The Kv1.3 channel is a recognized target for pharmaceutical development to treat autoimmune diseases and organ rejection. ShK-186, a specific peptide inhibitor of Kv1.3, has shown promise in animal models of multiple sclerosis and rheumatoid arthritis. Here, we describe the pharmacokinetic-pharmacodynamic relationship for ShK-186 in rats and monkeys. The pharmacokinetic profile of ShK-186 was evaluated with a validated high-performance liquid chromatography-tandem mass spectrometry method to measure the peptide's concentration in plasma. These results were compared with single-photon emission computed tomography/computed tomography data collected with an 111In-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid-conjugate of ShK-186 to assess whole-blood pharmacokinetic parameters as well as the peptide's absorption, distribution, and excretion. Analysis of these data support a model wherein ShK-186 is absorbed slowly from the injection site, resulting in blood concentrations above the Kv1.3 channel-blocking IC50 value for up to 7 days in monkeys. Pharmacodynamic studies on human peripheral blood mononuclear cells showed that brief exposure to ShK-186 resulted in sustained suppression of cytokine responses and may contribute to prolonged drug effects. In delayed-type hypersensitivity, chronic relapsing-remitting experimental autoimmune encephalomyelitis, and pristane-induced arthritis rat models, a single dose of ShK-186 every 2 to 5 days was as effective as daily administration. ShK-186's slow distribution from the injection site and its long residence time on the Kv1.3 channel contribute to the prolonged therapeutic effect of ShK-186 in animal models of autoimmune disease.
The Kv1.3 channel has been an active target of pharmaceutical development for more than 15 years (Chi et al., 2012). The interest in this channel derives from its important function in activated effector-memory T (TEM) cells, which are major mediators of autoimmune disease (Wulff et al., 2003; Beeton et al., 2006). Engagement of the T cell receptor by antigen-presenting cells results in an influx of calcium into the cytoplasm, initially from the endoplasmic reticulum but subsequently from the extracellular space via the Ca2+ release-activated Ca2+ channel (Cahalan and Chandy, 2009). The opening of voltage-gated Kv1.3 and calcium-activated KCa3.1 potassium channels in the T cell membrane and the resulting efflux of potassium ions promotes calcium entry and sustains intracellular calcium at concentrations necessary for T cell activation (Cahalan and Chandy, 2009). Resting T cells express a mixture of both K+ channels. However, upon activation, naive and central-memory T cells increase expression of the KCa3.1 channel, whereas TEM cells up-regulate Kv1.3 channel expression (Wulff et al., 2003). In the latter cell population, the degree of Kv1.3 expression is a measure of cell activation, and the channel is required for the maintenance of the TEM cell phenotype (Hu et al., 2007, 2012).
We and others have shown previously that Kv1.3HIGH CCR7−-activated TEM cells are present at sites of inflammation in autoimmune disease, and auto-reactive T cells from subjects with multiple sclerosis, type 1 diabetes, and rheumatoid arthritis express the Kv1.3HIGH phenotype of activated TEM cells (Rus et al., 2005; Beeton et al., 2006). In addition, specific Kv1.3 inhibitors have been found to be effective in numerous animal models of inflammation including chronic relapse-remitting experimental autoimmune encephalomyelitis (CR-EAE) and adoptive EAE (Beeton et al., 2005, 2006), pristane-induced arthritis (PIA) (Beeton et al., 2006), the delayed-type hypersensitivity (DTH) reaction (Koo et al., 1997; Beeton et al., 2005; Matheu et al., 2008), allergic contact dermatitis (Azam et al., 2007), allogeneic kidney transplant (Grgic et al., 2009), spontaneous autoimmune diabetes (Beeton et al., 2006), vascular neointima hyperplasia (Cheong et al., 2011), antiglomerular basement membrane glomerulonephritis (Hyodo et al., 2010), and psoriasis (Gilhar et al., 2011). For these reasons, numerous groups have focused on developing specific and potent inhibitors of the Kv1.3 channel for the treatment of inflammation and autoimmune disease (Cahalan and Chandy 2009, Rangaraju et al., 2009). Much of this effort has focused on developing small-molecule inhibitors of Kv1.3, but identifying compounds that are adequately specific has proved challenging, and to date, no drug specifically targeting Kv1.3 has entered clinical trials.
Our group is developing ShK-186, a 37-amino acid selective peptide inhibitor of Kv1.3, as a therapeutic for autoimmune diseases. To date, the Food and Drug Administration has approved more than 55 peptide drugs in approximately 33 mechanistic classes, making peptides one of the most active areas of biologics drug research. However, the peptide field has suffered from a perception that frequent dosing is required for sustained pharmacodynamic (PD) activity and patients have a poor acceptance of parenteral therapies. Therefore, careful attention to dose frequency and dose presentation are important to the development of a commercially viable peptide drug. Here, we report on the optimization of ShK-186 dose and dose frequency in animal models of autoimmune disease as part of our recently completed nonclinical program for the peptide.
Previous studies from our group have characterized the level of Kv1.3 channel-blocking activity in the serum of Lewis rats after single subcutaneous injections of three related analogs, ShK(L5), ShK-186, and ShK-192 (Beeton et al., 2005; Matheu et al., 2009; Pennington et al., 2009). Peak serum drug activity occurred 30 min after injection and returned to baseline by 7 h postdose, but ∼200 pM concentration of functionally active peptide was detected in the blood 24 and 48 h after injection, and approximately 100 pM was detectable at 72 h (Beeton et al., 2005; Matheu et al., 2009; Pennington et al., 2009). However, previous animal studies that investigated ShK peptides were performed by using a once-daily or more than once-daily dosing frequency (Beeton et al., 2001, 2005, 2006; Matheu et al., 2008; Pennington et al., 2009). Here, we demonstrate that ShK-186 has a long-lasting therapeutic effect in three different rat models of autoimmune disease because of its slow release from the site of subcutaneous injection and its tight binding to and slow release from the Kv1.3 channel on T cells.
Materials and Methods
DA and Lewis rats (8–10 weeks old) were purchased from Harlan (Indianapolis, IN) and housed under pathogen-free conditions with food and water ad libitum. DTH trials were approved by the Baylor College of Medicine and Infectious Disease Research Institute (IDRI; Seattle, WA) institutional animal use and care committees (IACUCs). EAE trials were approved by the IACUCs of the University of California at Irvine and the IDRI. PIA trials were approved by the Baylor College of Medicine IACUC.
Sprague-Dawley (Crl:CDSD) rats (6–9 weeks old) were purchased from Charles River Laboratories, Inc. (Wilmington, MA) and housed in a temperature (64°C-79°C)- and humidity (30–70%)-controlled facility. Food and water were ad libitum. PK studies in SD rats were approved by either the MPI Research or IDRI IACUCs.
Non-naive cynomolgus (Macaca fascicularis) and squirrel monkeys (Saimiri boliviensis) were between 2 and 5 years of age and transferred from the MPI Research stock colony to the study site (Mattawan, MI). All cynomolgus monkeys were of Chinese origin. The squirrel monkey was of Bolivian origin and obtained originally from the University of Texas MD Anderson Cancer Center (Houston, TX). Animals were housed individually in stainless-steel cages in an environmentally controlled room. The monkeys were provided environmental enrichment; fluorescent lighting was provided 12 h/day. Temperature was maintained between 64 and 84°C; humidity was 30 to 70%. Animals were provided Certified Primate Diet (PMI Nutrition International, Inc., St. Louis, MO) twice daily. Primatreats and other enrichment foods were provided on a regular basis. Water was available ad libitum. Primate studies were approved by the MPI Research IACUC.
Kv1.3 Peptide Inhibitors.
ShK-186 and ShK-198 (derivatives of ShK; accession no. P29187) were manufactured by using an fluorenylmethyloxycarbonyl-tertiary butyl solid-phase strategy on an amide resin. All of the coupling steps were mediated with 6-Cl-N-hydroxybenzotriazole in the presence of diisopropyl carbodiimide. Fluorenylmethyloxycarbonyl removal was facilitated with 20% piperidine in dimethylformamide containing 0.1 M N-hydroxybenzotriazole to buffer the piperidine and minimize potential racemization at the six Cys residues. After assembly, the peptide was cleaved from the resin and simultaneously deprotected by using a TFA cleavage cocktail (reagent K) containing aromatic cationic scavengers for 2 h at room temperature. The crude peptide was filtered from the spent resin and subsequently isolated by precipitation into ice-cold diethyl ether. The crude peptide was dissolved in 50% acetic acid and subsequently diluted into 3 liters of H2O. The pH of this peptide solution was adjusted to 8.0 with NH4OH and allowed to slowly stir overnight. Disulfide bond formation was mediated by air oxidation or through the addition of a glutathione exchange system. ShK and its derivatives spontaneously fold to a major thermodynamically favored isomer, which is the biologically active form of the peptides. The folded peptide was loaded onto a preparative RP-HPLC column and purified by using a gradient of ACN versus H2O containing 0.05% TFA. The fractions containing the desired peptide purity were pooled together and lyophilized to produce an acetate salt. Drug was formulated at 0.5 to 25 mg/ml in 10 mM sodium phosphate, 0.8% NaCl, and 0.05% polysorbate 20, pH 6.0.
Serum and Plasma Stability Studies.
Shk-186 peptide was solubilized at a concentration of 10 mg/ml in formulation buffer. Twenty five microliters of the stock peptide solution was spiked into 475 μl of a 1:1 dilution of serum, plasma, or whole blood with RPMI to a final concentration of 0.5 mg/ml and incubated at 37°C for the indicated period. Fifty microliters of the sample was then mixed with 50 μl of 4% trichloroacetic acid, and the sample was vortexed for 30 s and placed at 4°C for 15 min. The trichloroacetic acid mixture was centrifuged at 12,000 rpm for 5 min to pellet the precipitated proteins, and the supernatant was analyzed on an Agilent Technologies (Santa Clara, CA) 1100 HPLC system fitted with a Grace (Deerfield, IL) Vydac C18, 5.0-μm, 300-Å, 4.6 × 250-mm column. Mobile phase A was composed of 0.1% TFA in 95% water/5% ACN, and mobile phase B was composed of 0.1% TFA in 5% water/95% ACN. The system flow rate was 1 ml/min.
HPLC-MS/MS Method for Measuring ShK-186 and Metabolites in Plasma.
Whole-blood samples were collected into K2EDTA-containing tubes and processed by centrifugation to plasma. Plasma was supplemented 20:1 (v/v) with the HALT Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific, Waltham, MA) and stored frozen at −70°C until analysis. Fifty microliters of an internal standard [2 μg/ml ShK (parent peptide; accession number P29187)] in 90:10 H2O/ACN (v/v) (Bachem, Bubendorf, Switzerland) were added to 100 μl of plasma. The combined sample was diluted with 300 μl of H2O and purified by using a Waters (Milford, MA) Sep-Pak tC18, 25-mg, 96-well SPE plate. Samples were eluted in 500 μl of TFA/H2O/ACN (1:70:30, v/v) and evaporated under N2. The residue was reconstituted in 200 μl of TFA/H2O/ACN (0.02:90:10, v/v/v) and analyzed by HPLC-MS/MS. HPLC was performed on an Agilent 1200 series instrument fitted with an ACE 5 C18-PFP 50 × 2.1-mm, 5-μm column (Advanced Chromatography Technologies, Aberdeen, UK). Mobile phase A was formic acid/H2O (2:1000, v/v), and mobile phase B was formic acid/H2O/ACN (2:500:500, v/v/v). The flow rate was 300 μl/min. Mass spectrometry was performed on a SCIEX API 5000 instrument (AB Sciex, Foster City, CA) using turbo ion spray in the positive mode. Mass detection was performed by using multiple reaction monitoring (ShK-186, m/z 741.5 → 841.0; ShK-198, m/z 728 → 840.6; internal standard, m/z 676.8 → 769.0). The dwell times for the internal standard, ShK-186, and ShK-198 were each 500 ms.
SPECT/CT Scanning of Radiolabeled ShK-221.
ShK-221 (100 μg) was radiolabeled with 2 mCi 111indium chloride (GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK) in a 300-μl reaction containing 50 mM sodium acetate, pH 5.0 for 30 min at 95°C. The reaction was quenched by the addition of EDTA to a final concentration of 50 mM, and the radiolabeling efficiency was assessed by reverse-phase HPLC [Luna 5μ C18(2) 100A 250 × 4.6-mm column; Phenomenex, Torrance, CA] on an Agilent 1100 system using an IN/US Systems Gamma RAM model 4 radio-HPLC detector (LabLogic Systems, Brandon, FL). The labeling efficiency varied from 89 to 98% by this method. SPECT/CT scanning (NanoSPECT/CT Preclinical Imager; Mediso, Budapest, Hungary) was carried out on anesthetized animals in four 15-min scans during the first hour and one scan each at 4, 8, 24, 48, 72, 120, and 160 h postdose. The individual projection frame time for each helical SPECT was set such that the duration of each scan would last for approximately 15 to 45 min (varying by time point to account for isotope decay) and allow for significant collection of statistics within each frame. The characteristic peaks detected from the spectra for 111In were 245 and 171 keV (primary and secondary, respectively). The resulting projection data were reconstructed after each scan by using an iterative model that takes advantage of the pinhole geometry to achieve a resolution of approximately 2 mm.
Approximately 10 μl of blood samples were collected after each scan, and the amount of radioactivity in the sample was measured by using a Wallac Wizard 1470 scintillation counter (PerkinElmer Life and Analytical Sciences, Waltham, MA). Drug concentrations were computed by taking into account the specific activity of the administered dose, the half-life of 111In (67.3 h), and the counting efficiency of the instrument.
Suppression of Cytokine Responses in Peripheral Blood Mononuclear Cells.
PBMCs were isolated from human whole blood by using CPT Vacutainers (BD Biosciences, San Jose, CA) and dispersed into RPMI media. One hundred microliters of media containing 2 × 105 PBMCs were added to each well of a 96-well dish and treated with the addition of 50 μl of ShK-186 in media at varying concentrations for 1 h at 37°C, 5% CO2. Cells were washed twice with RPMI, resuspended in 200 μl of fresh media supplemented with 40 μM thapsigargin, and stimulated for 48 h at 37°C, 5% CO2. Cytokine production was measured in the overlying media by using a Luminex Assay (Millipore Corporation, Billerica, MA) specific for IL-2.
DTH Model in Rat.
Active DTH was induced and monitored as described previously (Beeton and Chandy, 2007). In brief, Lewis rats were immunized in the flanks with ovalbumin (OVA) (200 μg/rat) emulsified in complete Freund's adjuvant (CFA) (Sigma, St. Louis, MO). Seven to nine days later, animals were challenged under isoflurane anesthesia in the pinna of one ear with 20 μg of OVA dissolved in saline and in the other ear with saline. Animals received one or two subcutaneous injections of either ShK-186 or vehicle at the time of challenge or on the 4 days before challenge. Thickness of both ears was measured 24 h after challenge with a spring-loaded micrometer (Mitutoyo America Corporation, Aurora, IL).
CR-EAE Model in Rat.
The CR-EAE model in DA rats (Lorentzen et al., 1995) was used with minor modifications. In brief, animals were immunized by subcutaneous injection at the base of the tail with 0.2 ml of a 1:1 emulsion of homogenized SD rat spinal cord (Bioreclamation LLC, Westbury, NY) in CFA supplemented to 4 mg/ml Mycobacterium tuberculosis H37Ra (Sigma) under isoflurane anesthesia. Each rat received ∼80 mg of spinal cord and 400 μg of H37Ra. Either ShK-186 or placebo was administered subcutaneously on alternate flanks as appropriate. Animals were observed daily by measuring their body weight and assessing clinical signs of disease. Animals were included in the study and randomized into experimental groups sequentially as they reached a clinical score of 1. Scores were assigned as follows: 0, no illness; 0.5, no tail coil; 1, no tail coil and flaccid tail (tail dropped straight down five consecutive times); 2, mild paraparesis, wobbling; 3, moderate to severe paraparesis, falling on its side, unable to stand on hind legs; 3.5, one-limb paralysis; 4, two-limb paralysis; 5, two-limb paralysis with incontinence; 6, death. A cumulative clinical score (CS) was calculated for each rat by adding the daily scores from the day of disease onset (CS = >1) until the end of treatment and averaged to obtain a mean cumulative clinical score. Disease prevalence was calculated as the number of animals with CS >0.5 divided by the number of living animals per day and expressed as a percentage. During the chronic phase, an animal was considered to have a relapse if its clinical score was more than 1.
Pristane-Induced Arthritis in Rats.
Female DA rats received 150 μl of pristane (2,6,10,14-tetramethylpentadecane; CosmoBio, Carlsbad, CA) by subcutaneous injection in two sites at the base of the tail under isoflurane anesthesia. ShK-186 or vehicle was administered subcutaneously in the scruff of the neck daily or every other day for the duration of the trial. All four limbs were monitored for arthritis as described previously (Rintisch et al., 2009). In brief, a score of one point was given for each swollen and red toe, and each midfoot, digit, or knuckle, and five points for each swollen ankle or wrist (maximum score per limb, 15; maximum score per animal, 60). All of these studies were done under an IACUC-approved protocol at the Baylor College of Medicine.
Statistical and Computational Analysis.
Statistical analysis was carried out by using one-way ANOVA (EAE model and cytokine expression studies), the two-tailed, Mann-Whitney U test (DTH model), or the paired t test (pharmacokinetic studies). Goodness of model fit was determined by using the R2 statistic. Pharmacokinetic calculations were as follows: Cmax and Tmax were as observed in the dataset. AUC was computed by using a linear trapezoidal method. The terminal elimination half-life was computed from the slope of the regression with the best adjusted R2 value. AUCt-∞ was calculated by dividing the last observed drug concentration by the terminal elimination slope.
Pharmacokinetic Properties of ShK-186 in Representative Species.
As a precursor to PK studies of ShK-186 in nonclinical species, we evaluated the stability of the peptide in serum, plasma, and whole blood from humans, SD rats, and cynomolgus monkeys (M. fascicularis). Spiking-in studies showed the formation of a single metabolite in all samples analyzed from all three species. The metabolite was characterized by mass spectrometry and shown to be the dephosphorylated form of ShK-186 (referred to as ShK-198; Supplemental Fig. 1, A-C). ShK-198 is identical to the previously described analog ShK(L4) with the exception of containing a C-terminal amide in place of a carboxyl (Beeton et al., 2005). ShK(L4) blocks Kv1.3 with an IC50 of 48 pM (Beeton et al., 2005), which is similar to the potency of ShK-198 (Fig. 1, A and C; IC50 = 41.4 ± 7.25 pM; n = 5). Conversion of ShK-186 to ShK-198 occurred most readily in serum and in plasma samples treated with citrate or heparin as the anticoagulant. ShK-186 is most stable in plasma containing K2EDTA or specific phosphatase inhibitors (sodium fluoride, sodium orthovanadate, sodium pyrophosphate, and β-glycerophosphate), suggesting that endogenous phosphatases are responsible for its conversion.
We developed and validated an HPLC-MS/MS method to measure concentrations of ShK-186 and ShK-198 in K2EDTA plasma of rats and cynomolgus monkeys. The method had a lower limit of quantitation of 2 ng/ml (∼450 pM) for each analyte and showed equivalent recovery of ShK-186 from both spiked plasma and spiked buffer QC samples. Using this method, we measured the in vivo pharmacokinetic properties of ShK-186 after a single subcutaneous administration (the intended clinical route of administration) to rats and monkeys. Both rats and monkeys have TEM cells that express large numbers of the Kv1.3 channel after activation and are therefore relevant nonclinical species for testing ShK-186 (Chi et al., 2012).
The PK profile of ShK-186 is characterized by a short residence time in the central compartment of rat and monkey species. In rats, after a single administration, ShK-186 reaches a maximum plasma concentration between 1 and 5 min, whereas its metabolite reaches its Cmax between 1 and 15 min (Fig. 2, A and B; Tables 1 and 2). Although the Cmax and AUC for both the parent and metabolite increase with dose in rats, the relationship between Cmax and dose was largely nonlinear (R2 = 0.8), whereas the relationship between AUC and dose was approximately linear through the 500 μg/kg dose (R2 = 0.97; Fig. 2E). The half-life in rat ranged from 4.4 to 9.2 min for ShK-186 and from 6.1 to 16.4 min for ShK-198 (Tables 1 and 2). Clearance values for ShK-186 and ShK-198 in rats were not significantly different (paired t test; t4 = 2.2; p = 0.1) and ranged from 441.4 to 1453.3 ml/kg · min.
The PK of ShK-186 (Fig. 2C) and ShK-198 (Fig. 2D) after a single subcutaneous administration to monkeys demonstrated a linear relationship between dose and both AUC and Cmax (slope = 1.02–1.27; R2 = 0.97; Fig. 2F) through the entire range of doses from 35 to 1000 μg/kg. The monkey half-life ranged from 8.8 to 23.8 min for ShK-186 and from 16.7 to 73.5 min for ShK-198. Clearance rates in the monkey were generally lower for ShK-198 than ShK-186 and ranged from 18.7 to 141.3 ml/kg · min (Tables 1 and 2).
These data indicate that ShK-186 and its metabolite ShK-198 reach a maximum concentration in plasma 1 to 15 min after a single subcutaneous injection of drug in both species, and both are rapidly cleared from the central compartment. The limit of quantitation of the HPLC-MS method is approximately 10-fold higher than the IC50 of ShK-186 and ShK-198 on the Kv1.3 channel and thus lacks the sensitivity to estimate the terminal elimination phase at low, but potentially therapeutic, concentrations.
Absorption, Distribution, Metabolism, and Excretion Studies with a Radiolabeled Analog of ShK-186.
We developed a radiolabeled analog of ShK-186 to measure the biodistribution of the peptide and evaluate its total concentration in whole blood. ShK-186 contains a single iodinatable tyrosine at position 23. However, iodine incorporation into the ring, which is predicted to interact within the pore region of the Kv1.3 channel (Pennington et al., 1996), results in disruption of the channel binding properties of the peptide. We therefore modified the amino terminus of ShK-198 with a six-carbon linker attached via a peptide bond to one of the carboxylic acids of a 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid chelate (Supplemental Fig. 2A). The 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid conjugate, designated ShK-221, was readily coordinated with indium or gadolinium (Supplemental Fig. 2, B and C). In patch-clamp experiments, the gadolinium- and indium-labeled ShK-221 peptides blocked Kv1.3 with IC50 values similar to ShK-186 and ShK-198: Gd-ShK-221, 58.23 ± 1.38 pM (n = 5); In-ShK-221, 63.8 ± 2.25 pM; (n = 3); ShK-186, 68.99 ± 4.01 pM (n = 5); and ShK-198, 41.4 ± 7.25 pM (n = 5) (Fig. 1C). We prepared and administered 111In-labeled ShK-221 by subcutaneous injection to SD rats (1.0 mCi; 100 μg/kg) and squirrel monkeys (0.83 mCi; 35 μg/kg). The radiolabeling efficiency ranged from 89 to 98% over the series of experiments as determined by HPLC. Biodistribution of radiolabeled ShK-221 was evaluated by SPECT imaging continuously for the first hour postdose, and then at 4, 8, 24, 48, 72, 120, and 160 h (Supplemental Videos 1–12). The background level in the detection system was approximately 0.1μCi/ml (∼5 ng/ml of ShK-221 at the initial time point and 26 ng/ml at the last time point). Blood samples were collected after each scan, and total radioactivity in whole blood was measured by gamma counting. Computed tomography was performed at each time point to enable colocalization of the radiolabel with key anatomical structures.
Studies in Squirrel Monkeys.
Biodistribution of 111In-ShK-221 in squirrel monkeys was characterized principally by slow absorption from the injection site over the entire 160-h period (Fig. 3, A and E). The quantity of 111In-ShK-221 present at the injection site followed a biphasic exponential decay (R2 = 0.95) with an initial half-life of approximately 1 to 1.5 h and a terminal half-life of >48 h (Fig. 3E). During the first hour, significant radioactivity could be observed in the kidney, increasing in intensity through 1 h [∼1% injected dose (ID)/g; Supplemental Table 1] and slowly declining to approximately baseline by 48 h. Radioactivity in the monkey kidney was observed primarily in the cortical and medullary regions during all time points and was comparatively absent in the renal pelvis except for the first hour (Fig. 3B). Significant bladder associated radioactivity (Tmax = 0.75–1 h; 0.34% ID/g) was observed only during the first 4 h, after which relatively little radiolabel was detected in bladder. No other organ showed significant levels of radioactivity except for liver, which peaked at 0.75 to 1 h after dose administration (0.166% ID/g). Muscle, heart, and brain all had <0.1% ID/g at all time points measured (Supplemental Table 1).
Evaluation of blood-associated radioactivity in monkeys at each time point also demonstrated a biphasic exponential decay (R2 = 0.99) with an initial half-life of approximately 1 h and a terminal half-life of >64 h (Fig. 3F). In monkeys much of the terminal elimination phase was reflected by blood concentrations of 111In-ShK-221 (>200 pM) well above the IC50 for ShK-186 (69 pM), ShK-198 (41 pM), and In-ShK-221 (64 pM) for 6 days postdose. Even though ultrafiltration studies demonstrated significant binding of ShK-186 to plasma proteins, the channel-blocking activity of ShK-186 and In-ShK-221 were unaffected by the presence of serum (Fig. 1, B, D, and E), suggesting that the bulk of the peptide detected in blood is functionally active.
Studies in Rats.
Biodistribution of 111In-ShK-221 in the rat was similar to that in monkeys and characterized by slightly faster absorption from the injection site and excretion through the urine over the first 24 h (Fig. 3, C and E). Significant radioactive label was observed in rat bladder (9.4% ID/g), kidney (2.9% ID/g), and liver (0.4% ID/g) during the first hour (Fig. 3C; Supplemental Table 2). Although little label was identified in the bladder at later time points, the amount of drug in liver and kidney was relatively constant through the first 24 h. Cross-sectional views of the rat kidney showed that, with the exception of the first hour, radioactivity was concentrated primarily in the cortical regions similar to the monkey (Fig. 3D).
The whole blood-associated radioactivity in the rat also showed a biphasic exponential decay (R2 = 0.99) with an initial half-life of approximately 1.7 h and a terminal half-life of >72 h (Fig. 3F). The ShK-221 concentrations in the rat (200 pM) were well above the IC50 for ShK-186 and ShK-198 until 5 days postdose. In summary, the biodistribution of radiolabeled ShK-221 in squirrel monkeys and rats is characterized by a slow absorption from the injection site, significant concentrations of drug peripherally in the injection site, kidney, and liver, and a long terminal elimination phase in whole blood.
Suppression of Cytokine Responses in Human PBMCs.
Another way that ShK-186 could exert a durable PD effect is through a sustained interaction with the Kv1.3 channel on lymphocytes. We therefore assessed the duration of lymphocyte suppression after exposure to ShK-186. We have previously shown that ShK-186 suppresses thapsigargin-induced expression of IL-2, IL-17, interferon-γ, and IL-4 in freshly isolated human PBMCs (Chi et al., 2012). Here, we exposed PBMCs to the drug either continuously during a 48-h period of thapsigargin stimulation, for 1 h before 48 h of stimulation, or for 1 h up to 16 h before stimulation and measured the IL-2 responses by enzyme-linked immunosorbent assay (Fig. 4). In cases of transient exposure, the cells were thoroughly washed with media before thapsigargin treatment. There was no statistically significant difference (one-way ANOVA; F3,12 = 0.40; p = 0.76)between continuous drug treatment versus transient drug exposure up to 16 h before thapsigargin stimulation. These data are consistent with a model where ShK-186 rapidly associates with, but slowly dissociates from, the Kv1.3 channel of lymphocytes.
Suppression of the DTH Response In Vivo: Dose Frequency and Dose-Response Studies.
Our absorption, distribution, metabolism, and excretion studies suggested that a low, but therapeutically, relevant drug concentration persists in the blood for several days after a single administration of ShK-186. We therefore evaluated the relationship between dose frequency and therapeutic efficacy in animal models of effector-memory T cell-mediated disease. The simplest model for assessing drug effects is the DTH reaction, an immune response largely mediated by skin-homing effector-memory CD4+ T cells (Soler et al., 2003; Matheu et al., 2008). We immunized groups of five to eight Lewis rats with OVA in CFA followed 7 to 9 days later by elicitation of the DTH response in the ears of immunized animals. Ear swelling was measured 24 h postelicitation. The durability of drug effect was assessed by treating animals with ShK-186 at varying times before or coincident with the elicitation phase. The day of ear challenge is referred to as day 0 for reference.
Animals that were treated with two doses of 10 or 100 μg/kg ShK-186 on days 0 and 1 (the standard regimen) were compared with single 100 μg/kg doses administered on days 0, −1, −2, −3, and −4. Two doses administered on days 0 and 1 and single doses administered on days −1 to −4 yielded statistically significant reductions in ear swelling relative to placebo-treated animals (two-tailed Mann-Whitney test; Fig. 5A). Single doses administered before day −4 did not result in a significant reduction in ear swelling relative to the control group (data not shown). We conclude that a single subcutaneous injection of ShK-186 is sufficient to suppress the DTH response in rat for 5 days, consistent with the long terminal elimination half-life of radiolabeled ShK-221.
To establish a relationship between dose level and the time of dosing, we evaluated the suppressive effect of a single ascending dose (0.1–100 μg/kg) of ShK-186 administered 2 days before ear challenge. A dose-dependent reduction in ear swelling was observed over the three-log dose range, with doses ≥1 μg/kg achieving statistical significance relative to vehicle-treated animals (Fig. 5B).
Treatment Strategies in the CR-EAE and PIA Models.
CR-EAE is widely used as a model of multiple sclerosis. This model has been extensively characterized in guinea pigs (Raine et al., 1977; Keith et al., 1979) and DA rats (Feurer et al., 1985; Lorentzen et al., 1995), and unlike the acute EAE models includes relapses and some degree of demyelination (Lassman and Wisniewski, 1979; Lorentzen et al., 1995; Matheu et al., 2008). Moreover, because of its protracted nature, it is an adequate model for testing dosing schedules. The initial wave of disease is largely adjuvant mediated and composed primarily of central memory T cells, whereas subsequent waves of disease are principally TEM cell-mediated and therefore Kv1.3-dependent (Matheu et al., 2008). We have previously shown ShK-186 to be effective in the CR-EAE model by using daily dosing at 100 μg/kg (Matheu et al., 2008). In the present study, we explored the effect of less frequent dose administration on drug efficacy in this model.
Two study designs were explored. In the first, a lead-in period with daily drug administration was initiated at the time of disease induction and continued throughout the first wave of disease. Seven days after disease onset, animals were randomized to receive 100 μg/kg ShK-186 every 2 or 3 days through the remainder of the study. Daily drug treatment caused a statistically significant reduction in mean clinical score compared with placebo-treated animals during the first wave of disease (Fig. 6A). After lead-in, drug administration every 2 or 3 days resulted in continued significantly lower mean daily clinical score in treated relative to control animals. The average cumulative clinical score was 161.1 ± 162.3, 72.5 ± 98.1, and 71.3 ± 103.3 for the placebo, every-2-days, and every-3-days dosing groups, respectively. Animals were scored twice daily.
In a second series of experiments, we evaluated the effect of reduced dose frequency on EAE without the drug lead-in period. In this study, groups of DA rats were randomized to receive placebo, daily ShK-186, or ShK-186 every 3 days beginning at a clinical score ≥1. Both drug-treated groups exhibited a significantly lower mean daily clinical score than placebo-treated animals (Fig. 6B). However, there was no statistically significant difference between daily and every-third-day dose administration. Mean cumulative clinical scores were 49.7 ± 26.7, 36.9 ± 20.6, and 30 ± 19.3 for the placebo, daily, and every-third-day dosing groups, respectively. The percentage of rats with one or more relapses was 70% (9/13), 21% (3/14), and 21% (3/14), and the total number of relapses observed during the chronic phase were 15, 6, and 5 for the placebo, ShK186 daily, and every-third day-treated groups, respectively. Animals were scored once daily.
Finally, we wanted to evaluate the durability of disease suppression in the CR-EAE model after cessation of drug administration. Disease was elicited in groups of DA rats, and the animals were allowed to develop a relapsing disease. On day 10 after disease onset (clinical score ≥1), animals were randomized to receive daily placebo or 100 μg/kg ShK-186 for 14 days. Drug treatment was discontinued on day 24, and the animals were monitored for an additional 15 days. There was no difference between the placebo- and drug-treated groups before drug treatment (Fig. 6C). During the 14 days of daily drug administration, there was a statistically significant reduction in mean clinical score in drug-treated versus control animals. Once treatment was withdrawn, the mean clinical score in the drug-treated group slowly returned to the level of placebo over approximately 5 days (Supplemental Table 3). However, there was no rebound effect observed in the previously treated animals.
As a final test of the effectiveness of less than daily dosing, we evaluated ShK-186 administration every other day in the PIA model of rheumatoid arthritis (Beeton et al., 2006). ShK-186 administered once daily (100 μg/kg) has been previously reported to reduce the number of joints affected and the severity of joint swelling in the PIA model in DA rats (Beeton et al., 2006). Here, we demonstrate that ShK-186 administered on alternate days (100 μg/kg) was as effective in ameliorating disease (Fig. 7) as our previously reported effect with once-daily administration (Beeton et al., 2006).
The data from these four studies are consistent with a model of durable drug effects after a single subcutaneous dose of ShK-186 and observations made using the DTH model.
Here, we present five complementary types of data that relate the PK properties of ShK-186 to its therapeutic efficacy in rat models of autoimmune diseases. Through serum/plasma stability studies, we identify the sole metabolite of ShK-186 in rat, monkey, and human samples as the dephosphorylated peptide ShK-198. ShK-198 blocks Kv1.3 (IC50 40 pM) with approximately the same potency as ShK-186 (69 pM). Second, we demonstrate that the Cmax of both ShK-186 and ShK-198 is rapidly reached after subcutaneous injection to rats and monkeys. Both parent and metabolite have a short apparent half-life in the central compartment with >90% of the Cmax eliminated by 2 h in both species at all tested doses. The half-life is shorter and the apparent clearance greater for ShK-186 than ShK-198, probably reflecting the independent action of endogenous phosphatases on the conversion of ShK-186 to its metabolite. Third, SPECT imaging of rats and monkeys administered an 111In-labeled ShK-186 analog (ShK-221; IC50 65 pM) by subcutaneous injection revealed that the peptide is released slowly from the injection site and has an extended terminal half-life, with whole-blood levels above the IC50 value for Kv1.3 block for 5 days in rats and 7 days in monkeys. The Kv1.3 channel-blocking affinity of ShK-186 and ShK-221 is not affected by the presence of serum, indicating that plasma protein binding, if any, does not affect the functional activity of these peptides. Fourth, in vitro proliferation assays demonstrate that brief 1-h exposure of PBMCs to ShK-186 is sufficient to suppress interleukin-2 production 64 h later. This suggests that the peptide, once bound to the Kv1.3 channel, dissociates very slowly. Finally, in three different rat models of TEM cell-mediated inflammatory diseases (DTH, CR-EAE, and PIA) ShK-186 is as effective in ameliorating disease when administered every 2 to 5 days as when it is administered once daily. The durable PD effect of ShK-186 is contrary to conventional wisdom that frequent administration of peptide therapeutics is required to sustain PD activity.
Our findings highlight a potential advantage of developing peptides from venoms as drug candidates. SPECT imaging studies with 111In-labeled ShK-221 revealed a biphasic (fast then slow) release from the subcutaneous injection site over 7 days in rats and monkeys. This type of rate-limiting absorption resulting in prolonged plasma exposure has been described for whole animal venoms (Barral-Netto et al., 1990; Audebert et al., 1994). The long absorption phase has been used to explain the frequent relapse in envenomation patients long after exposure and/or treatment with antivenom (Dart et al., 2001; Seifert and Boyer, 2001; Gutiérrez et al., 2003), and venoms are often described as having a depot effect. The parent peptide of ShK-186 was originally isolated from the venom of Stichodactyla helianthus (Castañeda et al., 1995), and ShK-186 may share some of the in vivo characteristics of the complex peptide mixtures found in animal venom. Venom and toxin PK parameters are also characterized by large apparent volumes of distribution and slow elimination from deep and shallow peripheral compartments (Ismail et al., 1996).The apparent volume of distribution for ShK-186 is large in both rats (2743–16,458 ml/kg) and monkeys (1729–2664 ml/kg), consistent with that reported for other animal venoms. The blood concentration of 111In-ShK-221 mimicked the absorption of the peptide from the injection site and was characterized by a rapid initial phase and a very long terminal phase. The terminal half-life computed by using 111In-ShK-221 was >64 h in monkeys with sustained blood levels for 7 days above the IC50 values for ShK-186 (69 pM) and ShK-198 (40 pM). The presence of serum does not affect channel-blocking affinity of these peptides, indicating that plasma protein binding, if any, does not affect functional activity.
Glomerular filtration is the principal elimination pathway for the peptide shortly after subcutaneous injection. Significant amounts of radioactivity were observed in the bladder of both rats (∼17% injected dose) and monkeys (∼1% injected dose) at the earliest time points after administration of 111In-ShK-221. The large amount of drug excreted by the rat in the first hour is most likely a reflection of the increased metabolism of the rat compared with the monkey. After 1 h in rats and approximately 4 h in monkey, little radioactivity was observed in bladder or renal pelvis, whereas significant amounts of radioactivity could still be observed in the kidney cortex. Cortical concentration has been reported for numerous radiolabeled versions of peptide drugs including octreotide, bombesin, exendin, and gastrin (Gotthardt et al., 2007). The mechanism of cortical retention has been most thoroughly described for octreotide. Tubular reabsorption of the cationic octapeptide is mediated by megalin, a scavenger receptor expressed in the proximal kidney tubule (de Jong et al., 2005). Mice with a kidney-specific disruption of the receptor lack the cortical retention of radiolabeled octreotide seen in wild-type mice. Renal uptake of octreotide is partially mediated by charge and can be disrupted by coinfusion of the positively charged amino acids l-lysine and l-arginine (Bodei et al., 2003). ShK-186 carries a net +6 charge at physiological pH, so its cortical retention may be mediated by a similar mechanism.
Another contributing factor to the long PD effect of ShK-186 along with an extended terminal half-life could be its slow dissociation from the Kv1.3 channel. Studies with PBMCs show that there is essentially no difference between continuous ShK-186 treatment during thapsigargin stimulation and treatment up to 16 h before stimulation. Although we did not formally measure an off-rate of the drug on the Kv1.3 channel by using traditional methods, we did show that ShK-186's PD effect can persist for at least 72 h after its brief exposure to PBMCs. The receptor dissociation half-life of some animal toxins has been reported to be as long as 1 week (Berg and Hall, 1975, Chang and Huang, 1975). Future receptor binding studies using the radiolabeled analog will allow determination of the off-rate of ShK-186 from Kv1.3.
Absorption, distribution, metabolism, and excretion studies with 111In-labeled ShK-221 suggest that a single dose of peptide can provide therapeutically meaningful blood concentrations for up to 5 days in rats and 7 days in monkeys. These data are consistent with data from the rat DTH model where a single administered dose of ShK-186 provided a statistically significant reduction in ear swelling for up to 5 days. Likewise, dose administration every 2 to 3 days in the CR-EAE and PIA models were as effective in ameliorating disease as daily ShK-186 administration. Because ShK-186's therapeutic effects are long-lasting, we anticipate administering the peptide to humans once weekly or less frequently. From a therapeutic and commercial perspective, ShK-186 will compete favorably with pills taken daily or multiple times a day or injectables.
Participated in research design: Tarcha, Chi, Muñoz-Elías, Bailey, Ruppert, Boley, Beeton, Slauter, Knapp, Pennington, Chandy, and Iadonato.
Conducted experiments: Tarcha, Chi, Muñoz-Elías, Bailey, Upadhyay, Norton, Banks, Beeton, Tjong, Nguyen, Hu, Kentala, Hansen, and Iadonato.
Contributed new reagents or analytic tools: Tarcha, Bailey, and Pennington.
Performed data analysis: Tarcha, Chi, Muñoz-Elías, Bailey, Upadhyay, Nguyen, Knapp, Beeton, Chandy, and Iadonato.
Wrote or contributed to the writing of the manuscript: Tarcha, Chi, Muñoz-Elías, Chandy, and Iadonato.
We thank Srikant Rangaraju for testing the Kv1.3 channel-blocking activity of ShK-186 in the presence or absence of serum.
This work was supported by the National Institutes of Health National Institute of Allergy and Infectious Diseases Extramural Activities [Grant R43-AI085691] (to S.I.); the National Institutes of Health National Institute of Neurological Disorders and Stroke [Grant R01-NS48252] (to K.G.C); and the Regents of the University of California (Irvine) [UC Discovery Grant UCOP Bi009R-156245] (to K.G.C.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- effector-memory T
- analysis of variance
- area under the concentration-time curve
- complete Freund's adjuvant
- clinical score
- computed tomography
- single-photon emission CT
- dark agouti
- delayed-type hypersensitivity
- experimental autoimmune encephalomyelitis
- chronic-relapsing EAE
- high-pressure liquid chromatography
- institutional animal care and use committee
- injected dose
- Infectious Disease Research Institute
- mass spectrometry
- tandem MS
- peripheral blood mononuclear cell
- pristane-induced arthritis
- trifluoroacetic acid.
- Received January 18, 2012.
- Accepted May 23, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics