99mTc-MAG3-Aptamer for Imaging Human Tumors Associated with

Article pubs.acs.org/bc

99m

Tc-MAG3-Aptamer for Imaging Human Tumors Associated with High Level of Matrix Metalloprotease‑9 Sonia Da Rocha Gomes,†,‡,▽,⬢ Julie Miguel,■,▽ Laurent Azéma,†,‡ Sandrine Eimer,‡,# Colette Ries,‡,§ Eric Dausse,†,‡ Hugues Loiseau,‡,⊥ Michèle Allard,‡,§,∥,¶ and Jean-Jacques Toulmé*,†,‡ †

INSERM U869, ARNA, Institut Européen de Chimie et Biologie, 33607 Pessac, France Université de Bordeaux, 33076 Bordeaux, France § CNRS, INCIA, UMR 5287, 33400 Talence, France ∥ Pôle d’imagerie, ⊥Département de Neurochirurgie, #Département de Pathologie, CHU de Bordeaux, 33076 Bordeaux, France ¶ Ecole Pratique des Hautes Etudes, Bordeaux, France ■ Pôle des Produits de Santé, CHU de Bordeaux, Pessac, France ‡

S Supporting Information *

ABSTRACT: The human matrix metalloprotease 9 (hMMP9) is involved in many physiological processes such as tissue remodeling. Its overexpression in tumors promotes the release of cancer cells thus contributing to tumor metastasis. It is a relevant marker of malignant tumors. We selected an RNA aptamer containing 2′-fluoro, pyrimidine ribonucleosides, that exhibits a strong affinity for hMMP-9 (Kd = 20 nM) and that discriminates other human MMPs: no binding was detected to either hMMP-2 or -7. Investigating the binding properties of different MMP-9 aptamer variants by surface plasmon resonance allowed the determination of recognition elements. As a result, a truncated aptamer, 36 nucleotides long, was made fully resistant to nuclease following the substitution of every purine ribonucleoside residue by 2′-O-methyl analogues and was conjugated to S-acetylmercaptoacetyltriglycine for imaging purposes. The resulting modified aptamer retained the binding properties of the originally selected sequence. Following 99mTc labeling, this aptamer was used for ex vivo imaging slices of human brain tumors. We were able to specifically detect the presence of hMMP-9 in such tissues.

■

fluorothymidine (FLT), [11C]-methionine or O-2-[18F]-fluoroethyl-tyrosine (FET). New imaging modalities are required for a better monitoring of tumor malignancy associated with extracellular damage in surrounding brain tissue, especially functional imaging reflecting intimate biological mechanisms of tumor cell proliferation. One of the possible mechanisms involved in surrounding tissue invasion is the overexpression of matrix metalloproteinases (MMPs) capable of degrading extracellular matrix components, permitting cell migration.4,5 An alternative approach for imaging tumors would be to follow MMPs expression as a surrogate marker of malignant tumor cell invasion within brain parenchyma.6 Tumor cells form mass lesions in the central nervous system, and enzymatic degradation of extracellular matrix by MMPs is necessary for the malignant tumor cells to migrate into normal brain tissue. Indeed, MMP inhibitors constitute attractive potential anti-

INTRODUCTION Primary brain tumors can be classified in various ways: topographic (i.e., extraparenchymal or intraparenchymal), neuropathologic (according to cell type),1 and prognostic (benign, unpredictable behavior or malignant). For the latter, few gliomas are benign tumors (pilocytic astrocytoma), but the vast majority of gliomas are malignant and infiltrative tumors and present poor treatment results (grade 3 or anaplastic astrocytoma and grade 4 or glioblastoma). Unfortunately, lower-grade gliomas can transform into glioblastoma.2 The transformation to a more malignant phenotype involves a multistep process including tissue invasion, resulting from tumor cell ability to migrate in brain parenchyma.3 Uncontrolled cellular proliferation is a cardinal feature of neoplasia. The ability to measure the proliferation rate in tumors in patients in vivo will help with tumor grading and staging, and assessing the effect of therapy. Today, modern imaging techniques enable the investigation of pathological changes within tumors themselves in vivo, exploiting the increased cellular metabolism in these cells using positon emission tomography (PET) imaging of 3′-deoxy-3′-[18F]© 2012 American Chemical Society

Received: March 21, 2012 Revised: October 8, 2012 Published: October 8, 2012 2192

dx.doi.org/10.1021/bc300146c | Bioconjugate Chem. 2012, 23, 2192−2200

Bioconjugate Chemistry

Article

cancer agents.7,8 hMMP-2 and hMMP-9 constitute a subgroup of MMPs called gelatinases that degrade the basal lamina around capillaries, and enable angiogenesis and neurogenesis, participating in extracellular matrix degradation and facilitating tumor cells migration in brain parenchyma.9 Because changes in extracellular matrix surrounding a brain tumor can occur before a change in magnetic resonance imaging (MRI) data of the tumor itself, it can be postulated that hMMP-9 PET or SPECT (single photon emission computed tomography) molecular imaging would be an accurate means for monitoring therapy in grade 2 glioma or diffuse astrocytoma. Aptamers were developed in the early 1990s.10−12 These structured DNA, RNA, or modified oligonucleotides are identified after iterative cycles of selection/amplification through a process named SELEX (systematic evolution of ligands by exponential enrichment) from a random oligonucleotide library. Aptamers were successfully selected for a wide range of targets (proteins, nucleic acids, peptides, small molecules, cells, etc.) and were shown to display both high affinity and specificity.13,14 Aptamer-based tools were designed for diagnostic or therapeutic applications over the past decade and are a promising alternative to monoclonal antibodies in many applications15,16 including molecular imaging.17 Aptamers can be modified to make them resistant to nucleases and conjugated to fluorescent tags or radioelements. The first aptamer for in vivo imaging was developed in 1997 for the detection of human neutrophil elastase in a rat model of inflammation.18 Since these encouraging results, aptamers have been successfully applied to target tumor cells for detection or real-time imaging.19−26 Most aptamer imaging probes have been selected against cells for cancer detection particularly with aptamer-conjugated nanoparticles.22,27−29 Recently, an antibody-like nanostructure composed of two aptamers and a dendrimer was developed with temperature-dependent binding to cancer cells.30 Histological analyses have been carried out with fluorescent or biotinylated aptamers.31−35 A new strategy based on activable aptamer showed less fluorescence background with specific tumor retention.36 In this work, a specific aptamer directed against the human MMP-9 protein has been identified and converted into a 99mTclabeled hMMP-9 probe following truncation, backbone modification, and end functionalization. We characterized the binding properties of this aptamer and evaluated its potential for in vitro imaging brain tumors by assessing the intensity of radiotracer labeling and detectability of various primary tumor slices and compared these with immunohistochemistry analysis.

The in vitro selection against hMMP-9 protein (Calbiochem) was performed at 23 °C in SP buffer (50 mM Tris HCl, pH 7.4, 50 mM NaCl, 100 mM KCl, 5 mM CaCl2, 1 mM magnesium acetate) using the filter retention technique (HAWP 0.45 μM, Millipore).38 Filters were pretreated with alkali as described by McEntee et al. 39 in order to reduce nonspecific adsorption of nucleic acids. The library was first incubated with the alkali-treated filters for 20 min then with hMMP-9 protein for 20 min. The mixture was filtered and filters washed with SP buffer. Candidates bound to the protein were eluted with 500 μL phenol/urea 7 M for 20 min at 65 °C and reverse transcribed with 200 U M-MLV reverse transcriptase RNase H− Point Mutant (Promega): 1 μL was used for 25 cycles of PCR at 63 °C with 1 U of AmpliTaq Gold DNA polymerase (Applied Biosystems) and the two P20 and 3′SL primers. 2′-F-RNA candidates were obtained by in vitro transcription of the PCR products with the DuraScribe T7 transcription kit (Epicenter Technologies). During the successive in vitro selection rounds, candidates and protein concentrations were progressively decreased, whereas the number of washes, used to eliminate weak binders, was increased. This resulted in a tougher competition between the candidates for binding as evolution proceeded. Monitoring the evolution of the binding properties of the selected population after every cycle indicated an increase in binding efficiency of the candidates up to round 15. After 14 selection cycles against the hMMP-9 protein, selected candidates were cloned using the TOPO TA cloning kit (Invitrogen) and sequenced (Genome Express Company). Oligonucleotide Synthesis. The truncated variants F3B, F3C1, and F3C2 of the aptamer F3, the 2′-O-methyl purine/2′fluoropyrimidine F3B (F3Bomf) derivative, the 3′-end biotinylated aptamers (F3B, F3Bomf), the control sequence 5′ UGCCAAACGCGUCCCCUUUGCCCGGCCUCCGCCGCA 3′ and the mutants F3BΔ and F3BA were chemically synthesized on an Expedite 8909 in our laboratory according to standard procedures. All oligonucleotides were purified by electrophoresis on denaturing 20% polyacrylamide, 7 M urea gels. Secondary structure prediction of aptamers was determined using the mfold web server (http://mfold.rna. albany.edu/?q=mfold). NMR Analysis of F3B Variant. 1H NMR spectra were recorded at pH 6.4 and 5.5, in 10 mM sodium phosphate buffer containing 90/10 H2O/D2O. Imino protons were assigned based on the analysis of NOESY spectra recorded at 4 and 15 °C. Oligonucleotide Conjugation. Oligonucleotides F3Bomf and the control sequence, bearing a 5′ hexylamino function, were synthesized on a 1 μmol scale with an ABI Expedite 8909 synthesizer, using conventional β-cyanoethyl phosphoramidite chemistry. Once purified (HPLC, Macherey-Nagel Nucleodur column, 0.1 M triethylammonium acetate, pH 7.0, (acetonitrile/0.1 M triethylammonium acetate, pH 7.0: 80/20) gradient), they were conjugated to MAG3.40 Briefly, 20 nmol of oligonucleotide were suspended in 100 μL of binding buffer (sodium bicarbonate/sodium carbonate 0.25 M, pH 8.3, sodium chloride 1 M, sodium ethylenediaminetetraacetate 1 mM) and gently stirred at room temperature. MAG3-NHS (3 mg, in 30 μL of DMF) was added in portions at room temperature over 3 h. After complete addition, the suspension was stirred for an additional hour, and the crude was directly purified by HPLC under the same conditions to afford the oligonucleotide−MAG3 conjugate in 50−90% yield. Conjugate

■

MATERIALS AND METHODS Materials. Aptamer Production and Analysis. SELEX Conditions. The oligonucleotide library was obtained by transcription from a DNA library, synthesized by Proligo, containing 30 random nucleotides (N30) flanked by invariant primer annealing sites: 5′-GTGTGACCGACCGTGGTGCN30-GCAGTGAAGGCTGGTAACC-3′. Two different primers P20 5′GTGTGACCGACCGTGGTGC and 3′SL 5′TAATACGACTCACTATAGGTTACCAGCCTTCACTGC containing the T7 transcription promoter (underlined), were used for PCR amplification. The modified 2′-fluoropyrimidine RNA library used for the selection and aptamer F3 were obtained by transcription (DuraScribe T7 transcription kit from Epicenter Technologies containing 2′-F-CTP and 2′-F-UTP). The mutant T7 RNA polymerase Y639F37 was also used. 2193



Article

indicating good stability. Average specific activities of 2.48 MBq/μg were obtained (SD = 10%). Tissue Samples and in Vitro Binding Assay. Tumor tissues used in these studies were obtained from the department of Pathology, University Hospital Bordeaux, France. Nine different types of well-characterized tumors: pilocytic astrocytoma grade 1, meningioma grade 1, fibrillary astrocytoma grade 2, ependymoma grade 2, anaplastic astrocytoma grade 3, medulloblastoma grade 4, primitive central nervous system lymphoma grade 4, and glioblastoma grade 4 were collected from surgical samples, including one case of normal brain tissue from a patient undergoing autopsy. Tumor grade was done according to the 2007 WHO classification of tumors of the central nervous system.1 All tissue samples were formalin-fixed and paraffin-embedded. Representative 2.5-μm-thick tissue sections were obtained from blocks of paraffin-embedded tissue and subjected to immunohistochemistry and autoradiography analysis. Binding studies were performed using these tissue sections incubated in the presence of either 99mTc-MAG3-F3Bomf aptamer or 99mTc-MAG3-control sequence according to the following procedure: after deparaffinization and rehydration, tissue slices were incubated with 0.037 MBq (0.00125 nmol) of 99m Tc -MAG3-F3Bomf and adjacent section with 99mTc -MAG3-control for one hour in a humidified chamber at room temperature before being washed twice in PBS + 0.1% Tween and then twice in purified water. Then, sections were imaged using a Beta Imager 2000 (Biospace Mesures, Paris, France). Immunostaining with hMMP-9 Antibodies. Immunohistochemical hMMP-9 detection was performed on serial 2.5μm-thick sections, using a purified antimouse hMMP-9 monoclonal antibody (ab58803, Abcam). Immunohistochemical procedures were carried out with a DAKO Envision Peroxidase System (DAKO Diagnostica) according to the following protocol: paraffin-embedded sections were deparaffinized with xylene, dehydrated through a graded alcohol series, and washed with distilled water. They were then treated with 0.3% hydrogen peroxydase for 5 min to block endogenous peroxidase activity. After washing with PBS, the slides were incubated for 30 min with hMMP-9 antibody diluted 1:150 in a humidified chamber at room temperature and then washed twice in PBS. En Vision multilink was then applied as the secondary antibody for 30 min before washing and incubation with diaminobenzydine DAB substrate for 10 min, and hematoxylin counterstaining. Appropriate positive and negative controls omitting the primary antibody were included with every slide run. Immunoreactivity was evaluated in the cell cytoplasm, cytoplasmic membrane, and in the extracellular matrix.

characterization was performed with a MALDI-ToF mass spectrometer (Reflex III, Bruker). Human Matrix Metalloprotease-9. The human MMP-9 was purchased to Calbiochem; samples were checked for purity by SDS polyacrylamide gel electrophoresis. Batch-to-batch variation was noticed resulting in the presence of breakdown fragments likely related to self-cleavage of the protease. Only samples with low fragment content were used in our study. Binding and Specificity Assays. The dissociation constant (Kd) of the complexes, formed by the aptamers and the hMMP9 protein, was determined using the filter retention method. One nanomole 32P 5′-end-labeled aptamer was incubated with increasing concentrations of hMMP-9 (10, 20, 40, 72, 160, 320, 500 nM) for 20 min at 23 °C in 20 μL SP buffer. Complexes were filtered and the radioactivity retained on the filter was quantified using a scintillation counter (LS 6000 IC, Beckman). Kd values were deduced from data point fitting with Kaleidagraph 3.0 (Abelbeck software), according to the equation: B = (Bmax[L]0)/([L]0 + Kd), where B is the proportion of complex, Bmax is the maximum of complex formed, and [L]0 is the total concentration of unlabeled ligand. Surface plasmon resonance (SPR) experiments were performed on a BIAcore 3000 apparatus (Biacore AB, Sweden). Two micrograms of hMMP-9 protein was injected on a carboxymethylated dextran CM5 sensorchip for immobilization, and aptamers F3B, F3C1, and F3C2 were injected at 200 nM (20 μL/min) in SP buffer. Alternatively, CM5 sensorchips were functionalized with streptavidin. 3′-End biotinylated F3B and F3Bomf were immobilized on the functionalized CM5, and hMMP-9 protein was injected at 100 nM in SP buffer. hMMP-9 was injected at different concentrations (from 5 to 160 nM) in PBS buffer. In another series of experiments, 3′-end biotinylated aptamer F3Bomf was immobilized, and hMMP-9, pro-hMMP-9, mouse pro-MMP-9, human MMP-2, or MMP-7 proteins were injected at 50 nM in PBS buffer. SPR experiments were performed at 23 °C, at 20 μL/min, and the complexes were dissociated with a pulse of a solution containing 40% formamide/3.6 M urea/30 mM EDTA.41 99m Tc Oligonucleotides Radiolabeling. The MAG3 F3Bomf and the control sequence were labeled with 99mTc as described by Winnard et al.:40 two fresh solutions were prepared: (i) sodium tartrate (50 mg/mL) in sterile 0.5 M sodium bicarbonate, 0.25 M ammonium acetate, 0.18 M ammonium hydroxide, pH 9.2 (The high pH of the tartrate solution was necessary so that the final pH is approximately 7.6) and (ii) a 1 mg/mL SnCl2·2H2O in 10 mM hydrochloric acid just prior to use. 99mTc pertechnetate solution (2−10 μL) (Elumatic III − Cis Bio International) was added to the MAG3-aptamer (10−100 μg) to provide about 3.7 MBq/μg of aptamer followed by the addition of the tartrate solution to a final concentration of 6−7 μg/μL. The stannous ion solution was added immediately thereafter (1 μg of SnCl2·2H2O for each 10 μg of aptamer) and left at room temperature for 15 min. The labeled aptamer was then purified by microspin column (MicroSpin G-25 columns, GE Healthcare), and the radiochemical purity was determined using thin-layer chromatography (TLC plates RP-18, Merck). Under the above set of conditions, average labeling efficiency of 70% (N = 15, SD = 14%) was achieved. Radiochemical purity (RCP) determined by thin layer chromatography was 77% (SD = 8%). The stability of 99mTc-MAG3-aptamer over time was determined using TLC. The radiolabeled oligonucleotides RCP was about 70% at 6 h following radiolabeling,

■

RESULTS AND DISCUSSION Characteristics and Binding Properties of hMMP-9 Aptamers. The SELEX strategy has been carried out against the recombinant human MMP-9 protein (gelatinase B), using a library of RNA candidates containing 2′-fluoropyrimidine nucleoside residues, as described in Material and Methods. The 30 nt random window is flanked by fixed regions that display partial complementarity, generating hairpin-like candidates through the formation of a weak duplex between the 5′ and 3′ parts (Figure 1A). This is expected to limit the contribution of the fixed parts to the interaction of the selected aptamer with the target and to some extent preorganize the 2194



Article

candidates as hairpins. After 15 selection rounds, the binding properties of the selected populations improved as monitored by SPR analysis (not shown): when flowing the successive pools on a hMMP-9-grafted biochip, the resonance signal increased for rounds 9 to 14 and decreased markedly for the 15th round. Seventy-seven clones were sequenced from the 14th round of selection and compared. Three sequences named F1, F2, and F3 represented 74% of the candidates. Secondary structure prediction of the three sequences using mfold showed a very high degree of similarity; selected candidates appeared as imperfect hairpins (Figure 1A). The bottom part of F1, F2, and F3 adopts a mismatched double-stranded structure contributed by the fixed regions, the folding of which corresponds to the random region of the library shown from bottom to top: a 5 or 6 nt long pyrimidine-rich internal loop and a 6 base pair G-Crich stem interrupted by a mismatch (F1 and F3) or a 5 nt internal loop (F2). F1 and F3 are predicted to form the same 10 nt long apical loop, whereas the F2 one is only 8 nt in length; both loops are pyrimidine-rich (Figure 1A). F1 and F3 display about 80% sequence homology in the 30 nt random region. Binding curves of 32P 5′-end-labeled F1, F2, and F3 aptamers to hMMP-9 were determined by filter retention assay revealing a similar affinity of F1, F2, or F3 for hMMP-9 (Figure 1B). Aptamer F3, with an equilibrium dissociation constant of 8.1 ± 3.4 nM, was a slightly stronger hMMP-9 binder than F2 (Kd = 15.4 ± 2.8 nM) or F1 (Kd = 18.3 ± 3.7 nM) and was chosen for further investigations. MMPs constitute a large family of closely related enzymes. In order to assess the specificity of the aptamer F3 for hMMP-9, we monitored its binding efficiency to the human MMP-2 (hMMP-2), the matrix metalloprotease closest to hMMP-9 also called gelatinase A, and to the human MMP-7 (hMMP-7) by the filter retention procedure. Binding of F3 to hMMP-9 and pro-hMMP-9 was specific: no retention was detected by either hMMP-7 or its proform, whereas a light signal was noticed with hMMP-2 (SI Figure S1). In order to make the synthesis and the study easier, we undertook the truncation of F3 down to the minimal size compatible with hMMP-9 binding. On the basis of the predicted structure, F3 was shortened from 68 to 36 nucleotides (variant F3B; Figure 1A), thus getting rid of most of the primer sequences except 3 residues on each side of the bottom stem. The 1H NMR analysis of the resulting F3B showed a spectrum of exchangeable protons consistent with the hairpin structure shown on Figure 1A retaining a C,C mismatch in the upper part of the stem and a pyrimidine internal hexaloop in the bottom part, under the conditions of the experiment (not shown). F3B was further shortened, thus generating two variants: F3C1 (nt 3−34 of F3B, deletion of the two first base pairs of F3B; predicted abolition of the bottom double-stranded stem) and F3C2 (nt 7−30 of F3B, deletion of the first internal loop of F3B) (Figure S2). BIAcore analysis demonstrated that F3B showed the best binding efficiency for hMMP-9, whereas F3C1 was a weaker binder and F3C2 hardly yielded a SPR signal (Figure 2A, top). This suggests that the bottom internal loop of the aptamer is essential for the interaction with hMMP-9. We then synthesized a number of mutated F3B derivatives in order to delineate the structural elements of the aptamer contributing to its binding to the target protein. Loop regions are generally crucial for the formation of aptamer−target complexes. The importance of the apical loop was first

Figure 1. Predicted secondary structure and binding characteristics of hMMP-9 aptamers. (A) Secondary structure prediction of aptamers F1, F2, and F3 selected against hMMP-9 protein and of the truncated variant F3B. G-U pairs have been taken into account. Nucleotides 1− 19 and 50−68 correspond to fixed flanks of the candidate sequences. (B) Radiolabeled aptamers, F1, F2, or F3 (1 nM), were incubated with increasing hMMP-9 concentration (from 10 to 500 nM) in the selection buffer. Binding constants (Kd) were deduced from filter retention assays as described in Materials and Methods. 2195



Article

stranded upper stem either. Second, the strands of the stems of F3B were exchanged, the original 5′ strand being placed on the 3′ side and vice versa, leading to a preserved secondary structure (SI Figure S2). This did not alter the association between hMMP-9 and the derived aptamer (SI Figure S3). We conclude that apical and internal loops were the two major F3B elements ensuring the formation of the F3B/hMMP-9 complex. Next, we optimized the F3B chemistry in the perspective of its use in biological media. In order to supply a nucleaseresistant molecular tool, the original purine ribonucleosides were substituted by 2′-O-methyl residues. As shown on SI Figure S4, incubation of an RNA hairpin in 10% fetal bovine serum known to contain endo- and exonuclease activities led to its total degradation after 30 min. On the contrary, F3B and F3Bomf remained mostly intact demonstrating nuclease resistance compared to the unmodified sample. However, faint high-mobility bands for F3B and F3Bomf samples indicated limited digestion after 3 and 6 h. The resulting aptamer F3Bomf was still able to interact with hMMP-9 with a similar efficiency to the parent F3B as indicated by SPR signal (750 RU and 805 RU under the conditions of our experiment, respectively; Figure 2A, bottom). Both sensorgrams have close profiles, with a slower dissociation phase observed for the parent F3B. The binding of this F3Bomf derivative to hMMP-9 is specific: a control scrambled sequence with the same length, same base composition, and same chemistry did not lead to a detectable SPR signal (not shown). Sensorgrams for the complex F3Bomf/hMMP-9 carried out at different protein concentrations could not be properly fitted to a 1:1 model preventing the accurate determination of kon and koff (Figure 2B, top). However, the equilibrium constant was evaluated to be in the low nanomolar range from the variation of the amplitude of the SPR signal as a function of the MMP-9 concentration. It reached a plateau at concentrations higher than 160 nM MMP-9. The Kd was taken as the concentration leading to half saturation, i.e., about 20 nM. This aptamer bound to pro-hMMP-9 and processed hMMP9 with a similar efficiency but with different binding and dissociation behavior (Figure 2B, bottom) suggesting that it likely recognizes a site exposed in both active and inactive forms of the protein. In contrast, F3Bomf was able to discriminate between the human and the murine zymogen pro-MMP-9: a weak signal (49 RU) to the mouse pro-MMP-9 was detected, compared to about 1 300 RU with the human proenzyme (Figure 2B, bottom). The specificity of the parent aptamer was also maintained following modification, as F3Bomf did not bind to either hMMP-7 or hMMP-2 (Figure 2B, bottom). Surprisingly, the fully modified 2′-O-methylribo aptamer does not allow the formation of a complex with hMMP-9 (not shown). The modified aptamer F3Bomf was then functionalized by conjugation at its 5′-end to S-acetylmercaptoacetyltriglycine (MAG3) through a hexylamino linker. This modification did not interfere with target recognition; the MAG3-F3Bomf aptamer did bind with hMMP-9, whereas preincubation of the protein with the functionalized aptamer abolished the SPR signal (SI Figure S5). Both the aptamer and the control oligonucleotide were then labeled with 99mTc as described in Materials and Methods for imaging hMMP-9 in tissues. Human Central Nervous System Tumor Imaging. MMP-9 expression in several human tumors from central nervous system was investigated either by immunohistochemistry using specific hMMP-9 antibody or by binding assay with

Figure 2. Surface plasmon resonance analysis of F3B/MMP complexes. (A) Top panel: hMMP-9 was immobilized on the sensorchip as described in Materials and Methods. F3B (red), F3C1 (blue), and F3C2 (pink) were injected at 200 nM in SP buffer. Bottom panel: biotinylated F3B (red), F3Bomf (blue) were immobilized (240 RU and 245 RU, respectively) on the biosensor, and hMMP-9 was injected at 100 nM in SP buffer. (B) Top panel: 3′-end biotinylated F3Bomf was immobilized on a sensorchip and hMMP-9 was injected in PBS buffer at different concentrations (from 5 to 160 nM as indicated). Bottom panel: hMMP-2 (black), hMMP-7 (orange), proMMP-9 (human and murine form, in blue and pink respectively), or hMMP-9 (red) (injected at 50 nM each) were assayed for interaction with aptamer F3Bomf.

investigated. In F3BΔ, the nt 14−23 of F3B were substituted by a hexaethylene glycol linker whereas the variant F3BA displayed an A10 loop (SI Figure S2). This resulted in a tremendously decreased (F3BA) or abolished (F3BΔ) interaction with hMMP-9 (SI Figure S3). We then modified the composition of the internal loop. This region originally composed of pyrimidine residues was replaced by six abasic sites (F3Bds) or replaced by 6 A residues (F3BP) (SI Figure S2). We also noticed a drastically decreased binding signal for either derivative (SI Figure S3). Finally, F3B stem was also modified. First, the C(9),C(28) mismatch in the upper stem was replaced either by an A,A mismatch (F3BAA), allowing the conservation of the secondary structure, or by a CG pair (F3BGC), allowing formation of a perfect double-stranded stem (SI Figure S2). Both variants bound to hMMP-9 with an efficiency similar to that of the parent F3B indicating that the C,C mismatch was not essential for the interaction with hMMP-9 (SI Figure S3). Indeed, this element was not conserved in F1 or F2 even though these aptamers did not display a perfect double2196



Article

99m

ependymoma, anaplastic astrocytoma, medulloblastoma, lymphoma, and glioblastoma. As shown in Figure 4 (left panels),

Tc-MAG3-F3Bomf anti-hMMP-9 aptamer. Immunohistochemical analysis revealed that hMMP-9 was highly expressed in glioblastomas (Figure 3A). Strong cytoplasmic reactivity was

Figure 3. Labeling of human glioblastoma sections with antibody or aptamer. (A) Immunohistochemical detection of hMMP-9 using a mouse anti-MMP-9 antibody (20×). Note the specific cytoplasmic immunoreactivity in tumor cells, endothelial cells, an extra-tumoral stroma. Labeling of brain section with 1.25 pmol (0.037 MBq) of the 99m Tc-MAG3-F3Bomf aptamer (B1), the 99mTc-MAG3-control sequence (B2), or 99mTc -MAG3 (C). Competition assay was carried out: incubation with 99mTc-MAG3-F3Bomf after presaturation with 1.25 pmol of either cold MAG3-F3Bomf (D1) or cold MAG3-control sequence (D2). Note that panels (B1) and (B2) on one hand and (D1) and (D2) on the other hand were prepared with the same brain section. See Materials and Methods for details.

observed for numerous tumor cells. Immunopositivity was also present in the extracellular matrix, as well as in the endothelial cells of blood vessels in the tumor environment. Of note, the antibody used for this experiment was raised against the mouse MMP-9 that does not discriminate between murine and human enzymes in contrast to the aptamer F3Bomf (Figure 2B, bottom). Therefore, the aptamer shows a higher degree of specificity than the antibody. Incubation of glioblastoma slices adjacent to the ones used for immunohistochemical analysis with the radiolabeled aptamer F3Bomf revealed a strong signal (Figure 3B1), whereas the radiolabeled control sequence induced a weaker signal (Figure 3B2). Incubation with 99mTc -MAG3 alone did not produce any detectable signal (Figure 3C). Presaturation of the slice with unlabeled MAG3-F3Bomf almost abolishes the radiolabeling by the 99mTc-aptamer (Figure 3D1), whereas presaturation with unlabeled MAG3control sequence did not prevent the labeling with 99mTcMAG3-F3Bomf (Figure 3D2). This indicates that, first, the aptamer is able to bind to its target in the environment of the tumor and, second, its binding to hMMP-9 is specific. A range of other human central nervous system (CNS) tumor types also express MMP-9.42−45 We investigated the imaging properties of the anti-hMMP-9 aptamer against pilocytic astrocytoma, meningioma, fibrillary astrocytoma,

Figure 4. Labeling of human brain tumor slices. Tumor sections were incubated either with 1.25 pmol of 99mTc-MAG3-F3Bomf aptamer (central panels) or with 1.25 pmol of 99mTc-MAG3-control sequence (right panels) and compared with hMMP-9 immunoreactivity (left panels). See Materials and Methods for details.

the hMMP-9 expression, monitored by the immunohistochemical method, revealed a cytoplasmic staining dependent on the tumor grade within the group of glial infiltrative tumors. For the other primitive brain tumors explored (pilocytic astrocytoma grade 1, fibrillary astrocytoma grade 2, anaplastic astrocytoma grade 3, glioblastoma grade 4, ependymoma grade 2, meningioma grade 1, and medulloblastoma grade 4), hMMP-9 immunostaining showed a variable intensity of cytoplasmic expression. In all cases (glial and other tumor 2197



■

type), immunostaining for hMMP-9 was also clearly observed within both extracellular environment and endothelial cells. For primitive central nervous system lymphoma, hMMP-9 expression was weak in cytoplasmic compartment and in extracellular matrix. Healthy brain was used as control: no immunoreactivity for hMMP-9 was detected. The same tumors were incubated with the labeled anti-hMMP-9 aptamer. Generally, 99mTc-MAG3-F3Bomf induced a strong signal on the tissues (Figure 4, central panel), whereas a much weaker signal was recorded with the control sequence (Figure 4, right panel). No signal was detected with healthy brain tissue. We were not able to quantitate the signal obtained by immunostaining. Visual analysis revealed that immunohistochemical expression of MMP-9 increased roughly in a tumor grade-dependent manner. We also generally observed an increasing binding of F3Bomf aptamer with the grade of malignancy. We estimated that binding intensities roughly correlate with the amount of MMP-9 by comparison with the immunohistochemical signal. Therefore, whatever the tumor type a good agreement was observed between antibody fixation and radiolabeled F3Bomf aptamer binding.

Article

ASSOCIATED CONTENT

S Supporting Information *

Five figures: Binding specificity of F3 aptamer; Secondary structure of F3B hMMP-9 aptamer variants; Surface Plasmon Resonance analysis of F3B hMMP-9 variants; nuclease resistance of oligonucleotide derivatives; Surface Plasmon Resonance analysis of MAG functionalized F3Bomf aptamer binding to hMMP-9. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Tel 335 4000 3034; Fax 335 4000 3004. Present Address ⬢

Novaptech, Pessac, France.

Notes

The authors declare no competing financial interest. ▽ Co-first authors.

■

■

ACKNOWLEDGMENTS We thank I. Lebars for 1H NMR analysis of the aptamer F3B, Nathalie Pierre and Serge Moreau for oligonucleotide syntheses, and Carmelo Di Primo for SPR expertise. NMR, mass spectrometry, and SPR analysis were carried out in the IECB central facility (UMS3033/US001). This work was supported by “Régions Aquitaine et Midi Pyrénées, France”.

CONCLUSION Aptamers are attractive in biomedicine because of their advantages over antibodies, which rely in particular on their reproducible chemical production, low immunogenicity, reversible denaturation, and small size. Aptamers show many of the requested criteria for the ideal imaging probe. They are high-affinity binding ligands and show high tissue-specific retention and rapid blood clearance for in vivo imaging.17 Because of their easy conjugation to the appropriate label (fluorophore, radionuclide), aptamers afford a valuable alternative to antibodies for protein detection. MMPs are relevant markers of tumor malignancy. So far, molecular imaging of MMPs has been performed in tumorbearing mice with fluorescent peptide substrates46 or in human carotid47 using MMP inhibitor radiotracers48,49 or proteolytic nanobeacons.50 Investigation of MMP-2 and hMMP-9 expression with a 64Cu radiolabeled cyclic peptide by microPET failed to demonstrate a specific uptake in gelatinase-expressing tumors,51 whereas the same cyclic peptide 68Ga-DOTA conjugate showed acceptable plasma stability and good visualization of tumor xenografts.52 Recently, a 99mTcmonoclonal antibody was developed to target a membrane MMP for imaging atherosclerosis.53 In this work, we have successfully obtained and characterized aptamers displaying high affinity for hMMP-9 protein. Aptamer F3 was shortened and modified to generate MAG3-F3Bomf, an aptamer-based imaging probe. We could detect specifically hMMP-9 protein, a tumor biomarker, on different human tumor slices with 99mTc-MAG3-F3Bomf. It is the first aptamer application for hMMP-9 detection. Its high specificity will improve the signal-to-noise ratio compared to broad-spectrum MMP inhibitors, which lead to high uptake in tissues with non pathogenic MMP expression. Our goal is to develop an aptamer-based imaging tool for specific tumor monitoring in clinical studies. Chemistry and size have been optimized, but future improvements are still needed to enhance its retention in vivo (i.e., multimers, pegylation). Aptamer injections in human tumor-bearing mice are scheduled.

■

REFERENCES

(1) Louis, D. N., Ohgaki, H., Wiestler, O. D., Cavenee, W. K., Burger, P. C., Jouvet, A., Schelthauer, B. W., and Kleihues, P. (2007) The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol. 114, 97−109. (2) Laperriere, N., Zuraw, L., and Cairncross, G. (2002) Radiotherapy for newly diagnosed malignant glioma in adults: a systematic review. Radiother. Oncol. 64, 259−73. (3) Heimberger, A. B., McGary, E. C., Suki, D., Ruiz, M., Wang, H., Fuller, G. N., and Bar-Eli, M. (2005) Loss of the AP-2alpha transcription factor is associated with the grade of human gliomas. Clin. Cancer Res. 11, 267−72. (4) Rao, J. S. (2003) Molecular mechanisms of glioma invasiveness: the role of proteases. Nat. Rev. Cancer 3, 489−501. (5) Stamenkovic, I. (2003) Extracellular matrix remodelling: the role of matrix metalloproteinases. J. Pathol. 200, 448−64. (6) Binder, D. K., and Berger, M. S. (2002) Proteases and the biology of glioma invasion. J. Neurooncol. 56, 149−58. (7) Auge, F., Hornebeck, W., and Laronze, J. Y. (2004) A novel strategy for designing specific gelatinase A inhibitors: potential use to control tumor progression. Crit. Rev. Oncol. Hematol. 49, 277−82. (8) Bachmeier, B. E., Iancu, C. M., Jochum, M., and Nerlich, A. G. (2005) Matrix metalloproteinases in cancer: comparison of known and novel aspects of their inhibition as a therapeutic approach. Expert Rev. Anticancer Ther. 5, 149−63. (9) Forget, M. A., Desrosiers, R. R., and Beliveau, R. (1999) Physiological roles of matrix metalloproteinases: implications for tumor growth and metastasis. Can. J. Physiol. Pharmacol. 77, 465−80. (10) Tuerk, C., and Gold, L. (1990) Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 249, 505−10. (11) Ellington, A. D., and Szostak, J. W. (1990) In vitro selection of RNA molecules that bind specific ligands. Nature 346, 818−22. (12) Robertson, D. L., and Joyce, G. F. (1990) Selection in vitro of an RNA enzyme that specifically cleaves single-stranded DNA. Nature 344, 467−8.

2198



Article

(13) Mayer, G. (2009) The chemical biology of aptamers. Angew. Chem., Int. Ed. Engl. 48, 2672−89. (14) Syed, M. A., and Pervaiz, S. (2010) Advances in aptamers. Oligonucleotides 20, 215−24. (15) Soontornworajit, B., and Wang, Y. (2011) Nucleic acid aptamers for clinical diagnosis: cell detection and molecular imaging. Anal. Bioanal. Chem. 399, 1591−9. (16) Dausse, E., Da Rocha Gomes, S., and Toulmé, J. J. (2009) Aptamers: a new class of oligonucleotides in the drug discovery pipeline? Curr. Opin. Pharmacol. 9, 602−7. (17) Gomes, S. D. R., Azéma, L., Allard, M., and Toulmé, J. J. (2010) Aptamers as imaging agents. Expert Opinion on Medical Diagnostics 4, 511−518. (18) Charlton, J., Sennello, J., and Smith, D. (1997) In vivo imaging of inflammation using an aptamer inhibitor of human neutrophil elastase. Chem. Biol. 4, 809−16. (19) Lupold, S. E., Hicke, B. J., Lin, Y., and Coffey, D. S. (2002) Identification and characterization of nuclease-stabilized RNA molecules that bind human prostate cancer cells via the prostatespecific membrane antigen. Cancer Res. 62, 4029−33. (20) Hicke, B. J., Stephens, A. W., Gould, T., Chang, Y. F., Lynott, C. K., Heil, J., Borkowski, S., Hilger, C. S., Cook, G., Warren, S., and Schmidt, P. G. (2006) Tumor targeting by an aptamer. J. Nucl. Med. 47, 668−78. (21) Shangguan, D., Li, Y., Tang, Z., Cao, Z. C., Chen, H. W., Mallikaratchy, P., Sefah, K., Yang, C. J., and Tan, W. (2006) Aptamers evolved from live cells as effective molecular probes for cancer study. Proc. Natl. Acad. Sci. U. S. A. 103, 11838−43. (22) Herr, J. K., Smith, J. E., Medley, C. D., Shangguan, D., and Tan, W. (2006) Aptamer-conjugated nanoparticles for selective collection and detection of cancer cells. Anal. Chem. 78, 2918−24. (23) Borbas, K. E., Ferreira, C. S., Perkins, A., Bruce, J. I., and Missailidis, S. (2007) Design and synthesis of mono- and multimeric targeted radiopharmaceuticals based on novel cyclen ligands coupled to anti-MUC1 aptamers for the diagnostic imaging and targeted radiotherapy of cancer. Bioconjugate Chem. 18, 1205−12. (24) Phillips, J. A., Lopez-Colon, D., Zhu, Z., Xu, Y., and Tan, W. (2008) Applications of aptamers in cancer cell biology. Anal. Chim. Acta 621, 101−8. (25) Li, W., Yang, X., Wang, K., Tan, W., He, Y., Guo, Q., Tang, H., and Liu, J. (2008) Real-time imaging of protein internalization using aptamer conjugates. Anal. Chem. 80, 5002−8. (26) Sefah, K., Tang, Z. W., Shangguan, D. H., Chen, H., LopezColon, D., Li, Y., Parekh, P., Martin, J., Meng, L., Phillips, J. A., Kim, Y. M., and Tan, W. H. (2009) Molecular recognition of acute myeloid leukemia using aptamers. Leukemia 23, 235−44. (27) Smith, J. E., Medley, C. D., Tang, Z., Shangguan, D., Lofton, C., and Tan, W. (2007) Aptamer-conjugated nanoparticles for the collection and detection of multiple cancer cells. Anal. Chem. 79, 3075−82. (28) Medley, C. D., Bamrungsap, S., Tan, W., and Smith, J. E. (2011) Aptamer-conjugated nanoparticles for cancer cell detection. Anal. Chem. 83, 727−34. (29) Ko, H. Y., Choi, K. J., Lee, C. H., and Kim, S. (2011) A multimodal nanoparticle-based cancer imaging probe simultaneously targeting nucleolin, integrin alphavbeta3 and tenascin-C proteins. Biomaterials 32, 1130−8. (30) Zhou, J., Soontornworajit, B., and Wang, Y. (2010) A temperature-responsive antibody-like nanostructure. Biomacromolecules 11, 2087−93. (31) Blank, M., Weinschenk, T., Priemer, M., and Schluesener, H. (2001) Systematic evolution of a DNA aptamer binding to rat brain tumor microvessels. selective targeting of endothelial regulatory protein pigpen. J. Biol. Chem. 276, 16464−8. (32) Farokhzad, O. C., Jon, S., Khademhosseini, A., Tran, T. N., Lavan, D. A., and Langer, R. (2004) Nanoparticle-aptamer bioconjugates: a new approach for targeting prostate cancer cells. Cancer Res. 64, 7668−72.

(33) Shangguan, D., Meng, L., Cao, Z. C., Xiao, Z., Fang, X., Li, Y., Cardona, D., Witek, R. P., Liu, C., and Tan, W. (2008) Identification of liver cancer-specific aptamers using whole live cells. Anal. Chem. 80, 721−8. (34) Li, S., Xu, H., Ding, H., Huang, Y., Cao, X., Yang, G., Li, J., Xie, Z., Meng, Y., Li, X., Zhao, Q., Shen, B., and Shao, N. (2009) Identification of an aptamer targeting hnRNP A1 by tissue slide-based SELEX. J. Pathol. 218, 327−36. (35) Zhang, Z., Guo, L., Guo, A., Xu, H., Tang, J., Guo, X., and Xie, J. (2010) In vitro lectin-mediated selection and characterization of rHuEPO-alpha-binding ssDNA aptamers. Bioorg. Med. Chem. 18, 8016−25. (36) Shi, H., He, X., Wang, K., Wu, X., Ye, X., Guo, Q., Tan, W., Qing, Z., Yang, X., and Zhou, B. (2011) Activatable aptamer probe for contrast-enhanced in vivo cancer imaging based on cell membrane protein-triggered conformation alteration. Proc. Natl. Acad. Sci. U. S. A. 108, 3900−5. (37) Sousa, R., and Padilla, R. (1995) A mutant T7 RNA polymerase as a DNA polymerase. Embo J. 14, 4609−21. (38) Bartel, D. P., and Szostak, J. W. (1994) Study of RNA-protein recognition by in vitro selection, in RNA-Protein interactions (Nagai, K., and Mattaj, I. W., Eds.) pp 248−268, IRL Press, Oxford. (39) McEntee, K., Weinstock, G. M., and Lehman, I. R. (1980) recA protein-catalysed strand assimilation: stimulation by Escherichia coli single-stranded DNA-binding protein. Proc. Natl. Acad. Sci. U.S.A. 77, 857−61. (40) Winnard, P., Jr., Chang, F., Rusckowski, M., Mardirossian, G., and Hnatowich, D. J. (1997) Preparation and use of NHS-MAG3 for technetium-99m labeling of DNA. Nucl. Med. Biol. 24, 425−32. (41) Di Primo, C., Dausse, E., and Toulme, J. J. (2011) Surface plasmon resonance investigation of RNA aptamer-RNA ligand interactions. Methods Mol. Biol. 764, 279−300. (42) Comincini, S., Paolillo, M., Barbieri, G., Palumbo, S., Sbalchiero, E., Azzalin, A., Russo, M. A., and Schinelli, S. (2009) Gene expression analysis of an EGFR indirectly related pathway identified PTEN and MMP9 as reliable diagnostic markers for human glial tumor specimens. J. Biomed. Biotechnol. 2009, 924565. (43) Das, G., Shiras, A., Shanmuganandam, K., and Shastry, P. (2010) Rictor regulates MMP-9 activity and invasion through Raf-1-MEKERK signaling pathway in glioma cells. Mol. Carcinog. 50, 412−423. (44) Lee, E. J., Kim, S. Y., Hyun, J. W., Min, S. W., Kim, D. H., and Kim, H. S. (2010) Glycitein inhibits glioma cell invasion through down-regulation of MMP-3 and MMP-9 gene expression. Chem. Biol. Interact. 185, 18−24. (45) Veeravalli, K. K., Chetty, C., Ponnala, S., Gondi, C. S., Lakka, S. S., Fassett, D., Klopfenstein, J. D., Dinh, D. H., Gujrati, M., and Rao, J. S. (2010) MMP-9, uPAR and cathepsin B silencing downregulate integrins in human glioma xenograft cells in vitro and in vivo in nude mice. PLoS One 5, e11583. (46) Bremer, C., Tung, C. H., and Weissleder, R. (2001) In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat. Med. 7, 743−8. (47) Wallis de Vries, B. M., Hillebrands, J. L., van Dam, G. M., Tio, R. A., de Jong, J. S., Slart, R. H., and Zeebregts, C. J. (2009) Images in cardiovascular medicine. Multispectral near-infrared fluorescence molecular imaging of matrix metalloproteinases in a human carotid plaque using a matrix-degrading metalloproteinase-sensitive activatable fluorescent probe. Circulation 119, e534−6. (48) Furumoto, S., Takashima, K., Kubota, K., Ido, T., Iwata, R., and Fukuda, H. (2003) Tumor detection using 18F-labeled matrix metalloproteinase-2 inhibitor. Nucl. Med. Biol. 30, 119−25. (49) Razavian, M., Zhang, J., Nie, L., Tavakoli, S., Razavian, N., Dobrucki, L. W., Sinusas, A. J., Edwards, D. S., Azure, M., and Sadeghi, M. M. (2010) Molecular imaging of matrix metalloproteinase activation to predict murine aneurysm expansion in vivo. J. Nucl. Med. 51, 1107−15. (50) Scherer, R. L., VanSaun, M. N., McIntyre, J. O., and Matrisian, L. M. (2008) Optical imaging of matrix metalloproteinase-7 activity in vivo using a proteolytic nanobeacon. Mol. Imaging 7, 118−31. 2199



Article

(51) Sprague, J. E., Li, W. P., Liang, K., Achilefu, S., and Anderson, C. J. (2006) In vitro and in vivo investigation of matrix metalloproteinase expression in metastatic tumor models. Nucl. Med. Biol. 33, 227−37. (52) Ujula, T., Huttunen, M., Luoto, P., Perakyla, H., Simpura, I., Wilson, I., Bergman, M., and Roivainen, A. (2010) Matrix metalloproteinase 9 targeting peptides: syntheses, 68Ga-labeling, and preliminary evaluation in a rat melanoma xenograft model. Bioconjugate Chem. 21, 1612−21. (53) Kuge, Y., Takai, N., Ogawa, Y., Temma, T., Zhao, Y., Nishigori, K., Ishino, S., Kamihashi, J., Kiyono, Y., Shiomi, M., and Saji, H. (2010) Imaging with radiolabelled anti-membrane type 1 matrix metalloproteinase (MT1-MMP) antibody: potentials for characterizing atherosclerotic plaques. Eur. J. Nucl. Med. Mol. Imaging 37, 2093−104.

2200


99mTc-MAG3-Aptamer for Imaging Human Tumors Associated with

Recommend Documents