (Figure 2C, Table S3). Since 6α‐hydroxylation represents a major pathway of paclitaxel elimination in mice [12], it is conceivable that modification at the C7‐position in PB‐Gly‐Taxol and PB‐GABA‐Taxol with the linker‐fluorophore causes steric hindrance, impairing this mechanism of biotransformation of paclitaxel. The delayed clearance is also consistent with recent studies indicating that the hydroxyl group at the C7 position of paclitaxel plays a catalytic role by facilitating the hydrogen abstraction and rebound steps during the formation of 6α‐hydroxy‐paclitaxel [13]. Regardless of the mechanistic details, no paclitaxel as a metabolite was detectable after the administration of either PB‐taxoids, suggesting that, at the time points examined, the highest rate of formation of paclitaxel from these compounds was less than 0.2% (Figure 2D). This further substantiates the concept that the observed pharmacological effects are intrinsic features of the PB‐taxoids and are unrelated to appreciable biotransformation to paclitaxel.

FIGURE 2.

Characterization of PB‐taxoids in vivo. Mechanical allodynia was measured in wild‐type mice (A) and Oatp1b2‐deficient mice (B) at baseline and up to 1 week post administration of an equivalent single dose of PB‐Gly‐Taxol, PB‐GABA‐Taxol, or Taxol (10 mg/kg, iv). P values represent a one‐way ANOVA with Dunnett's post hoc test on percentage change relative to baseline values (N = 5 per group). *p < 0.05, **p < 0.01. Plasma concentration‐time profiles (C) of PB‐Gly‐Taxol, PB‐GABA‐Taxol and Taxol (10 mg/kg, iv) in wild‐type mice. The in vivo stability (D) of PB‐Gly‐Taxol and PB‐GABA‐Taxol in mouse plasma was determined by measuring the formation of free paclitaxel in mouse plasma over time. Data were represented as mean ± SD (N = 5).

4. Conclusion

Our findings demonstrate that PB‐Gly‐Taxol and PB‐GABA‐Taxol share requisite similarities with the parent compound, paclitaxel. Both compounds displayed significant uptake in cells overexpressing the main mammalian OATP1B/Oatp1b‐type transporters and exhibited cellular uptake kinetics and cytotoxicity comparable to paclitaxel against multiple malignant and normal cell lines. Moreover, nilotinib inhibited the uptake of PB‐taxoids in a non‐competitive manner, resembling the inhibitory profile of this agent reported previously with other substrates [10]. Importantly, PB‐taxoids induced mechanical allodynia only in mice with functional expression of Oatp1b2, highlighting the dependence of the peripheral neurotoxicity phenotypes on this transporter. The stability of PB‐taxoids in mouse plasma and cell culture media ensures that the observed effects stem from the compounds themselves and occur independently of any potential metabolism to paclitaxel. Overall, this study implies that PB‐Gly‐Taxol and PB‐GABA‐Taxol closely mimic the transport and toxicokinetic features of paclitaxel, and that both compounds have the potential to serve as fluorescent imaging tools for exploring drug–drug interaction liabilities and paclitaxel‐related toxicity profiles involving OATP1B/Oatp1b‐type transporters.

Author Contributions

A.S., M.R.N., Y.X., and S.H. wrote the manuscript. B.R.P. and S.H. designed the research. M.R.N., E.A.A., X.H., D.S.R., Y.J., P.H., B.S., and A.E.A. performed the research. M.R.N., Y.X., B.S., and S.H. analyzed the data.

Funding

This project was supported by R01CA238946 (SH), R01CA272254 (SH, BRP), and a College of Pharmacy Dean's Innovation Award (SH, BRP). The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: cts70484‐sup‐0001‐AppendixS1.docx.

Nepal M. R., Xu Y., Jin Y., et al., “Pacific Blue Derivatives of Paclitaxel as Fluorescent Probes of OATP1B‐Type Transporters,” Clinical and Translational Science 19, no. 2 (2026): e70484, 10.1111/cts.70484.

Data Availability Statement

The data generated in this study are available upon request from the corresponding author.