Systemic Delivery and Biodistribution of Cisplatin in Vivo

Abstract

Cisplatin is widely used to treat a variety of cancers. However, ototoxicity and nephrotoxicity remain serious side effects of cisplatin-based chemotherapy. In order to inform the study of cisplatin’s off-target effects, a new drug-fluorophore conjugate was synthesized that exhibited utility as a tracer to determine the cellular uptake and in vivo distribution of cisplatin. This probe will serve as a useful tool to facilitate investigations into the kinetics and biodistribution of cisplatin and its associated side effects in preclinical models after systemic administration.

Graphical abstract

INTRODUCTION

Cisplatin (cis-diamminedichloroplatinum[II]) is a platinum-containing anti-cancer drug, the bioactivity of which was discovered by Rosenberg et al. in 1965.¹ It has become a cornerstone of antineoplastic chemotherapy to treat many types of cancer, including ovarian, cervical, stomach, bladder, head, and neck.^2–6 Cisplatin is especially efficacious for testicular cancer, which has an overall cure rate greater than 90% and nearly 100% at stage I.⁵ There are more than 1 million patients receiving cisplatin or its derivatives in North America and Europe every year. Unfortunately, the clinical use of cisplatin is limited by acquired drug resistance and severe dose-dependent side effects, such as ototoxicity and nephrotoxicity. Clinical studies have shown that, in >60% of the patients who have taken multiple doses of cisplatin, permanent hearing loss^7,8 and acute renal dysfunction were significant side effects.⁹ Drug-induced hearing loss can severely affect an individual’s quality of life. Consequences can include social isolation, depression, and loss of income. Children can exhibit delayed development of communication and learning skills, and have a relatively challenging time developing social interactions.^7,8 Therefore, an understanding of the events that lead to cisplatin-induced ototoxicity and nephrotoxicity can have a profound impact on human health, and can enable the design of potentially less harmful therapeutic alternatives.

The mechanism of cisplatin uptake is poorly understood. There have been several prior studies to identify the biodistribution of cisplatin in cells and tissues. These include the use of positron emission tomography, radiolabeled cisplatin and high-performance liquid chromatography (HPLC).^10–12 Morphologically, these techniques have relatively low anatomic resolution and cannot provide the cellular distribution of cisplatin in heterogeneous cellular tissues, such as the cochlea in the inner ear. In order to investigate the cellular uptake and intracellular distribution of cisplatin, the use of fluorogenic probes is a promising alternative to these techniques.

Prior investigations of fluorophore–cisplatin conjugates and their cellular uptake have been reported. For example, a fluorescein–cisplatin conjugate (1, Figure 1) entered cells via endocytosis.¹³ A tetramethylrhodamine–cisplatin conjugate entered the sensory hair cells of the zebrafish lateral line via their apical membranes, potentially via the mechanoelectrical transduction channel-dependent mechanism or calmodulin-dependent endocytosis.¹⁴ In addition, other conjugates, including cyanine, coumarin, and dinitrophenyl fluorophores, have been studied.¹⁵ Results of these prior studies have suggested that a cationic charge on the fluorophore can enhance electrostatic interactions of the cisplatin conjugate toward a guanine base. Therefore, we hypothesized that conjugation of cisplatin to a cationic dye will better preserve its desired bioactive properties, thus enabling in vivo studies of its cellular trafficking mechanisms involving ion channels. This hypothesis is also based on related investigations of other cationic pharmaceuticals, such as the aminoglycosides, which are also known to cause ototoxicity and nephrotoxicity.¹⁶ Specifically, it was envisioned that tagging cisplatin with a rhodamine-type fluorophore, akin to gentamicin–Texas Red,^17,18 would better preserve cisplatin’s properties related to its cationic charge as compared to conjugation to a negatively charged fluorophore.¹⁶

Figure 1.

Top: Structures of cisplatin conjugates 1 and 2. Bottom: Synthesis of 2.

Herein, cisplatin was conjugated to the commercially available fluorophore Texas Red to afford 2 (Figure 1). Following systemic injection, its distribution in fixed rodent tissues, including the cochlea, was determined using confocal microscopy. This is the first example of the successful use of a fluorescently tagged cisplatin probe in mammals, to the best of our knowledge. Compound 2 exhibited an unprecedented degree of functionality. For example, the compound crossed the blood–labyrinth barrier and entered the blood–brain barrier at the choroid plexus; therefore, it has substantial potential utility as a tracer.

RESULTS AND DISCUSSION

Platinum complex 3 (Figure 1) was selected as a target as it contains a short linker for conjugation and retains the chemical properties of cisplatin, as reported by Karasawa et al.¹⁹ The target cisplatin–Texas Red conjugate (2) was synthesized from the reaction of Texas Red N-succinimidyl ester (TR-SE, mixture of regioisomers) (4) with platinum complex 3 under basic conditions in 88% yield with 97% purity (Supporting Information).

In order to determine that 2 had spectral properties similar to those of Texas Red-derived 4, absorbance and fluorescence spectra were evaluated. As shown in Figure 2, conjugate 2 exhibited similar maximum absorption intensities as compared to 4, with the fluorescence excitation spectra slightly red-shifted by 5 nm in 2. Relatively negligible changes in the emission spectra were observed, and, as expected, there was no significant improvement in the quantum yield of 4 compared to 2.

Figure 2.

Top: Absorption spectra. Bottom: Excitation/emission fluorescence spectra of 1.25 μM solutions in MeOH of compounds 2 and 4.

To verify that 2 retained the cytotoxic properties of cisplatin, the relative toxicity values of cisplatin and 2 were compared using zebrafish neuromast hair cells. Larvae were treated with a dose range of 2 or 4 for 4 h and allowed to recover for 3 h.²⁰ Alexa Fluor 488-conjugated phalloidin labeling of the hair bundle and cuticular plate was used to determine the number of surviving hair cells. Hair cells can cleave their hair bundles following drug exposure, yet the loss of the hair bundle or cuticular plate is an ototoxic event. After treatment with 2 (10–200 μg/mL) for 4 h, plus 3 h of recovery, the number of surviving hair cells decreased in a dose-dependent manner (Figures 3A–C,E). Treatment with 4 revealed negligible fluorescence in the hair cells, suggesting that the forced uptake of Texas Red, via conjugation to cisplatin, was responsible for the enhanced toxicity of 2 (Figure 3D). To quantify and normalize this decrease, untreated controls provided a baseline to determine the ratio of surviving hair cells in zebrafish neuromasts. Both cisplatin and 2 decreased hair cell numbers in a dose-dependent manner (Figure 3E).

Figure 3.

Cytotoxicity in zebrafish neuromast hair cells with a range of dose after 3 h recovery. (A–C) Neuromast hair cells treated with various concentrations of conjugate 2 (50–200 μg/mL) for 4 h. Phalloidin labeling (green) revealed the actin cytoskeleton of individual cells. (D) Treatment with 4. (E) Quantification of surviving zebrafish neuromast hair cells after treatment with cisplatin or conjugate 2 (0–100 μg/mL). N ≥ 15 neuromasts per dose, error bars = 95% confidence intervals.

To determine the in vivo distribution of 2, rats were injected intravenously with 2 mg/kg of 2. In the renal cortex, proximal, but not distal cells avidly took up diffuse and punctate fluorescence within 1 h (Figure 4A). By 3 h, the intensity of cytoplasmic and punctate fluorescence had diminished, indicating clearance, which was further evident at 24 h (Figure 4B,C). Similar uptake and clearance kinetics were also seen in liver hepatocytes (Figure 4D,E). In the choroid plexus, a double endo- and epithelial barrier in the blood–brain barrier, 2 was diffusely dispersed in the cytoplasm of ependymal cells within 1 h (Figure 4G). By 3 h, fluorescent puncta occurred within the laden cytoplasm (Figure 4H). After 24 h, more intense puncta were present within the laden cytoplasm (Figure 4I), indicating an apparent lack of cellular clearance.

Figure 4.

Uptake and clearance of 2 (red) in kidney (A–C), liver (D,E), and choroid plexus cells (G–I) over time. (F) Liver tissues of rats treated in the absence of 2, showing only weak punctate autofluorescence. Alexa Fluor 488-conjugated phalloidin (green) revealed the actin cytoskeleton of individual cells. p, proximal tubule cells; d, distal tubule cells.

We then tested whether 2 crossed the cochlear blood–labyrinth barrier (akin to the blood–brain barrier) into the stria vascularis (SV) and entered hair cells. In cryostat sections of the cochlea, at 1 and 3 h (Figure 5) after injection, fluorescence was brightest in the SV in the lateral wall of the cochlea, compared to other cochlear locations, as for fluorescently tagged gentamicin.¹⁸ Renal and cochlear tissues did not exhibit any fluorescence from hydrolized forms of 4 (data not shown) as previously described.^18,21 In whole mounted tissues of the cochlear lateral wall, 1 h after injection, fluorescence was visible in strial marginal cells, intrastrial tissues (particularly capillary endothelial cells), and basal cells, with negligible fluorescence in spiral ligament (SL) fibrocytes (Figure 6A–D). After 3 h, the intensity of 2 in strial cells was increased, but less so in fibrocytes (Figure 6E–H). After 24 h, clearance was evident in strial tissues (Figure 6I–K), while fibrocytes continued to take up 2 (Figure 6L). The clearance of 2 was statistically significant at later times compared to earlier time points (Figure 7), as for fluorescently tagged gentamicin.¹⁸

Figure 5.

Three hours after injection of 2, fluorescence is most prominent in the stria vascularis (SV). SL, spiral ligament; OoC, organ of Corti; SpL, spiral limbus; M, modiolus.

Figure 6.

Uptake and clearance of 2 in marginal cells (A,E,I), intrastrial tissues (B,F,J), basal cells (C,G,K), and spiral ligament fibrocytes (D,H,L) over time.

Figure 7.

Uptake and clearance of 2 in marginal cells, intermediate cells, basal cells, and spiral ligament fibrocytes over time. *, significance difference (P < 0.05) versus all other time points (except with marginal cells at 24 h); error bars = standard error of the mean, N ≥ 3.

We then examined the uptake of 2 by cochlear hair cells in the organ of Corti. One hour after injection, outer hair cells (OHCs) displayed weak fluorescence, particularly on their stereocilia (Figure 8A). Inner hair cells (IHCs) did not exhibit fluorescence at 1 h after injection (Figure 8B), although punctate fluorescence was visible (in supporting cells surrounding IHCs, Figure 8C), and identified as nonspecific autofluorescence (likely lipofuscin) as previously described.²¹ Three hours after injection, OHC and IHC stereocilia and cell bodies displayed diffuse fluorescence from 2 (Figure 8D, E), and this fluorescence intensity was maintained 24 h later (Figure 8F, G). These data demonstrated that 2 crossed the blood–labyrinth barrier and entered cochlear hair cells of healthy rats, reminiscent of previous studies using Texas Red-conjugated gentamicin.^18,22

Figure 8.

Uptake and clearance of 2 in OHCs (A,D,F) and IHCs (B,E,G) over time. (C) Note the punctate autofluorescence (at 568 nm) in supporting cells adjacent to IHCs (not visible) of rats treated without 2 or 4.

Using induction-coupled plasma mass spectrometry, we found that the serum kinetics of cisplatin and 2 were similar following i.v. injection. The serum half-life (t_1/2) of total 2 was calculated to be 6.8 h from 60 min to 24 and 26.1 h from the 24-h to 5-day time points. The t_1/2 for native cisplatin in rats was initially 6.3 h,²³ and then 14.7 h between 12 and 48 h.²⁴ Thus, the serum kinetics of 2 was similar to the biphasic kinetics of cisplatin, although with a longer secondary t_1/2 than published studies, perhaps due to the removal of the hydration step²⁵ in our protocol, as well as the larger molecular mass of the conjugate (~1000) compared to native cisplatin (~300).

CONCLUSION

Cisplatin–Texas Red conjugate 2 was synthesized using platinum complex 3 and commercially available TR-SE 4 in 97% purity. Spectral evaluation showed that the cisplatin conjugate 2 essentially retained the spectral characteristics of the parent probe 4.

Cisplatin is a relatively small molecule; thus, conjugation could alter the kinetics of its uptake as well as its cytotoxicity. Cellular studies revealed that Texas Red conjugate 2 successfully undergoes uptake and exhibits cytotoxicity to hair cells similar to that of native cisplatin. Moreover, conjugate 2 enters and is cleared from a variety of mammalian organs, in addition to crossing the blood–labyrinth barrier to enter cochlear hair cells in rats. The cellular uptake and in vivo distribution of 2 largely replicate our previous studies using Texas Red-conjugated gentamicin.¹⁸ Despite a larger molecular mass (~1000), 2 crosses an intact blood–labyrinth barrier and enters cochlear hair cells within 1 h. Systemically administered 2 was preferentially associated with the SV compared to the spiral ligament and organ of Corti. Previous studies have shown cisplatin toxicity in the SV, with concomitant loss of cochlear function, implying strial uptake of cisplatin.²⁶ This conjugate will therefore facilitate investigations into the kinetics and biodistribution of cisplatin, and its associated side effects, in preclinical models after systemic administration, as for fluorescently tagged gentamicin.²⁷ One way to reduce the volume of distribution of cisplatin in vivo is to attach cisplatin to polymers, nanoparticles, or monoclonal antibodies that can more specifically target tumors compared to systemic delivery of native cisplatin.^28,29 This can be particularly effective if the binding ligand is labile and cleaved in the vicinity of the tumor or within cancer cells.³⁰ However, polymers can often be entrapped within the fibrous sheath surrounding the tumor, or otherwise heterogeneously distributed throughout the tumor.^31,32 These approaches, however, may ameliorate cisplatin-induced ototoxicity if they retain their tumoritoxic potential. The fluorescence tagging of cisplatin was designed to determine the cellular distribution of cisplatin within tissues with heterogeneous cell types, like the cochlea, to determine which cells had a preferential uptake of the conjugate, and thereby of the native drug, akin to similar studies for gentamicin and tagged gentamicin.^18,27

In summary, conjugate 2 facilitates the study of the origin of cisplatin’s side effects in mammals. Prior studies with fluorescently tagged cisplatin have been limited to in vitro investigations. In heterogeneous structures such as the cochlea and other biological tissues, conjugate 2 enables the microscopic analysis of cisplatin’s biodistribution within tissues, complementing related techniques, such as ICP-MS, for native cisplatin.

EXPERIMENTAL SECTION

General

All chemicals were used without further purification. ¹H NMR, ¹³C NMR, and ¹⁹⁵Pt NMR were acquired on an ARX-400 Avance Bruker spectrometer. Chemical shifts (δ) are given in ppm relative to TMS and DMF-d₇ was used as solvent. For ¹⁹⁵Pt-NMR, K₂PtCl₄ was used as external standard. Mass spectra was conducted at the PSU Bioanalytical Mass Spectrometry Facility on a ThermoElectron LTQ-Orbitrap Discovery high resolution mass spectrometer coupled to an Accela HPLC system. Another HPLC system was used for the spectrophotometric detection consisting on a 1525 binary delivery module (Waters), and a 2996 photodiode array detector (Waters). A Supelco Discovery C-18 (25 cm × 2.1 mm, 5 μm) column was used for the separation in both HPLC systems. Fluorescence measurements were carried out on a Cary Eclipse fluorescence spectrophotometer (Agilent Technologies), and absorption measurements were performed on a Cary Eclipse 50 Bio UV–vis spectrophotometer.

Synthesis of Cisplatin–Texas Red Conjugate, 2

Platinum complex 5 (10.55 mg, 26.93 μmol) was dissolved in 1.2 mL of anhydrous DMF and sonicated for 10 min. Et₃N (10.24 μL, 73.44 μmol) was added in one portion and the mixture sonicated for another 5 min, resulting in a clear pale yellow solution; if necessary, DI H₂O was added in increments of 20 μL and sonicated until completely dissolved. Compound 4 (TR-SE, 20 mg, 24.48 μmol; Invitrogen, CA) was added, and the reaction mixture stirred at rt for 1 h prior to evaporation under vacuum to afford a purple precipitate. The residue was centrifuged and washed with cold H₂O (2 × 10 mL) and Et₂O (3 × 10 mL) until the supernatant became almost clear, and dried under vacuum to afford a purple solid (yield: 88%, 22.8 mg). HRMS (ESI): m/z [M+H]⁺ calculated for C₄₀H₅₁Cl₂N₆O₇PtS₂: 1056.2282; observed 1056.2354.

Zebrafish

Wild-type zebrafish larvae, 5 days after fertilization, were treated with a dose range of 2 or native cisplatin (10–400 μg/mL in standard E3 medium) for 4 h and recovered for 3 h prior to fixation in 4% formaldehyde containing 0.5% Triton X-100.²⁰ Larvae were then labeled with Alexa Fluor 488-conjugated phalloidin to determine the number of surviving hair cell bundles, as described previously.¹⁸

Rats

Male Long–Evans rats (~200 g) received one i.v. injection of 2 mg/kg of compound 2 (in a 60/40% mixture of EtOH/PBS, pH 7.4), and at various time points (0.5, 3, 6, 24, and 96 h) were anesthetized, and serum was collected prior to cardiac perfusion with phosphate-buffered saline (PBS), followed by 4% formaldehyde containing 0.5% Triton X-100. Choroid plexi, cochleae, kidneys, and liver tissues were excised and fixed overnight. Alternatively, rats received an equivalent dose of 4 prior to the above procedures. All tissues were counter-labeled with Alexa Fluor 488-conjugated phalloidin to label filamentous actin. All specimens were whole-mounted in VectaShield (Vector Labs, CA) and observed using a Bio-Rad MRC 1024 ES laser scanning confocal system attached to a Nikon Eclipse TE300 inverted microscope.¹⁸ For cryostat sections, cochleae were decalcified in 10% EDTA, infiltrated with OCT, cryostat sectioned at 8 μm, and mounted on gelatin-coated glass slides.¹⁸ Alexa Fluor 488 and Texas Red images were collected sequentially. For each dosing regimen, all specimens were imaged at the same laser intensity and gain settings, including control tissues. Representative images from each dose regimen were identically prepared for publication using Adobe Photoshop.¹⁸ Platinum in cisplatin and 2 was detected using isocrative HPLC and ICP-MS (running in time-resolved mode).³³

ABBREVIATIONS

TR-SE

Texas Red N-succinimidyl ester

HPLC

high-performance liquid chromatography

DMF

N,N-dimethylformamide

i.v

intravenous

SV

stria vascularis

SL

spiral ligament

OHCs

outer hair cells

IHCs

inner hair cells

PBS

phosphate-buffered saline