SEPTEMBER 2008 Volume 19, Number 9 Copyright 2008 by the American Chemical Society
COMMUNICATIONS Dark Quenched Matrix Metalloproteinase Fluorogenic Probe for Imaging Osteoarthritis Development in ViWo Seulki Lee,†,§ Kyeongsoon Park,†,§ Seung-Young Lee,† Ju Hee Ryu,† Jong Woong Park,‡ Hyung Jun Ahn,† Ick Chan Kwon,† In-Chan Youn,† Kwangmeyung Kim,†,* and Kuiwon Choi†,* Biomedical Research Center, Korea Institute of Science and Technology, Seongbuk-Gu, Seoul, South Korea, Department of Orthopaedic Surgery, College of Medicine, Korea University, Danwon-Gu, Ansan, Gyeonggi, South Korea. Received June 30, 2008; Revised Manuscript Received July 24, 2008
The early detection of osteoarthritis (OA) is currently a key challenge in the field of rheumatology. Biochemical studies of OA have indicated that matrix metalloproteinase-13 (MMP-13) plays a central role in cartilage degradation. In this study, we describe the potential use of a dark-quenched fluorogenic MMP-13 probe to image MMP-13 in both in Vitro and rat models. The imaging technique involved using a MMP-13 peptide substrate, near-infrared (NIR) dye, and a NIR dark quencher. The results from this study demonstrate that the use of a dark-quenched fluorogenic probe allows for the visual detection of MMP-13 in Vitro and in OA-induced rat models. In particular, by targeting this OA biomarker, the symptoms of the early and late stages of OA can be readily monitored, imaged, and analyzed in a rapid and efficient fashion. We anticipate that this simple and highly efficient fluorogenic probe will assist in the clinical management of patients with OA, not only for early diagnosis but also to assess individual patient responses to new drug treatments.
Degenerative joint diseases including osteoarthritis (OA) are common, particularly in the elderly. General awareness of OA is low because this condition is typically considered to be an effect of aging. The initial stages of OA are clinically silent. Therefore, people may have the disease for years with no symptoms, and when symptoms do occur, the disease has often † Biomedical Research Center, Korea Institute of Science and Technology, Seongbuk-Gu, Seoul, South Korea. ‡ Department of Orthopaedic Surgery, College of Medicine, Korea University, Danwon-Gu, Ansan, Gyeonggi, South Korea. § These authors contributed equally to this work. * To whom correspondence should be addressed. Korea Institute of Science and Technology, 39-1 Hawolgok-dong, Seongbuk-gu, Seoul 136-791, Korea. Tel: +82-2-958-5912. Fax: +82-2-958-5909. E-mail:
[email protected] and
[email protected].
already reached an advanced stage. However, if it was possible to monitor and assess the disease states of OA, treatment could begin before significant damage occurs. For these reasons, early detection of OA is a key challenge in the field of rheumatology. The challenges posed in identifying OA are the detection of both early and subtle changes of joint and articular cartilage. Early signs of OA are characterized by the progressive loss of proteoglycan aggrecan, which is reflected by excessive damage to type II collagen. A decrease in proteoglycan aggrecan ultimately results in the loss of articular cartilage (1). X-rays and computed tomography (CT) can show large changes in a joint based on bone density; however, these techniques do not reveal detailed surface information (2). Although magnetic resonance imaging (MRI) (3) and ultrasound (US) (4) techniques recently provided both surface and internal details of articular
10.1021/bc800264z CCC: $40.75 2008 American Chemical Society Published on Web 08/26/2008
1744 Bioconjugate Chem., Vol. 19, No. 9, 2008
Communications
Figure 1. (A) Structure-based probe design. The molecular surface of hMMP-13 and the modeled binding pockets of the MMP-13 imaging probe 1. Arrow and italics indicate the cleavage site. (B) Quenching properties and recovery of NIR fluorescence intensities in various concentrations of 1 with or without the addition of activated MMP-13. Left, fluorescence imaging sections of 96-well microplate; right, recovery of fluorescence emission intensity (695 nm).
cartilage, the limited resolution of these techniques prevent them from being able to detect small differences in the early stages of OA. Recent progress in fluorescence-based optical imaging is paving the way for the development of novel methods in many biomedical areas, ranging from the detection of pathologies to biomolecular targets, i.e., biomarkers (5, 6). Biomarkers are molecules that reflect a specific pathological process or a response to therapeutic intervention. Biochemical studies in OA have indicated that there are various kinds of accessible biomarkers in OA, including matrix metalloproteinase (MMP), hyaluronic acid, osteocalcin, etc. (7). Therefore, the development of new technologies for the rapid, sensitive, and accurate optical imaging of specific biomarkers in OA would provide a method to diagnose OA early on and to assess the severity of the disease. One of the fundamental pathways in cartilage degradation is the up-regulation of expression of the MMP genes. Several lines of evidence have suggested that matrix metalloproteinase-13 (called MMP-13 or collagenase 3) promotes degradation of the cartilage extracellular matrix (8, 9). It was reported that MMP13 expression occurred at very low levels in normal cartilage but was particularly upregulated in degenerating cartilage (7). The presence of excess MMP-13 in OA offers the possibility for successful MMP-13 targeted OA imaging and also for detecting MMP-13 inhibition (10) for OA treatment. Conventional fluorogenic peptide kits for MMPs are available; however, they have had limited use in Vitro. We have previously reported on the use of target biomolecules that are activated by fluorescence probes based on a FRET mechanism and the selfquenching properties of NIRF. However, these targets provided insufficient resolution for in Vitro imaging (11-13). The limited resolution in this study was mainly attributed to a low signalto-noise (S/N) ratio. Therefore, high quenching efficiency and specific recognition properties by target biomolecules that have improved S/N ratio are essential for the development of sensitive
in Vitro biomolecular probes. In this study, we describe the potential use of MMP-13 activatable near-infrared fluorescence (NIRF) dark-quenched fluorogenic probe for early OA diagnosis in Vitro and in MMP-13 inhibitor screening. On the basis of the known MMP-13 substrate, PLGMRG (14), we first designed different probes using a combination of the linker peptide and two bulky chemicals, Cy5.5 and the black hole quencher-3 (BHQ-3) (15). By modeling the X-ray crystal structure of hMMP-13, 1 was synthesized using standard solidphase Fmoc peptide chemistry as an MMP-13 imaging probe (see Supporting Information, Figures 1S and 2S). As shown in Figure 1A, 1 is composed of (i) Cy5.5, which was used as the near-infrared (NIR) dye (ex/em; 675/695) and is adequate for in Vitro imaging since it emits fluorescence above 650 nm where autofluorescence is minimal, (ii) a short peptide sequence, GPLGMRGLGK, which was used as the MMP-13 cleavable substrate (the recognition site indicated by italics with the cleavage site between G and L; see SI for details), and (iii) (BHQ-3), which was used as the NIR dark-quencher specific for Cy5.5 (abs. 650 nm). As a control, a probe containing a noncleavable linker 2, which was composed of the scramble peptide sequence GRMGLPLGK was also prepared. There are two basic concepts of fluorogenic peptides when used for NIR imaging, which depends on the type of quencher used. In one case, the fluorescent pairs are composed of the dye-based donor and acceptor (fluorescence dye-dye quenching system), and in the second case, the acceptor is chemically different from the dye and does not fluoresce. Dark-quenched fluorogenic peptides have distinct advantages over conventional dye-dye quenched peptides. BHQ-3 is a dark quencher that has no native fluorescence and has maximal absorption in the 620 to 730 nm range, which can efficiently quenching fluorophores that emit in this range, i.e., Cy5.5. Therefore, Cy5.5 is spectrally matched with a BHQ-3, and FRET quenching can be maximized. The quenching properties of the probe were analyzed and visualized
Communications
Bioconjugate Chem., Vol. 19, No. 9, 2008 1745
Figure 2. MMP-13 specificity of 1 in solution. (A) Fluorescence emission kinetic spectra of 1 (80 pM) in the presence of various stimuli (MMP-13 and MMP-7, 11 nmol/L; MMP-13 inhibitor) after a 40 min incubation at 37 °C. Below, fluorescence image sections of the 96-well microplate containing different combinations. (B) Fluorescence emission spectra of 1 in the presence of various concentrations of MMP-13 after a 40 min incubation at 37 °C; inset, MMP-13 standard curve; below, corresponding fluorescence image sections of a 96-well microplate.
using a spectrofluorometer with a fixed excitation wavelength of 675 nm and a Kodak Image Station 4000MM equipped with filter for Cy5.5. As shown in Figure 1B, increased concentrations of 1 in the 96-wells microplate did not affect the background NIR fluorescence signal. To determine the effect of activated MMP-13 on the NIR fluorescent signal, 11 nmol/L MMP-13 was activated by incubation with 2.5 mmol/L of p-aminophenyl mercuric acid in the reaction buffer (100 mM of Tris, 5 mM of calcium chloride, 200 mM of NaCl, and 0.1% of Brij) for 1 h at 37 °C. MMP-13 treatment resulted in an increased recovery of the NIR fluorescence signal, which depended on probe concentration (approximately 9, 22, 26, 37, 48-fold for 1.6, 16, 32, 80, 160 pM of the probe vs without MMP-13). The use of cleavable, dark-quenched NIR fluorescent probes resulted in significantly lower fluorescent backgrounds, which were amplified by the addition of MMP-13, thereby resulting in a highly sensitive visual detection method. The selectivity of the MMP-13 probe was examined in Vitro by incubating 1 (80 pM) in a cuvette containing the reaction buffer and 11 nmol/L activated MMP-13, MMP-7, MMP-13 with the MMP-13 inhibitor (pyrimidine-4,6-dicarboxylic acid, bis-(4-fluoro-3-methyl-benzylamide)) (16), or MMP-13 with 1. After incubation in the reaction buffer at 37 °C for 40 min with the respective enzymes and inhibitor, the NIR fluorescence emission signals of the samples were measured using a spectrofluorometer. Simultaneously, each sample was moved into a 96-well microplate and imaged using a Kodak Image Station 4000MM. As shown in Figure 2A, the kinetics of the enzyme was evaluated and visualized through fluorescence images. The spectrofluometry plots clearly demonstrate that significant recovery of the NIR fluorescent signals only occurred in the samples containing MMP-13 (approximately 32-fold over that without MMP-13). The fluorescent signal recovery of 1 was inhibited in the presence of the MMP-13 inhibitor, thus providing clear evidence for the selectivity of 1. Measurements were then performed under the same experimental conditions at different concentrations of activated MMP-13 (0, 0.55, 1.1, 2.2, 5.5, and 11 nmol/L; Figure 2B) in the presence of a fixed concentration of 1 (80 pM). In these experiments, there was a linear relationship between the recovered NIR fluorescent signals and MMP-13 concentrations up to 5.5 nmol/L (r2 ) 0.995) when 1 was used. This was further confirmed by the fluorescence
images. Taken together, these results indicate that 1 could be used for quantitative and visual analysis of MMP-13 activity. After the systematic validation of the utility of 1, in Vitro, its use in ViVo was demonstrated using in ViVo fluorescence tomography (16) (Kodak Image Station and eXplore Optix; see Supporting Information) in an experimental rat model of intermediate (six week OA-induced rats) and late-stage OA (eight week OA-induced rats; see Supporting Information) (17). Representative images are depicted in Figure 3. NIR fluorescence signals were monitored using the Kodak Image Station 1 h after intracartilage injection of 1 (10 µg/100 µL saline/ cartilage, 4 nM) into OA-induced and normal cartilage rats. When 1 was injected into the normal side, NIR fluorescence signal intensity did not substantially increase. In contrast, 1 in OA-induced cartilage produced strong NIR fluorescence signals that were clearly visualized in the fluorescent images (1.3 ( 0.2, 3.3 ( 0.7, and 7.4 ( 1.4-fold vs normal side for 0, 6, and 8 weeks, respectively; n ) 3). NIR fluorescence signals in the eight-week OA rat had a 2.2-fold increase relative to the sixweek OA rat. Furthermore, the histological section of the normal cartilage showed a smooth articular surface with clear demarcation between the cartilage and subchondral bone. In contrast to the control, the images of the six and eight week OA-induced rats show a distinct erosion of the cartilage and a significant alteration in the bone-cartilage interface. These results confirm the feasibility of using 1 for the detection and visualization of upregulated MMP13 in relation to determining the stage of OA development in ViVo. For quantitative image analysis of the upregulation of MMP13 in the OA rat model, similar in ViVo imaging of OA was conducted 1 h after the injection of 1 with or without the MMP13 inhibitor in the eight week OA rat model using a preclinical optical imaging system, eXplore Optix (ART Advanced Research Technologies Inc., Montreal, Canada) configured for NIR fluorescence detection (see Supporting Information). Figure 4A shows typical NIR fluorescence images of normal and OA cartilage obtained during the 1 h time period after the intracartilage injection of 1. The total photon counts in normal and OA cartilage, which were quantified using Analysis Workstation Software (ART Advanced Research Technologies Inc.), were (3.6 ( 0.2) × 106 and (11.5 ( 1.73) × 106, respectively, and the NIR fluorescence intensity ratio between OA and normal
1746 Bioconjugate Chem., Vol. 19, No. 9, 2008
Communications
Figure 3. In ViVo imaging of upregulated MMP-13 in normal, six, and eight week OA-induced cartilages 1 h after intracartilage-injection of 1 (4 nM). Left: NIR fluorescence reflectance imaging of normal and OA cartilage after local injection of 1. Right: Histological evaluation of normal, six, and eight week OA joints by Safranin-O staining. Arrows: dotted line (normal) and solid line (OA).
cartilage (O/N) was approximately 3.2-fold (p < 0.001) (Figure 4B). However, fluorescence contrast was significantly reduced ((4.7 ( 0.8) × 106) when the MMP-13 inhibitor was administered 30 min before the injection of 1. To analyze the depth profiles from the fluorescence images, the three-dimensional (3D) images were constructed by examining the intensity in slices cut along the Z-axis, which represented the twodimensional (2D) slices, starting at Z ) 2 to 7 mm deep (Figure 4C). The images from the OA-induced rats displayed a strong fluorescence signal between the depths of 3 to 5 mm, whereas normal and inhibitor treated cartilage displayed significantly low signals. Displaying the images in 2D allowed for the visualization of the expression and distribution of MMP-13 in OA cartilage. This may be useful for the longitudinal monitoring of the OA response in diagnosis and therapy. Overall, these data show that 1 is activated by MMP-13 and could produce strong NIR fluorescence signals, which enables OA imaging in ViVo. In summary, this study clearly demonstrates the potential use of the dark-quenched fluorogenic MMP-13 probe for the facile visual detection of MMP-13 in Vitro and in ViVo. In particular, by targeting a specific OA biomarker, the symptoms and
development stages of OA can be readily and easily monitored, imaged, and analyzed in a rapid and efficient fashion. In addition, this technique can be widely applied to any target protease-based disease diagnosis and high-throughput drug screening simply by changing the specific peptide-substrate linker between the fluorophores and quenchers. Recently, the development of optical imaging instruments and sophisticated imaging probes have allowed researchers to detect and monitor molecular targets in live cells and in in ViVo in real-time. However, because of the depth limitation of current optical imaging instruments, these techniques are only applicable to small animal models and superficial human tissues. Therefore, in order to translate dark-quenched imaging probes from preclinical studies to patients, the problems associated with interrogating fluorescence in deep tissue need to be addressed. We can envision improvements in imaging technologies that will allow optical probes to be used in larger subjects, and we anticipate that this simple and highly efficient fluorogenic probe will assist in the clinical management of patients with OA in the near future, not only for early diagnosis but also for assessing the response to treatment.
Communications
Bioconjugate Chem., Vol. 19, No. 9, 2008 1747
Figure 4. (A) In ViVo NIR fluorescence tomographic images of normal and OA cartilage after local injection of 1 with or without the addition of the MMP-13 inhibitor in eight week OA-induced rat model. Left: 1 without MMP-13 inhibitor. Right: 1 with inhibitor in OA-induced cartilage. (B) Total photon counts in cartilage calculated from A. (C) 2D slices of the image from A and reconstruction in the Z direction. (a) Normal; (b) without or (c) with the addition of MMP-13 inhibitor. Arrows: dotted line (normal) and solid line (OA).
ACKNOWLEDGMENT This work was supported by the Seoul R&BD Program and the GLR and Real-Time Molecular Imaging Project of MEST, and Intramural Research Program of the KIST. Supporting Information Available: Synthetic, physicochemical properties, and experimental details. This material is available free of charge via the Internet at http://pubs.acs.org.
LITERATURE CITED (1) Poole, A. R., Alini, M., and Hollander A. (1995) Chondrocytes and Cartilage Destruction. Mechanisms and Models in Rheumatoid Arthritis (Henderson, B., Edwards, J., and Pettipher, R., Eds.) pp 163-204, Academic Press, San Diego, CA. (2) Blackburn, W. D., Chivers, S., and Bernreuter, W. (1996) Cartilage imaging in osteoarthritis. Semin. Arthritis Rheum. 25, 273–281. (3) Eckstein, F., Cicuttini, F., Raynauld, J. P., Waterton, J. C., and Peterfy, C. (2006) Magnetic resonance imaging (MRI) of articular cartilage in knee osteoarthritis (OA): morphological assessment. Osteoarthritis Cartilage 14, A46-A75. (4) Meenagh, G., Filippucci, E., Iagnocco, A., Delle, S. A., Riente, L., Bombardieri, S., Valesini, G., and Grassi, W. (2007) Ultrasound imaging for the rheumatologist VII. Ultrasound imaging in rheumatoid arthritis. Clin. Exp. Rheumatol. 25, 5– 10. (5) Weissleder, R. (2001) A clearer vision for in vivo imaging. Nat. Biotechnol. 4, 316–317. (6) Ntziachristos, V. (2006) Fluorescence molecular imaging. Annu. ReV. Biomed. Eng. 8, 1–33. (7) Rousseau, J. C., and Delmas, P. D. (2007) Biological markers in osteoarthritis. Nat. Clin. Pract. Rheumatol. 3, 346–356. (8) Neuhold, L. A., Killar, L., Zhao, W., Sung, M. L., Warner, L., Kulik, J., Turner, J., Wu, W., Billinghurst, C., Meijers, T., Poole, A. R., Babij, P., and DeGennaro, L. J. (2001) Postnatal expression in hyaline cartilage of constitutively active human collagenase-3
(MMP-13) induces osteoarthritis in mice. J. Clin. InVest. 107, 35–44. (9) Bau, B., Gebhard, P. M., Haag, J., Knorr, T., Bartnik, E., and Aigner, T. (2002) Relative messenger RNA expression profiling of collagenases and aggrecanases in human articular chondrocytes in vivo and in vitro. Arthritis Rheum. 46, 2648–2657. (10) Close, D. R. (2001) Matrix metalloproteinase inhibitors in rheumatic diseases. Ann. Rheum. Dis. 60 (Suppl 3), iii62-67. (11) Kim, K., Lee, M., Park, H., Kim, J. H., Kim, S., Chung, H., Choi, K., Kim, I. S., Seong, B. L., and Kwon, I. C. (2006) Cellpermeable and biocompatible polymeric nanoparticles for apoptosis imaging. J. Am. Chem. Soc. 128, 3490–3491. (12) Kim, J. H., Lee, S., Kim, K., Jeon, H., Park, R. W., Kim, I. S., Choi, I., and Kwon, I. C. (2007) Polymeric nanoparticles for protein kinase activity. Chem. Comm. 13, 1346–1348. (13) Kim, J. H., Lee, S., Park, K., Nam, H. Y., Jang, S. Y., Young, I., Kim, K., Jeon, H., Park, R. W., Kim, I. S., Choi, K., and Kwon, I. C. (2007) Protein-phosphorylation-responsive polymeric nanoparticles for imaging protein kinase activities in single living cells. Angew. Chem., Int. Ed. 46, 5779–5782. (14) Deng, S. J., Bickett, D. M., Mitchell, J. L., Lambert, M. H., Blackburn, R. K., Carter 3rd, H. L., Neugebauer, J., Pahel, G., Weiner, M. P., and Moss, M. L. (2000) Substrate specificity of human collagenase 3 assessed using a phage-displayed peptide library. J. Biol. Chem. 275, 31422–31427. (15) Johansson, M. K., and Cook, R. M. (2003) Intramolecular dimers: a new design strategy for fluorescence-quenched probes. Chemistry 9, 3466–3471. (16) Park, K., Kim, J. H., Nam, Y. S., Lee, S., Nam, H. Y., Kim, K., Park, J. H., Kim, I. S., Choi, K., Kim, S. Y., and Kwon, I. C. (2007) J. Controlled Release 122, 305–314. (17) Engel, C. K., Pirard, B., Schimanski, S., Kirsch, R., Habermann, J., Klingler, O., Schlotte, V., Weithmann, K. U., and Wendt, K. U. (2005) Structural basis for the highly selective inhibition of MMP-13. Chem. Biol. 12, 181–191. BC800264Z
