DOI: 10.1002/cmdc.200800205 Enzymatic Release of a Surface-Adsorbed RGD Therapeutic from a Cleavable Peptide Anchor Steven R. Meyers, [b] Daniel J. Kenan,* [a] and Mark W. Grinstaff* [b] Implantation of a polymeric, ceramic, or metallic implant will unavoidably activate a response from the surrounding tissue. [1] Means to attenuate this inflammatory response at the implant site are of significant interest and currently include strategies based on surface morphology, chemical modification, or drug delivery. This response is particularly evident at the site of stent deployment, where the overproliferation of smooth muscle cells can lead to restenosis—a re-narrowing of the lumen. [2] Drug-eluting stents (DES) were introduced to de- crease the rate and severity of this neointimal formation through passive diffusion of a drug physically entrapped in a nondegradable polymer coating over a metal framework. How- ever, recent studies have expressed concern over the wide- spread use of DES owing to their increased late-thrombotic po- tential of two to three times the rates for a traditional bare metal stent. [3] This clinical outcome is likely due to delayed healing and endothelium regeneration as a result of the poly- mer coating (e.g., poly(styrene-b-isobutyl-b-styrene)) and the delivery of non-phenotype-specific antimitotic/antiproliferative drugs (e.g., sirolimus and paclitaxel). With the goal of improv- ing implant performance through appropriate interactions with the surrounding biology, we previously reported the use of implant-specific peptide coatings to prevent nonspecific sur- face biofouling and to promote a pro-healing response through increasing cell adhesion and spreading. [4] Herein we report a third approach whereby a surface-adsorbed therapeu- tic is enzymatically released, resulting in drug elution (Figure 1). Engineering of an enzymatic recognition site into a material is an elegant approach to promote active degradation and has been used successfully with hydrogels, microspheres, bioplex- es, and interpenetrating networks, [5] as well as for evaluating enzyme kinetics in the degradation of peptides on surfaces. [6] The enzymatic release of an adsorbed or tethered therapeutic from an implant surface is an exciting idea which would likely be of interest for many medical devices, including stents. Cur- rent stenting applications rely on passive drug entrapment and diffusion, and a wide variety of therapeutics are under in- vestigation. [7] Some of these low-molecular-weight therapeu- tics include dexamethasone, methylprednisolone, 17-b-estra- diol, angiopeptin, paclitaxel, actinomycin D, sirolimus, and argi- nine-glycine-aspartic acid (RGD). [8] The last example is particu- larly interesting, as clinical trials have shown that elution or local delivery of RGD decreased neointimal hyperplasia through the recruitment of circulating endothelial progenitor cells to the site of implantation and promoted arterial re-endo- thelialization. [8h, 9] Building upon these observations, we de- signed a peptide-based coating that consists of three distinct peptide domains: an implant-adsorptive sequence, an enzy- matically cleavable recognition site, and a therapeutic to be delivered (i.e., RGD). Medical devices such as stents coated with this peptide could then be implanted in vivo and remain stable and non-eluting against endogenous enzymes until sys- temic injection of the selected exogenous enzyme to catalyze the controlled local delivery of the therapeutic. We selected the polystyrene (PS) binding sequence H 2 N- FFSFFFPASAWGS-CO 2 H, previously identified through phage display, as PS serves as a model polymeric substrate. [4a] To select an appropriate enzyme for use with this proof-of-con- cept, we exposed a solution of FFSFFFPASAWGSSGSSGK- (biotin) 1 to varied concentrations of trypsin, papain, chymo- trypsin, enterokinase, thrombin, and factor Xa to determine the stability of the base sequence using both ELISA and MALDI- ToF MS. As shown in Figure 2, papain and chymotrypsin signifi- cantly cleave the base sequence; trypsin, thrombin, and factor Xa have a more moderate effect, and enterokinase did not show any activity. No additional degradation was observed after continued incubation with enterokinase for up to one week. We thus synthesized a cleavable peptide with an internal enterokinase recognition sequence (DDDDK) terminated with an RGD trimer (Table 1). The resulting peptide, FFSFFFPA- SAWGSSSGDDDDKSSGK-(biotin)-RGD 2 showed good stability in solution with no evidence of hydrolysis for at least one month. To ensure the resultant peptide continued to express affinity for PS after inclusion of the hydrophilic DDDDK sequence, Figure 1. A drug elution mechanism whereby a tethered therapeutic can be actively released from a substrate, such as a stent, upon addition of an enzyme. [a] Dr. D. J. Kenan Department of Pathology, Duke University Medical Center Box 3712, Durham, NC 27710 (USA) Fax: (+ 1) 919-684-9825 E-mail : kenan001@mc.duke.edu [b] Dr. S. R. Meyers, Dr. M. W. Grinstaff Departments of Biomedical Engineering and Chemistry Boston University, 590 Commonwealth Ave., Boston, MA 02215 (USA) Fax: (+ 1) 617-358-3186 E-mail : mgrin@bu.edu Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.200800205. ChemMedChem 2008, 3, 1645 – 1648  2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1645