|Home | About | Journals | Submit | Contact Us | Français|
Peptoids are a rapidly developing class of biomimetic polymers based on oligo-N-substituted glycine backbones, designed to mimic peptides and proteins. Inspired by natural antimicrobial peptides, a group of cationic amphipathic peptoids has been successfully discovered with a potent and broad-spectrum activity against pathogenic bacteria; however, there are limited studies to address the in vivo pharmacokinetics of the peptoids. Herein, 64Cu labeled DOTA conjugates of three different peptoids and two control peptides were synthesized and assayed in vivo by both biodistribution studies and small animal positron emission tomography (PET). The study was designed in a way to assess how structural differences of the peptidomimetics affect in vivo pharmacokinetics. As amphipathic molecules, major uptake of the peptoids occurred at the liver. Increased kidney uptake was observed by deleting one hydrophobic residue in the peptoid, and 64Cu-3 achieved the highest kidney uptake of all the conjugates tested in this study. In comparison to peptides, our data indicated that peptoids had general in vivo properties of higher tissue accumulation, slower elimination, and higher in vivo stability. Different administration routes (intravenous, intraperitoneal, and oral) were investigated with peptoids. When administered orally, the peptoids showed poor bioavailability, reminiscent to that of peptide. But, remarkably longer passage through the gastrointestinal (GI) tract without rapid digestion was observed for peptoids. These unique in vivo properties of peptoids were rationalized by efficient cellular membrane permeability and protease resistance of peptoids. The results observed in the biodistribution studies could be confirmed by the PET imaging, which provides a reliable way to evaluate in vivo pharmacokinetic properties of peptoids noninvasively and in real time. The pharmacokinetic data presented here can provide an insight for further development of the antimicrobial peptoids as pharmaceuticals.
Over recent decades a number of antimicrobial peptides (AMPs) have been identified as a frontline defense in various host organisms.1, 2 AMPs have drawn great scientific interest because they not only belong to the intrinsic host-defense system,3-6 but also are promising candidates for the development of novel anti-infective agents.7, 8 Natural AMP-based therapeutics demonstrate activities against broad spectrum pathogens in vitro; however, they exhibit major drawbacks in vivo such as their low resistance against proteolysis and the rapid clearance by the kidneys.9, 10 Recently, the challenges associated with peptides are being addressed owing to the use of non-natural amino acids and modified peptide structures;11, 12 hence, various peptidomimetics that are designed to reproduce critical AMP’s structural characteristics, such as spatially segregated cationicity and hydrophobicity in an amphipathic secondary structure, have been developed.13-18 Particularly, short (<40 amino acids), cationic, linear, and α-helical AMPs are amenable to mimicry, and peptoids are among the front-runners in mimicking this class of AMPs.
Peptoids (oligo-N-substituted glycines) are a novel class of peptidomimetics that differ from peptides only in that side chains are attached to the backbone amide nitrogen instead of the α-carbon (Figure 1).19, 20 This unique structural feature of peptoids leads to quite different pharmacological properties from natural peptides: (1) they are highly stable against proteases;21, 22 (2) they lack hydrogen-bonding potential, which further prevents backbone-driven aggregation and thus increases bioavailability;23, 24 (3) they show increased cell permeability over peptides;23, 25 and (4) they may evade immune recognition.26 Peptoids can be readily synthesized using conventional solid phase peptide synthesis (SPPS) method, and diverse sequences of peptoids can be prepared at relatively low cost. They are also precisely tunable in conjugation with established synthetic methods.27 Since their development in the early 1990s, peptoids have become attractive scaffolds for many different biological applications.28-30 For example, the use of peptoids as antimicrobial agents,31-36 molecular transporters for intracellular drug delivery,37 or ligands for tumor receptors binding has been reported.25
Among many peptidomimetic systems under study, peptoids are particularly well suited for structural mimicry of AMPs31, 32 because they can readily form helical secondary structures via a periodic incorporation of bulky, α-chiral side chains.38-40 The structures of the peptoid helices have been characterized to be stable polyproline type I-like helices which exhibit a pitch of 6.0-6.7 and a periodicity of approximately three residues per turn.41-43 The threefold periodicity of the peptoid helix makes it easier to mimic the facially amphipathic structures similar to those found in many AMPs; for example, the trimer repeat (X-Y-Z)n forms a peptoid helix with three faces, composed of X, Y, and Z residues. Previously, a group of cationic amphipathic peptoids, with broad-spectrum antimicrobial activity against both Gram-positive and Gram-negative bacteria including clinically relevant biosafety level 2 (BSL-2) pathogens, has been discovered.31-34 Interestingly, these antimicrobial peptoids also show potential in treating multidrug resistant strains such as Pseudomonas aeruginosa biofilms35 and Micobacterium tuberculosis.36 Furthermore, we recently found that the cationic amphipathic peptoids were potent against several cancer cell lines at low-micromolar concentrations and their actions were not affected by multidrug resistance (MDR) developed in cancer cells.44 Therefore, cationic amphipathic peptoids represent an excellent molecular platform for the development of antimicrobials as well as cytotoxic chemotherapeutics and warrant systematic investigations on their in vivo behaviors.
The in vivo pharmacokinetic studies on antimicrobial peptides have been reported using radiolabeled peptides;45, 46 however, we cannot simply extrapolate peptoid biodistribution from previous peptide biodistribution data. Firstly, due to the lack of hydrogen-bond donors in the peptoid backbone, the hydrodynamic radii of peptides and peptoids are expected to be significantly different.24 Secondly, peptoids are protease-resistant and metabolically more stable than peptides.21, 22 For these reasons, the pharmacokinetic profiles of peptoids are likely to be different from what was observed with peptides, and a new set of pharmacokinetic data are required for better understanding the in vivo behavior of this unique family of peptidomimetics.
Recently, we found the antimicrobial peptoids showed different in vitro antimicrobial activities depending on their degree of helicity, chain length, and hydrophobicity (Table 1; Note that the biological data in Table 1 are for oligomers without DOTA conjugation). Then we reasoned that the different structural and physicochemical properties could also cause distinct in vivo pharmacokinetic profiles of the peptoids. Herein, three cationic amphipathic peptoids and two control peptides were selected, and their in vivo profiles were evaluated using both biodistribution and small animal positron emission tomography (PET). Peptoids 1, 2, and 3 represent N-terminal 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) conjugates of cationic amphipathic peptoids (Table 1 and Figure 1). Peptoids 1 and 3, where the latter is almost identical to 1 except that it has one less C-terminal Nspe monomer, are linear peptoid helices. Peptoid 3 (without DOTA) has lower hydrophobicity than peptoid 1 (without DOTA), and was recently found to have substantially higher selectivity in killing bacterial cells without significant loss in potency (Table 1). Peptoid 2 (without DOTA) is a non-structured peptoid, which nonetheless can adopt a helical conformation in the presence of the anionic bacterial membrane that serves as an organizing surface. For comparison, we tested two cationic amphipathic peptide controls: (1) peptide 4 has a direct sequence homology to peptoid 1, and the effect of the peptoid backbone versus the peptide backbone was examined; and (2) peptide 5 is based on a well-known antimicrobial peptide, namely pexiganan,47, 48 which advanced to phase 3 clinical trials.49 Also known as MSI-78, pexiganan exhibits a random coil conformation, which folds into a helical conformation upon binding to a bacterial membrane (Table 1). To obtain kinetic data for the biodistribution and clearance for 24 hours, a radionuclide with an appropriate half-life was needed. Therefore, DOTA-chelated 64Cu was used due to its relatively long half-life (762 minutes) and numerous precedent biodistribution data of the DOTA-64Cu labeled peptides.50-52 Using the radiolabeled peptoids and peptides, three different administration routes (intravenous, intraperitoneal, and oral) were examined. The results reported here can provide an insight for future development of peptoids as antimicrobial chemotherapeutics.
All reagents were purchased from Sigma-Aldrich, Fisher and Novabiochem. They were used without further purification unless stated otherwise. DOTA-tris(t-Bu ester) (or Tri-tert-butyl 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetate) was purchased from Macrocyclics, TX, USA. N,N’-Diisopropylcarbodiimide was purchased from Advanced ChemTech, KY, USA. Radioisotope 64Cu was purchased from the Department of Medical Physics, University of Wisconsin, Madison, WI, USA. Six-week old Balb/c mice were ordered from Harlan, CA, USA.
The peptoid oligomers were synthesized by solid-phase submonomer method using an automated peptide synthesizer (ABI 433A, Applied Biosystems, CA, USA).20, 53 Rink amide resin (0.60 mmol/g, Novabiochem, San Diego, CA) was used to generate C-terminal amide peptoids. After Fmoc deprotection, each monomer was added by a series of bromoacetylation and displacement by an amine. These two steps were iterated with appropriate amines until desired peptoid sequence was obtained. Typically, 0.06 mmol reaction scale was used (0.1 g of the resin). For bromoacetylation, the addition of bromoacetic acid [1 mL, 1.2 M in N,N-Dimethylformamide (DMF), 1.2 mmol] followed by N,N’-diisopropylcarbodiimide (0.15 mL, neat, 1.0 mmol), and then agitation at room temperature for 20 min were performed. For the displacement reaction, primary amine [1 mL, 1.5 M in N-methylpyrrolidone(NMP), 1.5 mmol] was added, and the reactor was agitated at room temperature for 1 h. The resin was washed with DMF (6 mL × 5) between each step. Peptide oligomers were prepared by the automated solid-phase peptide synthesis according to standard Fmoc/tBu-methodology and HBTU/HOBt/DIEA coupling.54 Rink amide resin (0.60 mmol/g) was used to obtain C-terminal amide peptides. After completion of the solid-phase synthesis, the DOTA chelator was manually conjugated at the N-terminus of the pepoids or peptides by activating the carboxylic acid of DOTA-tris(t-Bu ester) with HATU. Peptoid (or peptide) attached on resin (properly protected, 0.025 mmol) was swelled in DMF (2 mL). To this was added DOTA-tris(t-Bu ester) (0.15 mmol), 2-(7-Aza-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU) (0.30 mmol), and N,N-diisopropylethylamine (DIPEA) (0.75 mmol), and the reaction continued at room temperature overnight.55 Cleavage from the resin and deprotection was performed with 95:2.5:2.5 TFA/water/triisopropylsilane (v/v/v) for 2-3 hours at room temperature. It should be noted that shorter cleavage time resulted in incomplete tert-butyl ester deprotection of DOTA-tris(t-Bu ester). The cleavage solution was filtered by solid-phase extraction (SPE) cartridges (Applied Separations, Allentown, PA, USA) and the volatiles were removed by a stream of nitrogen. The crude peptoid was dissolved in acetonitrile/water, lyophilized, and analyzed by ESI-MS (or MALDI-TOF), and analytical HPLC.
Analytical HPLC was performed on a Waters 2695 separations module system with Waters 2998 photodiode array detector and Waters Alliance column heater. Phenomenex Jupiter C18 (250 × 2.0 mm, 5, 300 Å) column was used for the analytical HPLC and the column was heated at 55 °C. Sample purity was monitored by absorbance at 220 nm. The mobile phase was used as follows: [A, water + 0.1% trifluoroacetic acid (TFA); B, CH3CN + 0.1% TFA] 3 minutes using 1% B, then a linear gradient to 99% B over 30 minutes, a linear gradient back to 1% B over 10 minutes, and then 2 minutes using 1% B. Flow rate of the analytical HPLC was 0.2 mL/min.
The peptoids or peptides were purified by a preparative HPLC system (Waters PrepLC system, Waters 2489 UV/Visible detector, Waters fraction collector III) using a C18 column (Phenomenex Jupiter C18, 250 × 21.20 mm, 10 m, 300 Å) at a flow rate of 25 mL/min. Sample elution was detected by absorbance at 220 and 260 nm. The purity of the product fractions were confirmed by analytical HPLC, and fractions containing pure product (>98% purity) were collected, lyophilized, and stored at −80 °C.
For electrospray ionization mass spectrometry (ESI-MS) analysis of compounds, ThermoFinnigan LCQ “Classic” ion trap LC-MS instrument was used (Stanford University Mass Spectrometry facility). Voyager-DE RP Biospectrometry instrument was used for matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry (Stanford University PAN facility).
Typical procedure for the preparation of Cu-1 - Cu-4 (for CD study) is as follows. Compound 1 (1.0 mg) was dissolved in NaOAc buffer (0.1 M, pH 5.5, 940 L). To this was added 10 mM CuCl2 (60 L, approximately 2.0 molar equivalent). The reaction mixture was incubated at room temperature overnight with a gentle agitation. The reaction was monitored by analytical HPLC and stopped when the starting material (compound 1) was consumed. D-Salt™ polyacrylamide desalting column (Pierce, Rockford, IL, USA) was used to purify the crude product. The purification was carried out according to the manufacturer’s instruction. The column fractions were monitored by measuring the absorbance at 220 nm using Nanodrop spectrophotometer. Fractions containing Cu-1 were combined and lyophilized.
Circular dichroism measurements were carried out at 20 °C with a Jasco J-815 circular dichroism spectrometer (Jasco, Inc., Easton, MD, USA). CD spectra were obtained in an 1 mm pathlength quartz cell and recorded from 190 to 260 nm in 1 nm increments with a scanning speed of 20 nm/min and a response time of 4 second. Three scans for each sample were averaged. Sample concentrations were in 50 M. Data are expressed in terms of per-residue molar ellipticity (deg cm2/dmol) calculated per number of amides in a molecule.
DOTA-conjugated compounds (10 μg) were used for labeling. The bioconjugate was radiolabeled with 64Cu by addition of 64CuCl2 (111 MBq, 3 mCi) in NaOAc buffer (pH 5.5, 0.1 N) followed by a 40 min incubation at 40 °C. EDTA (5 μL, 10 mM) was then added to quench free 64Cu2+. The radiolabeled complex was purified by analytical radio-HPLC. The mobile phase was used as follows: [A, water + 0.1% TFA; B, CH3CN + 0.1% TFA] 3 minutes using 5% B, then a linear gradient to 65% B over 30 minutes, then going to 85% B over 3 minutes, maintaining this solvent composition for another 3 minutes, and then a linear gradient back to 5% B over 3 minutes (total 42 minutes). Flow rate of the radio-HPLC was 1.0 mL/min. Fractions containing pure radiolabeled peptoid (or peptide) were combined and dried under reduced pressure using rotary evaporator. The radiolabeled compound was reconstituted in phosphate-buffered saline (PBS) and passed through a 0.22 μm Millipore filter into a sterile vial for in vitro and animal experiments.
64Cu-DOTA-peptoids or peptides (30 - 40 μCi) were injected into a tail vein in Balb/c mice (n = 4 for each group), and then the mice were sacrificed at 2 h and 24 h p.i. For peptoid 64Cu-1, three administration routes including intraperitoneal (i.p.), intravenous (i.v.) or gavages needle (p.o.) injection were used to compare the in vivo profiles of the peptoid. Major organs were excised and weighed, and their radioactivity was measured in a Wallac 1480 automated γ–counter (Perkin Elmer, MA, USA). The radioactivity uptake in the organs was expressed as a percentage of the injected radioactive dose per gram of tissue (% ID/g).
Small animal PET imaging of Balb/c mice was performed on a microPET R4 rodent model scanner (Siemens Medical Solutions USA, Inc., Knoxville, TN, USA). The mice were injected with 64Cu-DOTA-peptoids or peptides (~40 μCi) through the tail vein. A dynamic image of first 35 min and a series of static images were acquired. During the scan, anesthesia was maintained with 2% isofluorane in 1 L/min oxygen, and imaging was performed in a prone position. For dynamic scan, image acquisition was initiated once the mice were centered in the field of view. Radiolabeled peptoid was administered with immediate data collection and image acquisition. Acquisition lasted about 35 minutes. Image reconstruction was performed by a two-dimensional OSEM (ordered subsets expectation maximum) algorithm with a spatial resolution of 1.66 -1.85 mm. No attenuation or partial volume corrections were performed. Regional of interests (ROIs) were selected using ASIPro VMTM software (Siemens Medical Solutions USA Inc., Knoxville, TN, USA). For static scan at different times p.i. (0.25, 1, 2, 4 and 24 h), the mice were anesthetized with 2% isofluorane and placed in the prone position and in the center of the field of view of microPET. The 5-min static scans were obtained, and the images were reconstructed by the OSEM algorithm. PET images were imported using ASIPro VMTM. ROIs were drawn manually over the organ of interest on decay-corrected whole-body coronal images. The mean counts per pixel per minute were obtained from the ROIs and converted to counts per milliliter per minute by using a calibration constant. No attenuation correction was performed.
Statistical analysis was performed using the Student’s t-test for unpaired data. A 95% confidence level was chosen to determine the significance between groups, with P < 0.05 being significantly different.
Five DOTA conjugated compounds (three different cationic amphipathic peptoids 1-3 and two peptide controls 4–5 in Figure 1) were successfully synthesized by solid-phase synthesis followed by N-terminal DOTA conjugation. Both peptoid and peptide sequences were synthesized by automated peptide synthesizer according to the submonomer protocol20 and standard Fmoc/tBu method,54 respectively (Figure 2A). Then, the DOTA chelator was manually conjugated at the N-terminus of the pepoids or peptides by activating the carboxylic acid of DOTA-tris(t-Bu ester) with HATU (Figure 2B).55 After deprotection and cleavage from the resin with trifluoroacetic acid, the DOTA conjugates were purified by preparative HPLC. It should be noted that less than 2 hours of cleavage/deprotection resulted in incomplete tert-butyl ester deprotection of DOTA-tris(t-Bu ester); therefore, 2-3 hours of cleavage reaction were used for all five compounds. The identity of each compound was confirmed by mass spectrometry (Supporting Information, Table S1).
Subsequently, the conjugates were radiolabeled with 64Cu in order to monitor their in vivo behaviors (Figure 2C). The purification of radiolabeled complex, using a PD-10 column, afforded 64Cu-DOTA-peptoid/peptide (64Cu-1 – 64Cu-5) with >95% radiochemical purity and modest specific activity of 3.98 - 6.83 MBq/nmol. Further mouse serum stability test demonstrated that over 95% radiolabeled complexes remained intact after 1 h incubation at 37 °C (Supporting information, Figure S5).
Conformational information of unmetallated 1 – 4 as well as copper incorporated conjugates Cu-1 - Cu-4 was obtained by circular dichroism (CD) spectroscopy (Figure 3). CD study of pexiganan was previously reported (Table 1);56 therefore, compound 5 was not included in this study. For the CD study, non-radioactive copper bound DOTA conjugates, namely Cu-1 – Cu-4, was prepared, and the physicochemical properties of radioactive and non-radioactive coppers-compounds are assumed to be the same.
For helical peptoid conjugates 1 and 3, the maintenance of helical folds was observed after copper incorporation (Cu-1 and Cu-3). In addition, conformations of 1 and Cu-1 at body temperature (37 °C) were examined and indicated minimal temperature dependence at 20 °C and 37 °C (Supporting information, Figure S2). As predicted, compounds 2 and Cu-2 indicated almost non-structured conformation due to the lack of structure inducing residue, Nspe or (S)-N-(1-phenylethyl)glycine, in the peptoid sequence. Uncharacteristic, but α-helix like, CD signatures were obtained for 4 and Cu-4, and the handedness was the opposite of the peptoid conjugates’ handedness. The CD spectra in this study clearly showed the influence of the copper incorporation on the overall conformation was minimal.
Radiolabeled peptoids and peptides 64Cu-1 – 64Cu-5 were injected intravenously into a tail vein in Balb/c mice. After 2 h and 24 h postinjection (p.i.), the mice were euthanized, and the radioactivity of the organs and the blood were determined. The activity in the selected organs was expressed as a percentage of the injected radioactivity per gram of tissue (%ID/g). The results of the biodistribution studies are summarized in Figure 4.
Peptoid conjugates 64Cu-1 and 64Cu-2 exhibited prominent uptake in the liver, with lower liver uptake for 64Cu-2 than 64Cu-1. The detected activity in the liver was 137.6 ± 13.4 %ID/g for 64Cu-1 and 125.3 ± 16.8 %ID/g for 64Cu-2 at 2 h p.i., and the activity remained at 95.6 ± 10.5 %ID/g for 64Cu-1 and 72.2 ± 11.3 %ID/g for 64Cu-2 at 24 h p.i.. Compared with 64Cu-1 and 64Cu-2, 64Cu-3 exhibited lower liver uptake at both 2 h and 24 h p.i. with 56.7 ± 1.8 %ID/g and 47.3 ± 4.4 %ID/g, respectively. Unlike the liver uptake, 64Cu-3 showed the highest kidney uptake which was around 31.2 ± 3.6 %ID/g at 2 h while 14.4 ± 0.9 %ID/g and 12.55 ± 2.3 %ID/g for 64Cu-1 and 64Cu-2, respectively. At 24 h p.i., the activity of 64Cu-3 in the kidney remained at 19.3 ± 2.6 %ID/g, which was the highest level of all the conjugates tested in this study.
As for the peptide controls, both 64Cu-4 and 64Cu-5 showed lower liver uptake and faster liver clearances compared to all three peptoid conjugates (P < 0.05). 64Cu-4 has a direct sequence homology to 64Cu-1 and a similar hydrophobicity to 64Cu-2 (Table 1). Compared to both peptoids, 64Cu-4 showed lower liver uptake and higher kidney uptake at 2 h p.i. (Figure 4A), reminiscent of the different hydrodynamic radii coming from the lack of hydrogen-bond donors in the peptoid backbone. Then the kidney activity of 64Cu-4 decreases faster and becomes similar with that of 64Cu-2 at 24 h p.i. (Figure 4B), which can be explained by the higher in vivo stability of peptoids over peptides. Between the two peptides, less hydrophobic 64Cu-5 exhibited lower liver uptake than 64Cu-4 did. In addition, 64Cu-5 had a much lower spleen uptake than the other four conjugates tested at both 2 h and 24 h. All three peptoids and two peptides showed minimal brain uptake (lower than 1 %ID/g), suggesting their incapability of crossing the blood-brain-barrier.
The effects of different administration routes (oral, intravenous, and intraperitoneal) on the in vivo biodistribution were then examined using 64Cu-1 (Supporting Information, Figure S3). When administered intraperitoneally, predominent disposition of 64Cu-1 was in the abdomen region. After intraperitoneal injection, activity was detected in most abdomen organs including intestine, colon, stomach, liver, spleen, and pancreas, and the activity persisted through 24 hours. Significant kidney and thymus uptake was also observed after the intraperitoneal injection. When administered orally, the activity in stomach reached up to 4.82 ± 1.44 %ID/g at 2 h and then dropped to 0.54 ± 0.22 %ID/g at 24 h. The uptake of 64Cu- 1 could only be measured in the digestive system (i.e., stomach, small intestine, colon), suggesting that 64Cu-1 cannot be absorbed into blood circulations when administered orally.
Imaging studies were conducted at 0.25, 1, 2, 4, and 24 h after tail vein injection of 64Cu-1 – 64Cu-5. Decay-corrected coronal microPET images are shown in Figure 5. Three peptoids and two peptides displayed different in vivo distribution patterns; particularly, uptakes in liver and in kidney were noticeable. For 64Cu-1 and 64Cu-2, the main excretion route is likely hepatic; little kidney activity was observed. 64Cu-3 demonstrated the highest kidney activity of all three peptoids. Compared to peptoids, higher kidney activity and lower liver activity were observed for peptides. Among the tested conjugates, 64Cu-5 showed fastest clearance both in hepatic and renal pathways. Activities in the gastrointestinal (GI) tract could also be observed for all tested compounds, but markedly lower for 64Cu-1.
The distribution patterns of 64Cu-1 with different administration routes were compared and shown by PET imaging (Figure 6). Radioactivity was found throughout the whole abdomen cavity when 64Cu-1 was administered intraperitoneally. The permissive activity in abdomen clears rather slowly; therefore, it was complicated to differentiate organs in 24 hours based upon PET imaging. Interestingly, a symmetrical anatomical structure was distinctly visible inside the upper body of the mice after the intraperitoneal administration of 64Cu-1, and biopsy confirmed the symmetrical organ as thymus. Thymic uptake has been reported for several radioactive pharmaceuticals such as 18F-fluorodeoxyglucose (18F-FDG), radioiodine, and gallium-67 citrate; however, the mechanism through which it occurs is not completely understood.57, 58 When 64Cu-1 was administered through the tail vein, the compound essentially accumulated in the liver through 24 hours as described above. Very low GI activity and kidney activity were observed in 24 hours. In mice with orally administrated 64Cu-1, only GI activity was observed initially, and the peptoid conjugate was eliminated after 24 h as seen from PET imaging at 24 h p.i. A similar distribution pattern was observed for the oral administration of 64Cu-3 (Supporting Information, Figure S4).
Designed to mimic naturally occurring proteins and peptides, peptoids are interesting materials whose properties lie in between natural biopolymers and non-natural synthetic polymers. The structural similarity to peptides enabled peptoids to exhibit the impressive diversity of biological activity;28-30 however, the slight variation of backbone structure provides peptoids with distinct pharmacological properties such as resistance to proteolysis.21-23
Previously, a library of cationic amphipathic peptoids were synthesized based upon defined sequences and variations of secondary structures. It was demonstrated that the main chain length, helicity, charge, and hydrophobicity of peptoids could alter the in vitro antimicrobial activity.31, 32 Similarly, we envisioned that the modulation of peptoid structure could influence the in vivo pharmacokinetic profiles and biodistribution which are critical information to further develop the antimicrobial peptoids as therapeutic agents.
Previous in vivo data on biodistribution of peptoids are limited. Wang et al. used a 3H-labeled tripeptoid to test the absorption and disposition in rats and compared the labeled tripeptoid to a 3H-labeled tetrapeptide.59 The tripeptoid appeared to be more metabolically stable, but showed lower oral absorption and more rapid biliary excretion. Although the in vivo pharmacokinetic properties of peptoid were investigated, the peptoids tested by Wang et al. can be classified as a small molecule with molecular weight less than 500 Da. In our present study, we examined antimicrobial peptoids that have secondary structures and molecular weight higher than 1500 Da, which render the peptoids similarity to small proteins rather than to small molecules. Since distinct pharmacological properties between small and large molecules are reported,60 biodistribution study of the antimicrobial peptoids is important for further in vivo applications.
When designing radiolabeled conjugates, we chose 64Cu-DOTA labeling at the N-terminus of peptoids because we expected (1) more straightforward synthesis and (2) less impact on the peptoid secondary structure. The latter was confirmed by CD study. The comparison of CD spectra before and after N-terminal DOTA conjugation indicated nearly identical CD signatures: the intensity of spectral maxima at 190 nm and minima at 220 nm did not change much.32 If the labeling was made on the side chain, the impact on the peptoid conformation could be larger. CD spectra in Figure 3 clearly demonstrate minimal conformational change after Cu2+ binding. Although our observation proves neither DOTA conjugation nor metallation changed the CD signature of peptoid, this does not prove that the peptoid with complexed Cu2+ has the same pharmacokinetic characteristics. Nonetheless, the CD data is useful as a preliminary indicator of intact peptoid conformation after radiolabeling.
The hepatobiliary system and the kidneys are the major routes by which drugs and their metabolites are eliminated from the body. Generally, small and hydrophilic organic compounds are preferentially excreted via the kidneys. In contrast, compounds with relatively high molecular weight (>500 Da) and amphipathic structure are mainly excreted via the liver (and then into bile).60, 61 As large and amphipathic compounds, 64Cu-1 – 64Cu-5 all showed the major uptake and excretion occurred in the liver; however, different uptake profiles in the liver and in the kidney are observed for each conjugate by the combined effect of the structure, hydrophobicity, and resistance to proteolysis (Figure 4 and and55).
After intravenous administration, biodistribution studies of 64Cu-1 demonstrated highest liver accumulations at both 2 h and 24 h (Figure 4). A slow elimination from the liver region was observed by PET images (Figure 5). Although the secondary structures of 64Cu-1 and 64Cu-2 are very distinct, they showed similar in vivo pharmacokinetics based on biodistribution and microPET studies, suggesting their overall hydrophobicity and amphipathicity. At both 2 h and 24 h p.i., 64Cu-3 showed the highest level of kidney uptake among all three peptoids and two peptides tested (Figure 4). With only one shorter hydrophobic residue than 64Cu-1, 64Cu-3 showed more than 50% lower liver uptake and noticeably higher kidney uptake; and this result exemplify the strong influence of structural modification on in vivo biodistribution. Both biodistribution and PET results indicated faster clearance of peptide controls compared to peptoids (Figure 4 and and5).5). 64Cu-5 indicated lowest uptake in both the liver and the kidney and fastest elimination. Along with the hydrophilic property of 64Cu-5, the random coil structure, therefore vulnerability to proteolysis, can explain the result.
Different administration routes were investigated with 64Cu-1 and 64Cu-3. When administered orally, the peptoids showed poor bioavailability. The minimal absorption was indicated by high activity in the GI tract and little activity in the liver or in the kidney system (Supporting Information, Figure S3 and S4). For peptides, poor oral bioavailability has been widely accepted,62 and our result suggests the similarity between peptides and peptoids in terms of the poor absorption after oral administration. However, it should be noted that peptides usually have rapid elimination kinetics in the range of minutes due to the extensive hydrolysis in the GI tract by peptidases.63, 64 Therefore, the 24 h passage through the GI tract without rapid digestion constitutes a unique aspect of peptoids, which originates from the protease resistance and the property as a non-natural biopolymer. The high in vivo stability of peptoids can be advantageous for some applications. If the site of action of a designed peptoid is in the GI tract, such as peptoid-based bacterial toxin sequestrants developed by Kodadek et al.65 or antimicrobial peptoids for bacterial digestive infections, oral administration may provide a better therapeutic effect. Prolonged localized exposure to the pathogens and less side-effect are expected due to the longer retention in the GI tract and the poor absorption to the systemic circulation, respectively.
Our data indicated that peptoids, compared to peptides, had a general in vivo property of higher tissue accumulation and slower elimination as well as higher stability. These observations can be rationalized in two ways; membrane permeability and protease resistance. Studies have shown that the cellular uptake could be improved by directly translating peptides into peptoids due to the lipophilic nature and the stability of peptoids compared to those of peptides.24, 37, 66 In addition, peptoids used in this study are cationic and amphipathic, which are recently found to be effective in cell membrane penetration.44 As the first step for the tissue uptake is transport across the cell membrane (i.e., proximal tubular cells for the kidney uptake or hepatocyte cells for the liver uptake),60, 61 the efficient membrane permeability of the cationic amphipathic peptoids appears to play an important role in the higher tissue uptake. Secondly, the in vivo stability of peptoids against proteolysis provides longer circulation time, which leads to the higher tissue uptake and the slower elimination.
Unique in vivo properties of cationic amphipathic peptoids are demonstrated in our study. Applications of the peptoids, for example, as a long-lasting liver targeting carrier or as an antimicrobial therapeutic agent against digestive infections or against urinary tract infections, now seem to be reasonable.
In conclusion, our results suggest that the in vivo pharmacokinetics of peptoids can be tuned by varying the structure and modifying the physicochemical properties of peptoids. In comparison to peptides, peptoids exhibited higher in vivo stability, higher tissue uptake, and slower elimination rate. The in vivo biodistribution of peptoids can be confirmed by the real time small animal PET imaging. Using the noninvasive in vivo imaging approaches described here, pharmacokinetic properties of various peptoids can be predicted, which will provide useful information for the suitable application and development of the peptoids.
This work was supported by the DOD-PCRP-NIA (PC094646 to Z. C.), the National Institutes of Health (RO1 AI072666 to A. E. B.), and the GIST specialized research program (GIST-K02360 to J. S.).
Supporting Information Available: Supplementary figures and tables including HPLC, ESI-MS, CD spectra, biodistribution data, and PET images. This material is available free of charge via the Internet at http://pubs.acs.org.