The organic cation/carnitine transporters OCTN1 and OCTN2 are related to other organic cation transporters (OCT1, OCT2, and OCT3) known for transporting oxaliplatin, an anticancer drug with dose-limiting neurotoxicity. In this study, we sought to determine whether OCTN1 and OCTN2 also transported oxaliplatin and to characterize their functional expression and contributions to its neuronal accumulation and neurotoxicity in dorsal root ganglion (DRG) neurons relative to those of OCTs. [14C]Oxaliplatin uptake, platinum accumulation, and cytotoxicity were determined in OCTN-overexpressing human embryonic kidney (HEK) 293 cells and primary cultures of rat DRG neurons. Levels of mRNA and functional activities of rat (r)Octns and rOcts in rat DRG tissue and primary cultures were characterized using reverse transcription-polymerase chain reaction and uptake of model OCT/OCTN substrates, including [3H]1-methyl-4-phenylpyridinium (MPP+) (OCT1–3), [14C]tetraethylammonium bromide (TEA+) (OCT1–3 and OCTN1/2), [3H]ergothioneine (OCTN1), and [3H]l-carnitine (OCTN2). HEK293 cells overexpressing rOctn1, rOctn2, human OCTN1, and human OCTN2 showed increased uptake and cytotoxicity of oxaliplatin compared with mock-transfected HEK293 controls; in addition, both uptake and cytotoxicity were inhibited by ergothioneine and l-carnitine. The uptake of ergothioneine mediated by OCTN1 and of l-carnitine mediated by OCTN2 was decreased during oxaliplatin exposure. rOctn1 and rOctn2 mRNA was readily detected in rat DRG tissue, and they were functionally active in cultured rat DRG neurons, more so than rOct1, rOct2, or rOct3. DRG neuronal accumulation of [14C]oxaliplatin and platinum during oxaliplatin exposure depended on time, concentration, temperature, and sodium and was inhibited by ergothioneine and to a lesser extent by l-carnitine but not by MPP+. Loss of DRG neuronal viability during oxaliplatin exposure was inhibited by ergothioneine but not by l-carnitine or MPP+. OCTN1 and OCTN2 both transport oxaliplatin and are functionally expressed by DRG neurons. OCTN1-mediated transport of oxaliplatin appears to contribute to its neuronal accumulation and treatment-limiting neurotoxicity more so than OCTN2 or OCTs.
Combination chemotherapy based on oxaliplatin plus a fluoropyrimidine (FOLFOX or XELOX) is widely used for palliative first-line systemic therapy for metastatic colorectal cancer and as adjuvant chemotherapy after surgical resection of locally advanced colorectal cancer (de Gramont et al., 2000; André et al., 2004; Cassidy et al., 2004). Addition of oxaliplatin to standard fluoropyrimidine chemotherapy improves objective response rates in previously untreated metastatic colorectal cancer from approximately 20 to 50% (de Gramont et al., 2000) and further reduces risk of colorectal cancer recurrence as adjuvant chemotherapy by 23% (André et al., 2004). Neurotoxicity is a major treatment-limiting side effect of oxaliplatin-based combination chemotherapy, occurring most commonly as chronic sensory neuropathy with glove and stocking paresthesia and dysesthesia, loss of peripheral tendon reflexes, vibration sensation and proprioception, and sensory ataxia (Grothey, 2003). This neurotoxicity cumulates with repeated oxaliplatin treatment, causes long-standing neurological symptoms and deficits, and is a major reason for the withdrawal, reduction, delay, and omission of colorectal cancer treatment (Gamelin et al., 2004; Green et al., 2005).
Platinum accumulation within the dorsal root ganglion (DRG) and its sensory neurons is a major determinant of the neurotoxicity of oxaliplatin (Screnci et al., 1997, 2000; Holmes et al., 1998; Luo et al., 1999; Ta et al., 2006). The biological processes responsible for platinum accumulation in DRG tissue and neurons are poorly understood, although candidate membrane transporters have been identified on the basis of their ability to transport platinum complexes in other cell types (Hall et al., 2008).
OCTN1 (SLC22A4) and OCTN2 (SLC22A5) are organic cation/carnitine transporters (OCTNs) of the SLC22 family that are also known as sodium-dependent ergothioneine and carnitine transporters, respectively (Tamai et al., 1997, 1998). Information in the literature so far concerning the role of OCTNs in the transport of oxaliplatin and the association of OCTN expression with cellular and tissue uptake and toxicity of oxaliplatin and other platinum drugs is limited (Yonezawa et al., 2006). However, oxaliplatin is a known substrate of a closely related class of transporter (Zhang et al., 2006; Koepsell et al., 2007; Yokoo et al., 2008; Burger et al., 2010), the organic cation transporters (OCTs), which also belong to the SLC22 family and include three facilitative transporters, OCT1 (SLC22A1), OCT2 (SLC22A2), and OCT3 (SLC22A3). OCTs have been shown to contribute to the accumulation of platinum in tissues targeted during the treatment-limiting toxicities of cisplatin, such as nephrotoxicity and ototoxicity (Ciarimboli et al., 2005, 2010; Yonezawa et al., 2005; Filipski et al., 2008, 2009), but there have been no previous studies of the expression or functional activity of OCTNs or OCTs in DRG tissue or neurons or of their role in the neuronal accumulation and neurotoxicity of oxaliplatin.
In the present study, we sought to first determine whether OCTN1 and/or OCTN2 transported oxaliplatin by measuring [14C]oxaliplatin uptake and oxaliplatin-induced cytotoxicity and its inhibition of OCTN1-mediated uptake of ergothioneine and OCTN2-mediated uptake of l-carnitine in HEK293 cells overexpressing their rat and human genes, specifically rOctn1, rOctn2, hOCTN1, and hOCTN2, relative to mock-transfected control cells. Then we studied the mRNA expression of rOctn1 and rOctn2 in DRG tissues from adult female Wistar rats compared with that in other normal tissues and that of rOcts. Adult female Wistar rats were used as a source of tissues because of the existence of a large body of published work on the neurotoxicity of oxaliplatin and other platinum drugs in this specific animal model to which transporter expression profiles to be generated could subsequently be related (Screnci et al., 1997, 2000; Cavaletti et al., 2001; Pisano et al., 2003; Ghirardi et al., 2005; Liu et al., 2009). Then, we studied the functional activity of rOctns and rOcts in primary cultures of rat DRG neurons from postnatal female Wistar rats, by measuring uptake of radiolabeled model OCTN and OCT substrates including TEA+, MPP+, l-carnitine, and ergothioneine. Experimental protocols are well established for culturing primary sensory neurons isolated from the DRG of rats (Delree et al., 1989), but the cell yield is very low from these tiny peripheral nervous ganglia, limiting the scale of work that could be undertaken on OCTN- and OCT-mediated transport in DRG tissue. Finally, we studied the cellular accumulation of [14C]oxaliplatin and platinum and the loss of viability of cultured DRG neurons during their exposure to oxaliplatin to determine whether the neuronal transport and neurotoxicity of oxaliplatin could be reduced by inhibition of OCTNs or OCTs.
Materials and Methods
[Ethyl-1-14C]tetraethylammonium bromide (55 Ci/mol), [3H]1-methyl-4-phenylpyridinium (80 Ci/mol), and l-[N-methyl-3H]carnitine hydrochloride (85 Ci/mol) were purchased from American Radiolabeled Chemicals (St. Louis, MO). [3H]Ergothioneine (100 Ci/mol) and [14C]oxaliplatin (54 Ci/mol) were purchased from Moravek Biochemicals (Brea, CA). Unlabeled oxaliplatin was obtained from both Wako Pure Chemicals (Osaka, Japan) and as a gift from Dr. Ryoichi Kizu at Doshisha Women's College (Kyoto, Japan). Other reagents were purchased from Sigma-Aldrich (St. Louis, MO) or Wako Pure Chemicals unless otherwise specified.
Lumbar dorsal root ganglia (L1–L6) were dissected from 20-day-old female Wistar rats euthanized with pentobarbitone, following guidelines from the institutional animal ethics committees, and dissociated DRG neurons were acquired as described previously (Delree et al., 1989). In brief, DRGs were first separated from adherent connective tissue and nerve trunks under microscope, cut into smaller pieces, and digested in collagenase (5 mg/ml)/dispase (5 mg/ml) for 30 min at 37°C. Subsequently, DRG were trypsinized using 0.25% trypsin in Neurobasal A medium (Invitrogen, Carlsbad, CA) containing 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 2 mM glutamine for 30 min at 37°C and then triturated using fire-softened Pasteur pipettes to dissociate tissue structure and release cells. DRG neurons were isolated by Percoll centrifugation (density, 1.040 g/ml) for 20 min at 800g at room temperature. The resultant cell pellet was resuspended in Neurobasal A medium containing 10% horse serum, 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM glutamine, and 1% N-2 supplement. Typically, one rat would yield approximately 30,000 DRG neurons. The cells were cultured at 37°C in a humid atmosphere of 5% CO2-95% air. After 24 h, fresh culture medium containing 20 μM 5-fluoro-2-deoxyuridine and 60 μM uridine was added for 2 days to remove non-neuronal cells from culture. Then, the normal culture medium was replaced every other day until the 6th postdissection day when the DRG neurons were ready for experiment. For uptake studies, 0.5 to 1.0 × 104 cells/well (2.5–5.0 × 103 cells/cm2) were seeded onto a collagen-coated 12-well plate or poly-l-lysine-coated 24-well plate.
HEK293 cells overexpressing rat and human organic cation transporter isoforms HEK/rOctn2, HEK/hOCTN1, and HEK/hOCTN2 were previously established in our laboratory (Tamai et al., 2001; Fujita et al., 2009), whereas HEK/rOctn1 was transfected for this study using methods similar to those described previously (Nakamura et al., 2008). HEK293 cells were cultured using a 10-cm2 culture dish in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 6 mg/ml HEPES, and 1 mM pyruvic acid under a humidified atmosphere at 37°C and 5% CO2-95% air. Cells were split every 3 to 4 days and, for uptake studies, were seeded at 1.5 × 105 cells/well (7.5 × 104 cells/cm2) onto a 24-well plate 2 days before experiments.
A sulforhodamine B (SRB) assay was used to detect the effect of oxaliplatin on cell proliferation of HEK293 cells expressing rOctn1, rOctn2, hOCTN1, and hOCTN2. Cells were seeded at 4000/well (1.1 × 104 cells/cm2) (n = 6) on 96-well plates and allowed to adhere in drug-free medium for 6 h before oxaliplatin exposure as described previously (Fan et al., 2010). In brief, oxaliplatin was serially diluted and added to the cell cultures and incubated for 72 h at 37°C with 5% CO2. Drug exposure was terminated by fixing the cells with 10% trichloroacetic acid solution for 1 h at 4°C, and cellular protein content was determined by staining with 0.57% SRB dissolved in 1% acetic acid for 30 min. Unbound dye was removed by washing with 1% acetic acid, and plates were allowed to air-dry. Bound dye was solubilized in 10 mM Tris base (pH 10) for 30 min, and optical density was measured at 540 nm. Normalized absorbance was plotted against concentration, and IC50 values were generated from curve fits using GraphPad Prism 5 software (GraphPad Software Inc., San Diego, CA).
Cultured DRG neurons were seeded at 5000 cells/well (1.4 × 104 cells/cm2) (n ≥ 6) on collagen-coated 96-well plates and cultured as described previously for 6 days before drug exposure. Oxaliplatin, ergothioneine, l-carnitine, and MPP+ were prepared in assay medium (Neurobasal A medium containing 2% horse serum) and incubated with cultured DRG neurons for 48 h. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was used to determine cell viability. In brief, medium containing drugs was aspirated, replaced with MTT (5 mg/ml) freshly prepared in 50% phosphate-buffered saline (PBS) and 50% Neurobasal A medium containing 10% horse serum, 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM glutamine, and 1% N-2 supplement and incubated for 1 h. The MTT solution was subsequently aspirated, cells were lysed in dimethyl sulfoxide, and optical density was measured at 570 nm.
Adult female Wistar rats (n = 6) were euthanized, and their lumbar DRG, brain, spinal cord, liver, kidney, and intestine were collected. Total RNA (250 ng) was extracted using an Aurum Total RNA Fatty and Fibrous Tissue Kit (Bio-Rad Laboratories, Hercules, CA) and reverse-transcribed to cDNA using a SuperScript First-Strand Synthesis System (Invitrogen) following the manufacturer's protocols. cDNA was subjected to conventional PCR amplification for rOct1, rOct2, rOct3, rOctn1, rOctn2, and rat glyceraldehyde 3-phosphate dehydrogenase (rGapdh) using a C1000 Thermal Cycler (Bio-Rad Laboratories) and gene-specific primers (Invitrogen) (Table 1). The temperature profile was as follows: 5 min at 94°C, and then 40 cycles of 15 s at 94°C, 30 s at 52°C, and 30 s at 72°C. PCR products were electrophoresed on 2% agarose gel, stained with ethidium bromide, and visualized using Gel Doc 2000 (Bio-Rad Laboratories).
cDNA (10 ng) of DRG and reference tissues synthesized from total RNA as described previously was added to TaqMan Universal PCR Master Mix, inventoried FAM-labeled probe for rOct1, rOct2, rOct3, rOctn1, and rOctn2, and VIC-labeled probe for 18S ribosomal RNA (Applied Biosystems, Foster City, CA). Multiplex real-time PCR was performed using a 7900HT Fast Real-Time PCR System and SDS 2.3 software (Applied Biosystems). The fluorescence value at each corresponding cycle number was plotted in SigmaPlot 10 software (Systat Software Inc., Chicago, IL) using a sigmoidal three-parameter curve fitting, and the cycle number at which the fluorescence value was 50% of the maximum was determined as the threshold cycle (Liu and Saint, 2002). Relative mRNA expression levels were determined using the 2−ΔCT method (Livak and Schmittgen, 2001), where CT denotes threshold cycles for each reaction, and ΔCT = CT, OCT − CT, rRNA.
Total protein was extracted from fresh lumbar DRG tissue in a lysis buffer containing 50 mM Tris-HCl, pH 7.4, 250 mM sucrose, 1 mM EDTA, 1 mM EGTA, 0.5% NP-40, 0.1% SDS, and Complete Protease Inhibitor Cocktail Tablets (Roche, Indianapolis, IN). The homogenates were sonicated for 10 min followed by centrifugation for 5 min at 10,000g at 4°C to remove cellular debris, and the protein concentration of the resultant supernatant was determined using the bicinchoninic acid assay as described by Liu et al. (2009). Protein samples (100 μg) were denatured for 5 min at 95°C, separated on 8% SDS-polyacrylamide gel electrophoresis, and transferred onto nitrocellulose membranes. After overnight blocking at 4°C with 2% ECL Advance Blocking Reagent (Amersham Biosciences, Chalfont St. Giles, Buckinghamshire, UK), membranes were incubated overnight at 4°C with polyclonal rabbit antibodies against rOct2, rOct3, and rOctn1 (1:500) (Alpha Diagnostic International, San Antonio, TX) and horseradish peroxidase-conjugated anti-rabbit secondary antibody (1:1000–1:2000) (Amersham Pharmacia, Tokyo, Japan). Immunoreactive bands were visualized by an ECL Advance Western Blotting Detection System (Amersham Biosciences). Membranes were subsequently stripped and reprobed with monoclonal mouse anti-β-actin (1:20,000) (Abcam Inc., Cambridge, MA) and horseradish peroxidase-conjugated anti-mouse secondary antibody (1:5000) (Amersham Biosciences) for the loading control.
Primary culture of DRG prepared on chamber slides as described above was washed with prewarmed PBS and fixed with 4% paraformaldehyde for 15 min at room temperature. The cells were subsequently permeabilized using 0.2% Triton X-100 in PBS for 15 min and treated with Image-iT FX Signal Enhancer (Invitrogen) for 30 min under humid conditions. Next, cells were incubated with blocking solution (PBS containing 0.2% Triton X-100, 3% goat serum, and 2% bovine serum albumin) for 60 min at room temperature with gentle agitation, followed by overnight incubation with primary antibodies, polyclonal rabbit anti-rOct2, rOct3, or rOctn1 (1:500) (Alpha Diagnostic International) diluted in immunobuffer (PBS containing 0.2% Triton X-100 and 3% goat serum) at 4°C. After several washes in PBS containing 0.2% Triton X-100, cells were incubated with the secondary antibody, Alexa Fluor 488-labeled anti-rabbit IgG (H+L) (1:500) (Invitrogen) diluted in immunobuffer for 3 h at 4°C, protected from light. After incubation, cells were washed and cover-slipped with Vectashield mounting medium with 4,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA). Images were captured using a Leica DMR upright fluorescence microscope (Leica Microsystems, Wetzler, Germany) attached to a cooled color Nikon digital camera and analyzed using Nikon ElipseNet (Nikon, Melville, NY) and ImageJ software (National Institutes of Health, Bethesda, MD).
Functional activity of OCTs and OCTNs in primary cultures of DRG was determined by uptake of radiolabeled substrates [14C]TEA+, [3H]MPP+, [3H]ergothioneine, and [3H]l-carnitine. DRG neurons were first washed and preincubated in prewarmed transport buffer containing 125 mM NaCl, 4.8 mM KCl, 5.6 mM d-glucose, 1.2 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, and 25 mM HEPES (pH 7.4) for 10 min. Transport buffer was subsequently aspirated and replaced with fresh transport buffer containing either [14C]TEA+ (0.5 μCi/ml; 9.1 μM), [3H]MPP+ (0.5 μCi/ml; 6.3 μM), [3H]ergothioneine (0.5 μCi/ml; 5.0 μM), or [3H]l-carnitine (0.5 μCi/ml; 5.9 μM) to initiate uptake. At designated times, transport buffer containing radiolabeled substrates was aspirated, and cells were washed three times with ice-cold buffer, air-dried, and lysed in 200 μl of 1% Triton X-100 solution for 6 to 12 h. After cell lysis, 150 μl from each well was collected for quantification of radioactivity using LSC-6100 liquid scintillation counter (Aloka Co Ltd., Tokyo, Japan). Uptake of radiolabeled substrates was presented as the cell/medium ratio (microliters per milligram of protein), the radioactivity accumulated in the cells as a fraction of initial concentration of radioactive compound in the transport buffer. Cellular protein content from each well was determined using a Bio-Rad Protein Assay (Bio-Rad Laboratories).
For characterization of sodium-dependent ergothioneine and l-carnitine uptake, 125 mM NaCl was replaced with equimolar N-methylglucamine.
To determine the effect of oxaliplatin on OCTN1 and OCTN2, a range of concentrations of unlabeled oxaliplatin were coincubated for 3 min at 37°C with [3H]ergothioneine (0.5 μCi/ml; 5.0 μM) and [3H]l-carnitine (0.5 μCi/ml; 5.9 μM) in HEK293 cells overexpressing rat and human OCTN1 and OCTN2, respectively.
Radiolabeled [14C]oxaliplatin (0.05 μCi/ml; 1.0 μM) uptake by HEK293 cells overexpressing rOctn1, rOctn2, hOCTN1, and hOCTN2 was measured with an approach similar to that described above to determine the contribution of rOctn1 and rOctn2 in the transport of oxaliplatin. Uptake of [14C]oxaliplatin (0.1 μCi/ml; 2.0 μM) by cultured DRG neurons was determined for 2 h, during which transport buffer was replaced by minimal essential medium (MEM) supplemented with MEM vitamin solution (100×) and MEM essential amino acids solution (50×) to maintain cell viability during the 2-h incubation.
Platinum uptake by cultured DRG neurons used inductively coupled plasma mass spectrometry (ICP-MS). In brief, cells were incubated in transport buffer containing oxaliplatin (30–100 μM) in the absence or presence of ergothioneine (1 mM), MPP+ (1 mM), or l-carnitine (1 mM) for 30 min at 37°C in 5% CO2 or up to 120 min for platinum uptake kinetics studies. After incubation, cells were washed once with ice-cold PBS and twice with PBS containing 10 mM EDTA and lysed in 70% nitric acid at room temperature for at least 2 h. Platinum content was determined using ICP-MS at LabPLUS (Auckland, New Zealand). Cellular protein content was quantified using a method described previously (Bible et al., 1999).
Results were shown to be reproducible through at least two independent experiments. Data were analyzed using descriptive statistics. The statistical significance of differences between means was determined from 95% confidence intervals or by unpaired Student's t tests in GraphPad Prism 5. P < 0.05 was considered statistically significant. Linear and nonlinear regression analyses were used to determine relationships between experimental variables.
Oxaliplatin Transport and Cytotoxicity in OCTN-Overexpressing HEK293 Cells.
HEK293 cells overexpressing rOctn1, rOctn2, hOCTN1, and hOCTN2 demonstrated higher oxaliplatin uptake than mock-transfected control cells (Fig. 1, a and b). The uptake of [14C]oxaliplatin by HEK293/rOctn1 and HEK293/rOctn2 cells was approximately 2- and 4-fold higher, respectively, after a 1- or 2-h incubation with oxaliplatin compared with that in mock control cells (P < 0.01). The uptake of oxaliplatin by HEK/hOCTN1 and HEK/hOCTN2 cells was approximately 1.5- and 2.5-fold higher compared with that in mock control cells (P < 0.001).
Oxaliplatin inhibited OCTN1-mediated uptake of ergothioneine and OCTN2-mediated uptake of l-carnitine in OCTN-transfected HEK293 cells. During a 3-min exposure, the concentration of oxaliplatin inhibiting ergothioneine uptake by 50% was 3.0 mM in HEK/rOctn1 cells and 0.95 mM in HEK/hOCTN1 cells (Fig. 1, c and d) and that inhibiting l-carnitine uptake by 50% was 3.0 mM in HEK/rOctn2 cells and 8.9 mM in HEK/hOCTN2 cells (Fig. 1, c and d).
HEK293 cells overexpressing rOctn1, rOctn2, hOCTN1, and hOCTN2 were more sensitive to growth inhibition by oxaliplatin than mock-transfected control cells (Fig. 1, e and f; Table 2). The IC50 values for oxaliplatin inhibition of cell growth in HEK/rOctn1 (1.32 μM) and HEK/rOctn2 (0.71 μM) were approximately 2- and 4-fold lower than that in HEK/mock cells (2.93 μM, P < 0.05), respectively (Fig. 1e). The IC50 values for oxaliplatin inhibition of cell growth in HEK/hOCTN1 (2.96 μM) and HEK/hOCTN2 (1.92 μM) were approximately 3- and 4-fold lower than those in control HEK/mock cells (8.09 μM, P < 0.05), respectively (Fig. 1f). Ergothioneine reduced oxaliplatin-induced growth inhibition of HEK293/rOctn1 and HEK293/hOCTN1 cells more so than that of the corresponding mock control cells (Table 2). Acetyl-l-carnitine reduced oxaliplatin-induced growth inhibition of HEK293/rOctn2 but not that of HEK293/hOCTN2 or mock cells (Table 2).
Functional Expression of rOctns in rat DRG Neurons.
rOctns displayed a specific pattern of functional expression in rat DRG tissue and primary culture. Rat DRG tissue showed readily detectable mRNA for Octn1 and Octn2 (Fig. 2, a and b; Table 3), comparable to reference tissues including liver, kidney, and intestine. Octn1 protein was detectable by Western blotting (Fig. 3a) localized to DRG neuronal cell bodies (Fig. 4, a and b). Primary cultures of rat DRG neurons showed linear uptake of the model Octn1 substrate ergothioneine during a 1-h exposure at 37°C that was higher than that at 4°C (5.4 versus 1.0 μl/mg protein/h, P < 0.05) (Fig. 2c). Cultured DRG neurons showed linear uptake of the Octn2 model substrate l-carnitine during a 1-h exposure at 37°C that was higher than that at 4°C (2.4 versus 0.2 μl/mg protein/h, P < 0.05) (Fig. 2d). [3H]Ergothioneine uptake by cultured DRG neurons was sodium-dependent and inhibited by unlabeled ergothioneine (P < 0.05) but not by acetyl-l-carnitine, l-carnitine, or MPP+ (P > 0.05) (Fig. 2e). l-Carnitine uptake by cultured DRG neurons was sodium-dependent and significantly inhibited by approximately 2-fold by known OCTN2 substrates including unlabeled l-carnitine, acetyl-l-carnitine, and MPP+ (P < 0.05) but to a lesser extent by ergothioneine (Fig. 2f).
In contrast, the organic cation transporters Oct1, Oct2, and Oct3 displayed lower transport activity and lower or undetectable mRNA levels in rat DRG tissue and primary cultures. Oct1 mRNA was undetectable in DRG tissue but was readily detected in the reference liver tissue (Fig. 5a), as reported previously (Koepsell et al., 2007). Oct2 mRNA expression was only faintly detectable in DRG tissue at levels almost 200-fold lower than that in kidney, a known Oct2-expressing tissue (Koepsell et al., 2007). Oct3 mRNA was detected in DRG tissue, but the expression level was lower than that of Octn1 and Octn2 and approximately 6-fold lower than that in intestine. Oct2 and Oct3 protein was detectable in DRG tissue by Western blotting when immunoreactive bands were detected between 60 and 70 kDa, corresponding to a predicted molecular mass of 66 and 61 kDa for rOct2 (Fig. 3b) and rOct3 (Fig. 3c), respectively. In DRG cultures, immunoreactivity to rOct2 (Fig. 4c) and rOct3 (Fig. 4d) was evident in DRG neuronal cell bodies as revealed by immunocytochemistry but apparently in fewer cells than for Octn1 (Fig. 4b). The uptake of MPP+, a model substrate of Oct1, Oct2, and Oct3 (Koepsell et al., 2007), by cultured DRG neurons was similar at 37 and at 4°C (7.2 versus 4.2 μl/mg protein/h, P > 0.05) (Fig. 5d), consistent with a transporter-independent process. The DRG neuronal uptake of TEA+, a model substrate of Oct1–3 and Octn1–2 (Koepsell et al., 2007), was numerically different at 37 and 4°C (0.9 versus 0.3 μl/mg protein/h), but this difference did not reach statistical significance overall and individually only for the final time point (Fig. 5e).
rOctn-Mediated Uptake and Toxicity of Oxaliplatin in Rat DRG Neurons.
Next, we showed that the uptake and toxicity of oxaliplatin in primary cultures of rat DRG neurons was mediated by rOctns. Uptake of [14C]oxaliplatin by cultured rat DRG neurons was reduced at low temperature (17% of control; P < 0.001) and in the absence of sodium (79% of control; P < 0.01) and was inhibited by ergothioneine (46% of control; P < 0.001) but not by l-carnitine (Fig. 6a). The accumulation of platinum by cultured rat DRG neurons during exposure to oxaliplatin showed a nonlinear dependence on time and oxaliplatin concentration (Fig. 6b). Platinum accumulation was also dependent on temperature (32% of control; P < 0.001) and inhibited by ergothioneine (63% of control; P < 0.001) and to a lesser extent by l-carnitine (90% of control; P < 0.05), but not by MPP+ (107% of control; P > 0.05) (Fig. 6c). Cultured rat DRG neurons lost viability during exposure to oxaliplatin at 30 μM (67% of control, P < 0.001) and 100 μM (50% of control, P < 0.001) for 48 h. The loss of neuronal viability induced by oxaliplatin was prevented by addition of ergothioneine to the culture medium (30 μM oxaliplatin, 102 versus 67%, P < 0.001; 100 μM oxaliplatin, 97 versus 50%, P < 0.001) but was unaffected by addition of l-carnitine or MPP+ (Fig. 6d). However, oxaliplatin inhibited the uptake of both ergothioneine and l-carnitine by cultured rat DRG neurons (Fig. 6e).
This study demonstrates that oxaliplatin is transported by rat and human OCTN1 and OCTN2. Direct evidence was provided by studies of isogenic HEK293 cell lines overexpressing rOctn1, rOctn2, hOCTN1, and hOCTN2, which accumulated more oxaliplatin than mock-transfected control cells. Isogenic HEK293 cells overexpressing OCTN1 and OCTN2 showed increased sensitivity to the growth inhibitory effect of oxaliplatin compared with mock-transfected control cells. Furthermore, oxaliplatin inhibited OCTN1-mediated uptake of ergothioneine and OCTN2-mediated uptake of l-carnitine, consistent with oxaliplatin interacting directly with OCTN1 and OCTN2. Taken together, these findings suggested that oxaliplatin is a competitive substrate for OCTN1 and OCTN2 and that OCTN1 and OCTN2 determine the sensitivity of some cell types to oxaliplatin.
Next, the study investigated whether OCTN1 and/or OCTN2 mediated the neuronal uptake and neurotoxicity of oxaliplatin. Neurotoxicity is the dose-limiting clinical side effect of oxaliplatin (Grothey, 2003). In previous studies, oxaliplatin neurotoxicity was associated with accumulation of platinum by DRG tissue and its sensory neurons in female Wistar rats or related animal models (Screnci et al., 1997, 2000; Holmes et al., 1998; Luo et al., 1999; Ta et al., 2006). In the current study, DRG tissue and primary cultures from female Wistar rats were first shown to have significant mRNA expression and functional activity of rOctn1 and rOctn2. Then, oxaliplatin was found to inhibit the DRG neuronal uptake of ergothioneine and l-carnitine at high concentrations of oxaliplatin that, during these short exposures (3 min), did not affect the viability or membrane integrity of DRG neurons. The neuronal uptake of oxaliplatin was shown to depend on temperature and the presence of sodium, which are features of OCTN-mediated transport (Tamai et al., 1998; Nakamura et al., 2008), and their accumulation of platinum showed a nonlinear dependence on time and concentration of oxaliplatin exposure. Next, ergothioneine and, to a lesser extent, l-carnitine, but not MPP+, were shown to inhibit the uptake of oxaliplatin, accumulation of platinum, and/or oxaliplatin-induced loss of viability in cultured rat DRG neurons. Taken together, these findings suggest that OCTN1-mediated transport is an important mechanism contributing to the neuronal accumulation and resulting neurotoxicity of oxaliplatin more so than that mediated by OCTN2 or OCTs.
The individual contributions of Octn1 and Octn2 to the uptake of oxaliplatin appeared to be different, depending on the different cell types used in the current study. In the HEK293 cells, OCTN2 conferred greater oxaliplatin uptake than OCTN1, resulting in increased sensitivity to oxaliplatin. Our previous studies showed that OCTNs localized to the plasma membranes of overexpressing HEK293 cell consistent with their role in the transport of organic cations into cells in this model (Tamai et al., 2001; Kawasaki et al., 2004; Sugiura et al., 2006). In contrast, oxaliplatin uptake and toxicity in DRG neurons from female Wistar rats were inhibited more by ergothioneine (rOctn1) than by l-carnitine (rOctn2). These apparent differences in the contributions of Octn1 and Octn2 to the uptake of oxaliplatin may be due to differences in the relative activities of Octn1 and Octn2 between these different cell types, because rOctn1 appeared to be more highly expressed and functionally active than rOctn2 in rat DRG. Differences in their respective binding affinities for ergothioneine (Km = 4.6 μM) and l-carnitine to (Km = 25.4 μM) (Sekine et al., 1998; Nakamura et al., 2008) may also have contributed to their apparent differing activities. Differences in the abundance of their respective proteins and their localization within the DRG neuron may have contributed to their differing activities but were not defined in the current study. Selective gene knockdown by RNA interference, although a conceptionally attractive approach for deconvoluting the contribution of individual transporters, may not be achievable in primary cultures of DRG neurons because of their high sensitivity to off-target toxicities of RNA interference reagents (Read et al., 2009; Ehlert et al., 2010).
Membrane proteins other than OCTNs have been implicated in the transport and toxicity of platinum drugs in other cell types, including the organic cation transporters OCT1, OCT2, and OCT3 (Ciarimboli et al., 2005; Yonezawa et al., 2005, 2006; Zhang et al., 2006; Filipski et al., 2008, 2009; Burger et al., 2010; Ciarimboli et al., 2010) and the copper transporters CTR1 (Ishida et al., 2002), ATP7A (Samimi et al., 2004) and ATP7B (Komatsu et al., 2000). In the current study, levels of mRNA in rat DRG tissue for OCT1, OCT2, and OCT3 were lower than the readily detectable levels for OCTN1 and OCTN2. The uptake of MPP+, a known substrate of OCT1, OCT2, and OCT3 (Koepsell et al., 2007), was the same at 37 and 4°C, consistent with a lack of functional expression of these OCTs in rat DRG. Likewise, uptake of TEA+, a model compound transported by all OCTs but with greater affinity to OCT1, OCT 2, and OCT3 than to OCTN1 or OCTN2 (Koepsell et al., 2007), was not significantly different at 37 and 4°C. This apparent lower functional expression of rOct1, rOct2, and rOct3 compared with that of rOctn1 and rOctn2 by rat DRG neurons suggests that OCTs may not have a significant role in mediating the neuronal uptake and neurotoxicity of oxaliplatin, at least in tissues and cells from the female Wistar rats that we examined. However, OCT expression is known to influenced by many factors, including animal species, gender, strain, disease, and exposures (Koepsell et al., 2007); therefore, our study does not rule out the possibility of OCTs having a significant role in the uptake of oxaliplatin by DRG neurons in contexts other than as explicitly studied. We recently showed a specific pattern of expression of copper transporters in rat DRG tissue (Ip et al., 2010) and that CTR1-expressing DRG neurons become atrophied after oxaliplatin treatment (Liu et al., 2009). The relative contributions of copper transporters, OCTs, and OCTNs to the neurotoxicity of oxaliplatin remain to be determined definitively. However, the findings of the current study suggest that OCTN1 may be responsible for approximately 50% of the uptake of oxaliplatin by rat DRG neurons, with a lesser contribution from OCTN2, and no significant uptake mediated by OCT1, OCT2, or OCT3, at least in the restricted context in which these studies were performed.
The new information we now report about OCTN-mediated mechanisms of the neuronal uptake of oxaliplatin could be applied to the development of strategies for improving patient outcomes from oxaliplatin-based cancer chemotherapy. In this article, we provide the first preclinical proof of OCTN-related mechanisms of oxaliplatin neurotoxicity and a rationale for targeting OCTNs with pharmacological inhibitors for the purpose of lowering this dose-limiting toxicity. We showed that a known OCTN1 competitive substrate, ergothioneine (Nakamura et al., 2008), reduced the uptake of oxaliplatin by cultured rat DRG neurons and attenuated their accumulation of platinum and loss of viability during in vitro exposure to oxaliplatin. OCTN substrates have been associated with a reduction in chemotherapy-induced neurotoxicity in previous studies (Pisano et al., 2003; Ghirardi et al., 2005; Song et al., 2010), but their exploratory findings occurred in the absence of the insights we now provide into OCTN transport-related neurotoxic and neuroprotective mechanisms. Whether oxaliplatin-induced neurotoxicity can be prevented in whole animals or human subjects using strategies based on inhibiting OCTN-mediated neuronal transport, without causing undue side effects or attenuating the antitumor chemotherapeutic activity of oxaliplatin, remains to be determined in future studies. However, this approach is supported by several additional considerations and data. First, the Octn1 knockout mouse is viable and has no obvious phenotypic abnormality (Kato et al., 2010), suggesting that OCTN1 inhibition may be tolerable. Second, tissue-selective expression profiles of oxaliplatin transporters such as that described in this article could be exploited to selectively reduce oxaliplatin-induced neurotoxicity because OCTs rather than OCTNs appear to contribute to oxaliplatin uptake and antitumor activity in colorectal carcinoma (Zhang et al., 2006; Yokoo et al., 2008), the major current clinical indication for oxaliplatin. Last, only transient and intermittent OCTN1 inhibition may be required to protect against oxaliplatin-induced neuropathy because oxaliplatin has a very short plasma half-life (Ip et al., 2008) and is given only once every 2 to 3 weeks. However, if OCTN1 is involved in the excretion of oxaliplatin, its inhibition might result in a pharmacokinetic drug-drug interaction leading to increased toxicity.
In conclusion, OCTN1 and OCTN2 both transport oxaliplatin and are functionally expressed by DRG neurons. OCTN1-mediated transport of oxaliplatin appears to contribute to its neuronal accumulation and treatment-limiting neurotoxicity more so than OCTN2 or OCTs.
Participated in research design: Jong, Nakanishi, Liu, Tamai, and McKeage.
Conducted experiments: Jong and Liu.
Performed data analysis: Jong, Nakanishi, Liu, Tamai, and McKeage.
Wrote or contributed to the writing of the manuscript: Jong, Nakanishi, Liu, Tamai, and McKeage.
We thank the research technician, Yaeseul Kim, and Dr. Ryoichi Kizu for kindly supplying the oxaliplatin.
This work was supported by the Auckland Medical Research Foundation; the Cancer Society of New Zealand; Kanazawa University Special admission quota for graduate students from Designated Universities Special Selection for the Enhancement of International Exchange Scholarship; and Research Activity Start-up.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- dorsal root ganglion
- organic cation/carnitine transporter
- solute carrier
- organic cation transporter
- human embryonic kidney
- sulforhodamine B
- 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
- reverse transcription
- polymerase chain reaction
- glyceraldehyde-3-phosphate dehydrogenase
- minimal essential medium
- inductively coupled plasma mass spectrometry
- [14C] tetraethylammonium bromide
- base pairs.
- Received March 7, 2011.
- Accepted May 20, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics