An Investigation into Retigabine (Ezogabine ... - ACS Publications

Subscriber access provided by Iowa State University | Library

Article

An Investigation into Retigabine (Ezogabine) Associated Dyspigmentation in Rat Eye by MALDI Imaging Mass Spectrometry (IMS) M. Reid Groseclose, and Stephen Castellino Chem. Res. Toxicol., Just Accepted Manuscript • DOI: 10.1021/acs.chemrestox.8b00313 • Publication Date (Web): 14 Jan 2019 Downloaded from http://pubs.acs.org on January 15, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

An Investigation into Retigabine (Ezogabine) Associated Dyspigmentation in Rat Eye by MALDI Imaging Mass Spectrometry (IMS) M. Reid Groseclose and Stephen Castellino* GSK, Department of Bioimaging, 1250 S Collegeville Rd, Collegeville, PA 19426 KEYWORDS: retigabine, ezogabine, Potiga, imaging mass spectrometry, MALDI, drug distribution, preclinical toxicology, melanin binding, ocular tissue, uveal tract, retina, and dyspigmentation

ABSTRACT: Retigabine (RTG) is an antiepileptic drug approved as an adjunctive treatment for refractory partial-onset seizures in adults. In April 2013, the Food and Drug Administration (FDA) issued a warning that RTG could cause changes in retinal pigmentation and discoloration of skin resulting in a blue appearance. As part of a larger preclinical effort to gain a mechanistic understanding as to the origins of retinal pigment changes associated with RTG, we conducted a long-term repeat dosing study in rats. Matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) was used to determine the distribution of RTG and its metabolites in the rat eye following 13 and 39 weeks of dosing. IMS revealed the presence of RTG, a previously characterized N-acetyl metabolite of RTG (NAMR), and several species structurally related through the dimerization of RTG and NAMR. These species were highly localized to the melanin containing layers of the uveal tract of the rat eye including the choroid, ciliary body and iris, suggesting that the formation of these dimers occurs from melanin bound RTG and NAMR. Furthermore, several of the RTG-related dimers have UV absorbance which give them a purple color in solution. We propose that the melanin binding of RTG and NAMR effectively concentrates the two compounds to enable mixed condensation reactions to occur when the binding provides the proper geometry and redox environment in the uveal tissues. High lateral resolution images illustrate that the blood-retinal barrier effectively restricts retinal access to RTG-related compounds. The spatial information provided by MALDI IMS was critical in contextualizing the homogenate concentrations of key RTG-related compounds and helped provide a basis for the mechanism of dimer formation.

1.

INTRODUCTION

Epilepsy is a disease associated with significant patient morbidity and mortality, with approximately one-third of patients being refractory to available medications. The majority of patients with epilepsy have partial seizures, accounting for about two thirds of the cases in epidemiologic studies in developed countries.1 Despite antiepileptic drug treatment, approximately one-third of patients will have persistent seizures and can be classified as medically intractable.1 Retigabine (international nonproprietary name; RTG in this article) or ezogabine (US adopted name) is approved as an adjunctive treatment for refractory partialonset seizures in adults. RTG is differentiated from other antiepileptic drugs by its unique mechanism of action. By activating voltage-gated potassium channels KV7.2-7.5 (KCNQ2/3) in brain neurons, RTG initiates hyperpolarization of the membrane potential resulting in decreased neuronal excitability under physiological conditions.2-6 In clinical trials, RTG demonstrated reasonable efficacy for adjunctive partial seizure control and appeared to be relatively safe and well tolerated. Efficacious doses range between 600 and 1200 mg per day with circulating Cmax concentrations generally less than 2500 ng/mL.7-11

In April 2013, the Food and Drug Administration (FDA) issued a warning that RTG could cause changes in retinal pigmentation and discoloration of skin resulting in a blue appearance (Figure 1).12-15 At the time of this warning, those affected had long RTG treatment intervals of over four years on average with a range of 0.8 to 7 years. Of the estimated 605 patients, 38 had developed skin discoloration (6.3%) and all but two had received RTG treatment for at least two years. Of the 89 patients that remained in ongoing studies, 36 had eye examinations that included a funduscopic and corrected visual acuity examination. Eleven of these 36 patients were found to have retinal pigmentary abnormalities; however, no visual acuity baseline was available for these patients. It was not known if the changes in retinal pigment could result in a loss of vision. Only one of the 11 patients with retinal pigment changes received a full diagnostic retinal evaluation; the findings were consistent with retinal dystrophy. Not all patients with retinal pigmentary abnormalities also developed blue-tinged skin. In October of that year, the FDA issued a “black box” warning on the RTG label due to the potential risk of changes in retinal pigmentation, potential vision loss, and skin discoloration.16 In 2014, the first article documenting the development of mucocutaneous dyspigmentation (abnormal discoloration of

ACS Paragon Plus Environment

Chemical Research in Toxicology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the skin and mucous membranes) in two patients who had been treated with a combination of antiepileptic drugs, including RTG, for several years was published.17 The main histopathological finding was similar to previously reported cases of dyspigmentation induced by psychotropic drugs and amiodarone: dermal cells heavily laden with coarse melanin granules. These granules were located mainly around blood vessels and adnexae and appeared mostly intracellular based on electron microscopy. The peculiar skin pigmentation observed in these patients was attributed to the Tyndall effect, where the dermal melanin appears blue, gray, or blue-gray due to the selective scatter of shorter wavelengths. Attempts to identify RTG or RTG-related metabolites in tissue extracts by 19F-NMR and mass spectrometry were not successful. The authors proposed the origins of skin discoloration were due either to RTG induced melanin synthesis or disruption of melanin degradation. One patient showed significant improvement of the dyspigmentation four months after discontinuing RTG. An additional case study of purple/blue discoloration of skin, nail beds, lips, and mucosa of the hard palate associated with RTG treatment was reported.18 In June of 2015, the FDA issued an additional safety communication that stated that the potential risks of vision loss due to pigment changes in the retina and of skin discoloration could be adequately managed by following current recommendations. The review of additional safety reports did not indicate that the retinal pigmentation changes observed in some patients affected vision. Furthermore, skin discoloration associated with the use of RTG appeared to be cosmetic and did not appear to be associated with more serious adverse effects. They concluded that modification of the Risk Evaluation and Mitigation Strategy was not needed at the time to ensure that the benefits of RTG outweigh the risks of retinal and skin pigment changes. Furthermore, GSK committed to a long-term observational study. 19 As a part of our investigational studies to gain a mechanistic understanding of the origins of retinal pigment changes associated with long-treatment intervals with RTG, we conducted a long-term repeat dosing study in rats. Matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) was used to determine the presence and distribution of RTG and its metabolites in the rat eye following 13 and 39 weeks of dosing. Herein, we report the findings of this investigation.

Page 2 of 16

All animal procedures were conducted in an American Association for the Accreditation of Laboratory Animal Care (AALAC)-accredited facility at GlaxoSmithKline (GSK) in accordance with GSK policies on the care, welfare and treatment of laboratory animals and they were reviewed and approved by GSK’s Institutional Animal Care and Use Committee (IACUC) as appropriate. The in-life phase of this study, including dosing and sample collection, was performed by Safety Assessment at GlaxoSmithKline, Research Triangle Park, NC, US. Briefly, RTG was administered to pigmented male Long Evans (HsdBlu:LE) rats for 91 days (Group 2) or 271 days (Group 3) by oral diet administration at a dose of 0 (vehicle; Group1) or nominal dose of 100 mg/kg/day. Rats were necropsied on days 91, or 272. Eye tissues were collected on days 91 or 272 for evaluation by MALDI IMS. Each of the tissues was wrapped in labelled aluminum foil and snap frozen in liquid nitrogen and stored at -80 °C until analysis. 2.3.

MALDI IMS

2.3.1.

Tissue Preparation

Eye tissues were embedded in pHPMA by the careful addition of cold polymer solution (150 mg/mL pHPMA in H2O) to the frozen eye in a cryomold cooled on crushed dry ice. Thin sections (6 m) of the embedded eye tissues were collected in a cryostat (CM3050S, Leica, Buffalo Grove, IL; Chamber temp. 20°C and Object temp. -18°C) and mounted onto ITO coated glass microscope slides. An optical image of each tissue section was generated using an Aperio ScanScope (Leica, Buffalo Grove, IL) digital slide scanner (20x magnification) prior to matrix application. Sections (6 m) serial to those collected for MALDI IMS were collected for hematoxylin and eosin (H&E) staining to correlate ion images with tissue histology. An optical image of each H&E stained serial section was generated using an Aperio ScanScope digital slide scanner (20x and 40x magnifications). 2.3.2.

Matrix Application

2.

MATERIALS AND METHODS

DHB matrix was applied to the tissues using a custom-built sublimation apparatus. The system was operated under vacuum (~200 mTorr) and a heating mantle was used to heat the chamber containing 50 mg of DHB to ~140° C until all matrix was sublimed (approximately 15 min). Following sublimation, the matrix coated slides were incubated for approximately 20 min in a chamber saturated with methanol.

2.1.

Materials and Reagents

2.3.3. Data Acquisition

GW582892X (Retigabine, RTG) (batch #: K282470 / Stated Purity: 99.84%), GSK2398706 (NAMR) (Batch #: U25399-1384 / Stated Purity: 99.8%), GSK3212509 (RTG-RTG) (batch #: EE717717 (N20797-86-2) / Stated Purity: 98.86%), and GSK3211985 (NAMR-NAMR) (batch #: EE721615 (EE657569120) / Stated Purity: 99.459%) were provided by GSK Product Development and Supply. 2,5-dihydroxybenzoic acid (DHB, purity 98%) was purchased from Sigma Aldrich and purified by recrystallization prior to use. Indium tin oxide (ITO) coated glass microscope slides were purchased from Bruker (Billerica, MA, USA). Superfrost® Plus glass slides and Richard-Allan Scientific™ Signature Series Hematoxylin 2, Eosin Y, Clarifier 1, and Bluing Reagent were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Tissue embedding polymer, poly(N-(2-hydroxypropyl)methacrylamide) (pHPMA), was purchased from Sigma Aldrich. 2.2.

In-life Phase

MALDI IMS was performed on a 7T Solarix Fourier transform – ion cyclotron resonance (FT-ICR) mass spectrometer (Bruker Daltonics, Billerica, MA). Positive ion mass spectra were acquired in full scan mode (m/z 200-1000) with pixel dimensions ranging from 5-25 m. All images were acquired using a 256K data acquisition size (0.2447 s transient) giving an estimated resolution of 33,000 at m/z 400. The 25 m pixel dimension images were acquired with minimum laser focus setting, 50 laser shots at 200 Hz, and data reduction of 97.5%. The 5 and 10 m pixel dimension images were acquired with minimum laser focus, 25 laser shots at 200 Hz, and data reduction of 97.5%. MALDI MS spectra of the RTG, NAMR, RTGRTG, and NAMR-NAMR standards are included in Figure S1. 2.3.4. Ion Image Visualization All ion images were generated using FlexImaging v5.0 (Bruker Daltonics) from the reduced raw data. All ion images are displayed as the maximum intensity in range with a mass


Page 3 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


tolerance of ± 0.0005 Da centered on the measured m/z value. The intensity threshold values used to generate ion distributions are shown within each respective ion image. The optical, H&E, and ion images were co-registered and arranged into figures using Photoshop CC (Adobe, San Jose, CA) and overlaid ion images are displayed with 80% opacity. 2.3.5. UV Spectra Calculations All calculations were completed using Spartan’16 (Wavefunction, Irvine CA) software. Chemical structures were built using the Spartan graphical interface and the equilibrium gas phase geometry was determined using the density functional B3LYP model with the 6-31G* basis set. The UV/Vis spectra were calculated by running a single point TD-DFT calculation after the main wave function was calculated. The absorption energies were the difference between the HF ground state and CIS excited state energies. All UV spectra were calculated with the key word “UVSTATES” equal to 10. 3.

RESULTS

3.1.

MALDI IMS Analysis Eye Tissues

MALDI IMS analysis of eye tissue sections from pigmented male Long Evans rats administered RTG revealed the presence of RTG (Figure 2A), an N-acetyl metabolite of RTG (NAMR) (Figure 2B), and several species corresponding to the dimerization of RTG and NAMR. These included: i) three phenazinium dimers (Figure 2C) corresponding to RTG-RTG, NAMR-NAMR, and RTG-NAMR, ii) the corresponding phenazine species, RTG-RTG*, NAMR-NAMR*, and RTG-NAMR* (Figure 2D) which are formed by the loss of the fluorobenzyl group through quaternary amine hydrolysis and iii) lower intensity ions associated with the amidine and imidazolidin-2one (benzimidazoles) cyclized dimer species (Figure 2E). The distribution of several of these species detected by MALDI IMS (25 µm pixel dimensions) analysis in ocular tissue from a rat selected from Group 2 (RTG 100 mg/kg/day; 91 days) is shown in Figure 3. At this pixel dimension, these species appeared to nominally be localized to the melanin containing layers of the uveal tract including the choroid, ciliary body and iris (Figure 4). The predominant ion detected for the dimers from eye tissue was the [M+K]+ ion of the neutral phenazine species, which results primarily from the hydrolysis of the phenazinium quaternary amine and loss of the fluorobenzyl group. The [M]+ of the intact phenazinium dimers were also detected but with lower signal intensity. The same spatial distribution was observed for these species in an eye tissue section collected from a different rat from the same dosing group (Group 2, 100 mg/kg/day; 91 days) as shown in Figure 5. No ions associated with RTG-related compounds were observed in control tissues (Group 1). Further investigation into the localized distribution of RTGrelated material was conducted through several high spatial resolution imaging experiments. For example, Figure 6 displays the 10 µm pixel dimension ion images for the three major dimer species, detected as the phenazine species [M+K]+, in ocular tissue from an animal from each group: Group 1 (vehicle; 272 days), Group 2 (100 mg/kg/day; 91 days), and Group 3 (100 mg/kg/day; 272 days). These images further show the highly-localized distribution of the dimer species in the eyes of animals administered RTG. The same species and signal intensity distributions were observed for the RTGrelated material between animals from Group 2 (91 days) and Group 3 (272 days).

The results from an imaging experiment conducted with 5 m pixel dimensions from a Group 2 rat (100 mg/kg/day; 91 days) are displayed in Figure 7. In the ion image for the phenazine RTG-NAMR* [M+K]+, the localization appears to be primarily to the choroid layer with little or no ion intensity detected in the RPE. The large blood vessels within the choroid do not appear to contain any RTG-related material. In Figure 8, the ion image for the phenazine RTG-NAMR* [M+K]+ (green) is overlaid with the ion image for bis-retinoid N-retinyl-N-retinylidene ethanolamine (A2E) (red). A2E is a by-product of the visual cycle and is known to accumulate in the RPE and Bruch’s membrane. 20 The distinct distributions observed for these species is further evidence that the RTG-related material is localized specifically to the choroid layer and excludes the RPE. 4.

DISCUSSION

This investigation was a part of a much larger preclinical effort to understand the mechanism and safety risk of mucosal tissue dyspigmentation and retinal pigmental changes in patients after chronic administration of RTG. The purple hue (absorption 500-560 nm) associated with the dyspigmentation suggested that it was not due directly to RTG (UVmax 304 nm) or any of the previously characterized metabolites (Figure S2).21 Compounds that absorb in the visible spectrum with large molar absorption coefficients can create the observation of color with little compound present, making structural characterization and isolation an analytical challenge. A key jumping-off point in this investigation was access to two previously characterized RTG-related phenazinium dimers, which appear purple in solution: RTG-RTG (UVmax 551 nm) and NAMR-NAMR (UVmax 549 nm) (Figure 2). A common element between reports on patients with dyspigmentation in mucosal tissues and retinal pigmental changes was an association with melanin containing tissue.17, 18 Additionally, in several nonclinical investigations (data not reported) using both pigmented (Long Evans) and albino (Wistar Han) rats, the dimer species were only detected in the eye tissues from the pigmented rats by LC-MS/MS; however, the plasma levels for RTG and NAMR were similar between the pigmented and albino rats. The dimer species were not detected in the plasma from either the pigmented or albino rats. Melanin is a naturally occurring indole polymer pigment that provides color to the hair, skin and eyes and has a high binding affinity for a wide range of systemically administrated nonocular drugs. Chemically, there are two structural classes of melanin: eumelanin (brown to black in color) and pheomelanin (yellow to red in color). Both eumelanin and pheomelanin polymers originate from the oxidation of tyrosine by tyrosinase enzyme. The pathway produces related indole quinones and hydroquinones which assemble to form eumelanin (Figure S3). The detailed macromolecular structure of melanin is not known due to the heterogeneity associated with the polymerization of the various monomers. However, in general eumelanin is the dominant form in ocular tissue and hence considered the most relevant structure type with regard to drug binding.22 There are three known protective biological functions of ocular melanin: i) attenuate tissue damage by absorbing UV light, ii) scavenge reactive free radicals (both oxidizing and reducing) and iii) chelate reactive redox metal ions. The absorption of light is the most important of these functions in the eye. Melanin reduces the amount of visible and UV light entering the eye and protects against harmful effects including DNA damage. Melanin is believed to protect ocular tissues from



oxidative stress by acting as an antioxidant and also has the capacity to chelate transition metal ions and prevent them from generating reactive oxygen species.22, 23 The eye is the most densely pigmented organ in the body. The pigmented epithelium and the uveal tract forms a continuous band which is located between the sclera and the nonpigmented neural retina. The uveal tract, which is the highly vascularized middle layer of the eye, consists of three continuous areas: the choroid layer, the iris and ciliary body (Figure 4). The choroid is the primary structure of the eye responsible for supplying blood to the external layers of the retina. The innermost melanin containing tissue in the eye contains the retinal pigment epithelium (RPE) and in the anterior region, the pigmented ciliary epithelium and the iris pigmented epithelium. In the human eye, it is estimated that combined, the choroid and RPE tissues constitute >60% of the ocular melanin.24 Systemic circulating xenobiotics can reach the uveal tract via branches of the ophthalmic artery including the short and long posterior ciliary arteries. The central retinal artery which also branches from the ophthalmic artery provides the interior surface of the retina with arterial blood.25 Endothelia of the retinal blood vessels contain tight junctions to form an inner blood-retinal barrier, similar to those of the blood-brain barrier, and restrict access to highly permeable low molecular weight compounds or substrates of uptake transporters. Retinal efflux transporters further restrict xenobiotic access. In addition, the RPE cells are connected by tight junctions forming an outer blood-retinal barrier between the choroid layer and the retina, which further restricts the movement of xenobiotics from the choroid layer into the retina.26 The ocular tissues analyzed by MALDI IMS from rats administered RTG for 91 days (Group 2) and 272 days (Group 3) revealed a well-defined and conserved distribution for all RTG-related material to the melanin containing uveal tract, further suggesting some possible role for melanin in the formation of the dimers. Melanin affinity for RTG and NAMR, as well as the derived dimers, is consistent with the most commonly reported structural descriptors of molecules with high melanin binding in the literature: basicity, hydrophobicity, and π−π stacking motifs conducive to charge–transfer.27 The structural elements of the three detected dimers are very similar to eumelanin. While many drugs bind melanin, the mechanism is not well understood. No correlation between melanin binding and the plasma protein binding has been demonstrated suggesting that these binding events are very different.28 While drug binding to melanin can be associated with enhanced retention in the pigmented tissues and prolonged drug exposure, the consequences of binding to ocular melanin could be protective or result in adverse effects.23 Many drugs with high melanin affinity do not show ocular toxicity.29 The detection of the three phenazinium dimers (RTG-RTG, NAMR-NAMR; RTG-NAMR) in the melanin containing ocular tissue, their absence in plasma or other non-melanin containing ocular tissues, and the high melanin association for both RTG and NAMR tissues suggests that the formation of these dimers occurs from melanin bound RTG and NAMR. In this hypothesis, the melanin binding of RTG and NAMR effectively concentrates the two compounds to enable mixed condensation reactions to occur when the binding provides the proper geometry in the oxidative environment of the uveal tract.30 The observation that the dimers slowly accumulate in the eye over the course of several weeks in rats and that

Page 4 of 16

patients affected by dyspigmentation or changes in retinal pigmentation had long RTG treatment intervals of over four years on average are consistent with dimerization as the rate limiting step, occurring infrequently relative to RTG and NAMR binding. Furthermore, the lack of detectable dimer levels in plasma suggest that melanin affinity for the dimers may be higher than the precursors of RTG and NAMR. A proposed mechanism for formation of the phenazinium dimers is outlined in Scheme 1. Dimer formation can be rationalized through a concerted or step-wise conjugate addition pathway to form the two new C-N bonds of the hydrophenazine ring following initial oxidation of RTG or NAMR to the diimide quinine. The final step would involve oxidation to aromatize the hydrophenazine ring to the phenazinium cationic species. The IMS and LC-MS/MS experiments also established the discrete presence of the corresponding phenazine dimers as downstream hydrolysis products. Calculated UV spectra for the phenazine dimers suggest that their presence would contribute to the discoloration of the tissue because they contain a similar absorbance UVmax to the corresponding phenazinium compounds (calculated gas phase NAMR-NAMR UVmax ~500 nm; NAMR-NAMR* UVmax ~495 nm). The intramolecular cyclization to form the two benzimidazole compounds has been shown to occur as a part of the biotransformation of RTG but are not necessarily enzymatically driven. However, given the low IMS intensity levels of the benzimidazole dimers it may be that this process occurs slowly after dimerization. It is also possible that the primary nitrogen of RTG, NAMR, and or the corresponding acetamide or carbamate groups are less available for further reaction because of melanin binding. The oxidative conditions necessary to promote dimerization are consistent with the processes associated with ocular tissue and melanin.26 LC-MS/MS quantification was conducted as part of a separate study (unpublished data) to estimate the levels of RTG, NAMR, RTG-RTG and NAMR-NAMR in plasma and various tissues. A complete quantitative assessment was not possible as no standards for the phenazine dimers and the RTG-NAMR dimer were available; however, some useful trends for the monomer and dimer binding as well as their differentiation between the plasma and uveal tract tissue compartments can be made. In ocular tissue homogenates, RTG and NAMR concentrations were similar on Day 91 (592 and 684 pmol/g, respectively) and Day 272 (893 and 816 pmol/g, respectively). For the RTG-RTG and NAMR-NAMR dimers, on Day 91 ocular tissue homogenate concentrations were 110 and 183 pmol/g and increased to 229 and 390 pmol/g, respectively, at Day 272. Neither dimer species were observed in plasma in any of the preclinical studies completed to date reflecting their high affinity for melanin. Furthermore, in studies that included off-dose periods of up to 13-weeks following the primary dosing, only a modest reduction (10-20%) in the dimer concentrations were observed in the eye tissue over the course of the off-dose period. Our efforts to achieve higher pixel resolution were driven by the desire to distinguish between the binding of RTG-related compounds between the melanin containing choroid layer of the uveal tract and the monolayer RPE of the retina. Melanin binding within the retina, the region of the eye containing the photoreceptors, could have greater potential risk for adverse events. Thus, the high resolution IMS experiments (Figure 7; 8) demonstrate that no detectable levels of RTG related material including the dimers were able to penetrate the blood-


Page 5 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


retinal barrier associated with the RPE. It is not clear how our findings in the rat model study relate to the FDA warning of changes in retinal pigmentation associated with long term use of RTG. In conclusion, we demonstrated that RTG, NAMR and three related phenazinium dimers (RTG-RTG, NAMR-NAMR; RTGNAMR) are associated with ocular melanin binding in the uveal tract following chronic dosing of RTG (91 and 272 days of dosing) in the rat model. In addition, related phenazine and benzimidazole compounds were also detected from the uveal tract. Based upon their purple appearance and UV spectra, we propose that these dimers could also be responsible for the dyspigmentation (purple-grey appearance) in melanin containing skin tissues of patients with long term RTG treatment. Low levels of the RTG-RTG dimer were observed in the skin by LC-MS/MS (Days 91 and 272). However, MALDI IMS analysis was not conducted on the skin tissues. High affinity melanin binding was observed for RTG, NAMR and their associated dimers in the in vitro melanin binding assays. It was important to observe that RTG and RTG-related species including the “purple” dimers were not observed in the RPE, the melanin containing region of the retina. This suggests that the blood-retinal barrier effectively restricts RTG-related compounds from entering the retinal compartment, the center for photoreceptors. In accordance with IHC guidelines31, during development an in vitro phototoxicity study in 3T3 cells

(data not presented) was negative for RTG, suggesting no phototoxic potential. The observed dyspigmentation in skin and pigment changes in the retina of some patients with long treatment intervals can be classified as adverse events but not phototoxicity. No vision impairment or dermal toxicity has been linked with the dyspigmentation. Furthermore, no ocular toxicity was reported for either treatment dose groups in this study (data not presented) or in any of the non-clinical studies conducted at GSK. The observation of dyspigmentation due to drug treatment, while known, is a rare event which has been restricted to a few structural classes of drugs. What has not been previously established is the molecular species and mechanistic origins for the tissue dyspigmentation. This knowledge provides a basis for risk assessment and mitigation. MALDI IMS analysis was a critical element in establishing the ocular tissue disposition of RTG in the rat model. The specific association with the uveal melanin helped contextualize the homogenate concentrations of key RTG-related compounds but also provided a basis for the mechanism of dimer formation. Furthermore, these data demonstrated the blood-retinal barrier excluded RTG-related material to the photoreceptors of the retina. In June 2017, GSK discontinued RTG permanently due to very limited usage and a continued decline in new patient initiation.



ASSOCIATED CONTENT Supporting Information

(9)

The Supporting Information is available free of charge on the ACS Publications website.

(10)

Standard Analysis and Reaction Mechanisms (PDF)

AUTHOR INFORMATION

(11)

Corresponding Author *Stephen Castellino, e-mail: [email protected], phone +1 267-472-1526 (12)

ACKNOWLEDGEMENTS The authors thank the following individuals for their contributions to these investigative efforts: Judith Prescott, Christopher Evans, Hermes Licea-Perez, Dave Melich, Sharon Boram, Dan Reynolds, Scott Sides, Chris Merrill, Dana Knecht, Kendra Hightower and Greg Waitt.

ABBREVIATIONS

(13) (14) (15)

RTG, retigabine; FDA, Food and Drug Administration; MALDI, matrix-assisted laser desorption/ionization; IMS, imaging mass spectrometry; NAMR, N-acetyl metabolite of retigabine; FT-ICR Fourier transform – ion cyclotron resonance

(16) (17)

REFERENCES (1)

(2) (3)

(4)

(5)

(6)

(7)

(8)

Ciliberto, M. A., Weisenberg, J. L., and Wong, M. (2012) Clinical utility, safety, and tolerability of ezogabine (retigabine) in the treatment of epilepsy. Drug Healthc. Patient Saf. 4, 81-86. Rundfeldt, C. (1997) The new anticonvulsant retigabine (D-23129) acts as an opener of K+ channels in neuronal cells. Eur. J. Pharmacol. 336, 243-249. Main, M. J., Cryan, J. E., Dupere, J. R., Cox, B., Clare, J. J., and Burbidge, S. A. (2000) Modulation of KCNQ2/3 potassium channels by the novel anticonvulsant retigabine. Mol. Pharmacol. 58, 253262. Tatulian, L., Delmas, P., Abogadie, F. C., and Brown, D. A. (2001) Activation of expressed KCNQ potassium currents and native neuronal M-type potassium currents by the anti-convulsant drug retigabine. J. Neurosci. 21, 5535-5545. Wuttke, T. V., Seebohm, G., Bail, S., Maljevic, S., and Lerche, H. (2005) The new anticonvulsant retigabine favors voltage-dependent opening of the Kv7.2 (KCNQ2) channel by binding to its activation gate. Mol. Pharmacol. 67, 1009-1017. Ihara, Y., Tomonoh, Y., Deshimaru, M., Zhang, B., Uchida, T., Ishii, A., and Hirose, S. (2016) Retigabine, a Kv7.2/Kv7.3-Channel Opener, Attenuates DrugInduced Seizures in Knock-In Mice Harboring Kcnq2 Mutations. PLoS One 11, e0150095. Blackburn-Munro, G., Dalby-Brown, W., Mirza, N. R., Mikkelsen, J. D., and Blackburn-Munro, R. E. (2005) Retigabine: chemical synthesis to clinical application. CNS Drug Rev 11, 1-20. Weisenberg, J. L., and Wong, M. (2011) Profile of ezogabine (retigabine) and its potential as an adjunctive

(18)

(19)

(20)

(21)

(22)

(23)

(24)

Page 6 of 16

treatment for patients with partial-onset seizures. Neuropsychiatr. Dis. Treat. 7, 409-414. Amabile, C. M., and Vasudevan, A. (2013) Ezogabine: a novel antiepileptic for adjunctive treatment of partialonset seizures. Pharmacotherapy 33, 187-194. Brodie, M. J., Lerche, H., Gil-Nagel, A., Elger, C., Hall, S., Shin, P., Nohria, V., Mansbach, H., and Group, R. S. (2010) Efficacy and safety of adjunctive ezogabine (retigabine) in refractory partial epilepsy. Neurology 75, 1817-1824. Porter, R. J., Burdette, D. E., Gil-Nagel, A., Hall, S. T., White, R., Shaikh, S., and DeRossett, S. E. (2012) Retigabine as adjunctive therapy in adults with partialonset seizures: integrated analysis of three pivotal controlled trials. Epilepsy Res. 101, 103-112. (2013) FDA Drug Safety Communication: Anti-seizure drug Potiga (ezogabine) linked to retinal abnormalities and blue skin discoloration (https://wayback.archiveit.org/7993/20170406043945/https://www.fda.gov/Drug s/DrugSafety/ucm349538.htm) (2013) Ezogabine (Potiga) toxicity. Med. Lett. Drugs Ther. 55, 96. Aschenbrenner, D. S. (2013) Retinal Abnormalities and Blue Skin from Antiseizure Drug. Am. J. Nurs. 113, 1. (2013) Retigabine: blue skin discoloration and retinal pigment abnormalities. Prescrire Int. 22, 269. FDA. (2013) POTIGA (ezogabine) FDA Approved Labeling dated 09/06/2013. Garin Shkolnik, T., Feuerman, H., Didkovsky, E., Kaplan, I., Bergman, R., Pavlovsky, L., and Hodak, E. (2014) Blue-gray mucocutaneous discoloration: a new adverse effect of ezogabine. JAMA Dermatol. 150, 984989. Beacher, N. G., Brodie, M. J., and Goodall, C. (2015) A case report: retigabine induced oral mucosal dyspigmentation of the hard palate. BMC Oral Health 15, 122. (2015) FDA Determines 2013 labeling adequate to manage risk of retinal abnormalities, potential vision loss, and skin discoloration with anti-seizure drug Potiga(ezogabine); requires additional study. FDA Safety Communication. Grey, A. C., Crouch, R. K., Koutalos, Y., Schey, K. L., and Ablonczy, Z. (2011) Spatial localization of A2E in the retinal pigment epithelium. Invest. Ophthalmol. Vis. Sci. 52, 3926-3933. Hempel, R., Schupke, H., McNeilly, P. J., Heinecke, K., Kronbach, C., Grunwald, C., Zimmermann, G., Griesinger, C., Engel, J., and Kronbach, T. (1999) Metabolism of retigabine (D-23129), a novel anticonvulsant. Drug Metab. Dispos. 27, 613-622. Karlsson, O., and Lindquist, N. G. (2016) Melanin and neuromelanin binding of drugs and chemicals: toxicological implications. Arch. Toxicol. 90, 18831891. Rimpela, A. K., Reinisalo, M., Hellinen, L., Grazhdankin, E., Kidron, H., Urtti, A., and Del Amo, E. M. (2017) Implications of melanin binding in ocular drug delivery. Adv. Drug Del. Rev. Menon, I. A., Wakeham, D. C., Persad, S. D., Avaria, M., Trope, G. E., and Basu, P. K. (1992) Quantitative determination of the melanin contents in ocular tissues from human blue and brown eyes. J. Ocul. Pharmacol. 8, 35-42.


Page 7 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(25) (26)

(27)

(28)

Chemical Research in Toxicology Koneru, P. B., Lien, E. J., and Koda, R. T. (1986) Oculotoxicities of systemically administered drugs. J. Ocul. Pharmacol. 2, 385-404. Nakano, M., Lockhart, C. M., Kelly, E. J., and Rettie, A. E. (2014) Ocular cytochrome P450s and transporters: roles in disease and endobiotic and xenobiotic disposition. Drug Metab. Rev. 46, 247-260. Reilly, J., Williams, S. L., Forster, C. J., Kansara, V., End, P., and Serrano-Wu, M. H. (2015) HighThroughput Melanin-Binding Affinity and In Silico Methods to Aid in the Prediction of Drug Exposure in Ocular Tissue. J. Pharm. Sci. 104, 3997-4001. Pelkonen, L., Tengvall-Unadike, U., Ruponen, M., Kidron, H., Del Amo, E. M., Reinisalo, M., and Urtti, A. (2017) Melanin binding study of clinical drugs with

(29)

(30)

(31)

cassette dosing and rapid equilibrium dialysis inserts. Eur. J. Pharm. Sci. 109, 162-168. Leblanc, B., Jezequel, S., Davies, T., Hanton, G., and Taradach, C. (1998) Binding of drugs to eye melanin is not predictive of ocular toxicity. Regul. Toxicol. Pharmacol. 28, 124-132. Vriezema, D. M., Comellas Aragones, M., Elemans, J. A., Cornelissen, J. J., Rowan, A. E., and Nolte, R. J. (2005) Self-assembled nanoreactors. Chem. Rev. 105, 1445-1489. International Conference on Harmonization (IHC) of technical requirements for registration of pharmaceuticals for human use: photosafety evaluation of pharmaceuticals (2013) https://www.ich.org/products/guidelines/safety/article/s afety-guidelines.html

For TOC Only



Page 8 of 16

Figure 1 Examples of blue skin discoloration associated with RTG (Photos from FDA Drug Safety Communication: Anti-seizure drug Potiga (ezogabine) linked to retinal abnormalities and blue skin discoloration (4/2013))


8

Page 9 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2 Structures of A) RTG B) NAMR C) phenazium dimers D) phenazine dimers E) benzimidazole dimers A

B O

O

O

HN

HN H2 N

H2 N

NH

NH F

F

RTG C16H18FN3O2

C

NAMR C15H16FN3O

F

H2 N

N

HN O

+

NH

HN

F

F H2 N

N

HN

N

N O

F

F

F

O

+

RTG-RTG C32H31F2N6O4+

H2 N

O

O

N

HN HN

O

O

NH

+

NH

N HN

O

NAMR-NAMR C30H27F2N6O2+

O

RTG-NAMR C31H29F2N6O3+

D F H2 N

N

HN

N

O

O

HN

F

F

NH

H2 N

N

HN

O

O

O

N

HN

N HN

O

O

NH

H2 N

NH

N HN

O

O

RTG-RTG*

NAMR-NAMR*

RTG-NAMR*

C25H25FN6O4

C23H21FN6O2

C24H23FN6O3

E F N

N

N H

N

F

NH

H N

N

N H

N

NH

O HN

Amidine

HN

Imidazolidin-2-one


9


Page 10 of 16

Figure 3 MALDI IMS analysis of whole rat eye tissue section at 25 µm pixel dimensions from a Group 2 animal (91 days) A) Prematrix optical scan B) Serial H&E stained section C) NAMR [M+K]+ (m/z 312.0909) D) RTG [M+K]+ (m/z 342.1015) E) NAMRNAMR*[M+K]+ (m/z 471.1342) F) RTG-RTG* [M+K]+ (m/z 531.1553) G) RTG-NAMR* [M+K]+ (m/z 501.14473) H) RTG-NAMR [M]+ (m/z 571.22637). Ion images are all displayed using a relative intensity scale of 5-60%.


10

Page 11 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4 Annotated rat eye H&E stained tissue section (BV – blood vessel RPE – retinal pigment epithelium)


11


Page 12 of 16

Figure 5 MALDI IMS analysis of whole rat eye tissue section at 25 m pixel dimensions from animal 43 (91 days) A) Pre-matrix optical scan B) Serial H&E stained section C) NAMR [M+K]+ (m/z 312.0909) D) RTG [M+K]+ (m/z 342.1015) E) NAMRNAMR*[M+K]+ (m/z 471.1342) F) RTG–RTG* [M+K]+ (m/z 531.1553) G) RTG-NAMR* [M+K]+ (m/z 501.14473) H) RTG-NAMR [M]+ (m/z 571.22637). Ion images are all displayed using a relative intensity scale of 5-60%.


12

Page 13 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6 MALDI IMS analysis of a posterior region of rat eye tissue sections at 10 m pixel dimensions from animals 23 (91 days/control), 41 (91 days/100 mg/kg/day), and 55 (272 days/100 mg/kg/day) A, F, K) Pre-matrix optical scan and serial H&E stained sections B, G, L) 5x magnification of analysis region on optical and H&E scans C, H, M) RTG–RTG* [M+K]+ (m/z 531.1553) D, I N) NAMR-NAMR*[M+K]+ (m/z 471.1342) E, J, O) RTG-NAMR* [M+K]+ (m/z 501.14473). Ion images are displayed using a relative intensity scale of 5-25%.


13


Page 14 of 16

Figure 7 MALDI IMS analysis of rat eye tissue section at 5 m pixel dimensions from animal 41 (91 days) A) Pre-matrix optical scan B) Ion image for RTG-NAMR* [M+K]+ (m/z 501.14473) overlaid with the pre-matrix optical scan C) Ion image for RTG-NAMR* [M+K]+ (m/z 501.14473). Ion image is displayed using a relative intensity scale of 5-25%


14

Page 15 of 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8 MALDI IMS analysis of rat eye tissue section at 10 m pixel dimensions from animal 41 (91 days) A) serial H&E stained section B) Pre-matrix optical scan C) Ion image for RTG-NAMR* [M+K]+ (m/z 501.14473) in green and A2E [M]+ (m/z 592.45129) in red overlaid with the pre-matrix optical scan D) Ion image for RTG-NAMR* [M+K]+ (m/z 501.14473) in green and A2E [M]+ (m/z 592.45129) in red. Ion images are displayed using a relative intensity scale of 5-25%.


15


Page 16 of 16

Scheme 1 Proposed mechanism for the formation of phenazinium dimers

hydrophenazine

phenazine

phenazinium

F F

F

F

2

H2 N R N H

NH

[O] melanin

R1 = COCH2CH3 R2 = COCH3

H2 N N R

N

F N HN NH R

F F

H2 N

N

HN R

N H

NH

NH R

[O]

F

+

H2 N

N

HN R

N

NH

H2 N

N

HN R

N

NH

NH R

+

NH R

F

HO


16

An Investigation into Retigabine (Ezogabine ... - ACS Publications

Recommend Documents