Fluorescent Nanoprobes as a Biomarker for Increased Vascular

Bioconjugate Chem. 2010, 21, 93–101

93

Fluorescent Nanoprobes as a Biomarker for Increased Vascular Permeability: Implications in Diagnosis and Treatment of Cancer and Inflammation Britto S. Sandanaraj,*,† Hans-Ulrich Gremlich,† Rainer Kneuer,*,† Janet Dawson,‡ and Stefan Wacha† Global Imaging Group and Department of Autoimmunity, Transplantation and Inflammation, Novartis Institutes for Biomedical Research, Basel, Switzerland. Received July 15, 2009; Revised Manuscript Received September 27, 2009

This article describes the use of a fluorescent nanoprobe as a functional biomarker for the identification of increased vascular permeability in cancer/arthritis disease models. Synthesis of the fluorescent nanoprobe was achieved by passive loading of a fluorophore inside the nanoparticle using thin film hydration method. The outer layer of the nanoprobe was decorated with poly(ethylene glycol) arms to increase the bioavailability of the fluorophore. Stability studies of the nanoprobe showed that the particles were stable up to 70 days. The uptake and internalization of the fluorescent nanoprobe inside target cells was confirmed by fluorescence microscopy studies. Co-localization of the probe with the target tissue in ViVo was unambiguously identified using intravital microscopy. Results from in ViVo imaging studies showed that the particles had a long half-life in the circulation and passively targeted tumor or arthritic tissue. The increased and specific uptake of the fluorescent nanoprobe in tumor/arthritic tissue is attributed to an enhanced permeation and retention (EPR) effect. Use of an optical method to validate antiinflammatory drugs in an arthritis disease model is demonstrated in this study. In general, this methodology could be used for detection of leaky vasculature in different pathological states.

INTRODUCTION Biomarkers play a crucial role in both preclinical and clinical evaluation of drugs (1–7). In particular, imaging-based biomarkers can speed up drug discovery by supplementing or replacing preclinical and clinical pharmacokinetic and pharmacodynamic assessment, including target selection and modulation. In addition, imaging-based biomarkers could be used for diagnosis, staging, prognosis, and treatment selection (8–10). Presence of leaky vasculature is a classical biomarker for some diseases which include cancer (11), inflammation (12, 13), and diabetes (14, 15). The intrinsic biological processes that drive the formation of leaky blood vessels can vary depending on various disease conditions. However, leaky blood vessels present in different disease conditions share some unique properties that could be used as biomarkers for the identification of pathological states. Dynamic contrast-enhanced MRI (DCE-MRI) (16, 17), positron-emission tomography (PET) (18), dynamic contrast-enhanced CT (DCE-CT) (19–21), and contrast enhanced ultrasound (CEUS) (22) have been routinely used as functional biomarkers for measuring the vascular leakiness in various disease conditions. Although these techniques have clearly paved the way for the use of imaging based biomarkers as a routine tool in both preclinical and clinical settings, they are not optimum in their performance: for example, DCE-MRI suffers from quantification and reproducibility, use of DCECT might be a problem because of the high exposure to X-ray radiation, and requirement for in-house cyclotron is a limitation in the case of PET. Therefore, there is a need for new functional biomarkers to overcome these issues. Fluorescence microscopy and near-infrared fluorescence (NIRF) imaging are important research tools in the area of in * Email: [email protected], rainer.kneuer@novartis. com. † Global Imaging Group. ‡ Department of Autoimmunity, Transplantation and Inflammation.

ViVo imaging applications (10, 23–28) These techniques provide information at a molecular level, thus offering a distinct advantage over other imaging modalities such as magnetic resonance imaging and computer tomography. The use of light in the near-infrared range (NIR; 650-950 nm) is well suited for in ViVo studies due to the reduced absorbance and scattering of endogenous substances (especially water, oxyhemoglobin, and melanin). The low cost and high sensitivity of this imaging modality has increased the widespread use of this technique in diagnostics (29), preclinical drug discovery (30, 31), and, to some extent, clinical settings (32, 33). The use of optical methods to assess vascular leakiness is an attractive imaging based biomarker to speed up drug discovery in preclinical settings. As a “proof of concept”, we have shown that “optical imaging combined with fluorescent nanoprobes” can be used effectively as a functional biomarker for the identification of increased vascular leakiness. The versatility of this method is demonstrated by using this technique for diagnosis and monitoring of two different disease conditions where vascular leakiness exists. In general, this methodology can be extended to any disease model where leaky blood vessels exist. This technology can be translated easily to clinical settings since all the components used for formulation of fluorescent nanoprobes are clinically approved for different applications (34). By encapsulating a clinically approved fluorescent dye (e.g., indocyanine green (35, 36) (ICG)) inside a nanoprobe, this nanoparticle can be used as a biomarker for functional imaging in clinical settings. The preclinical studies described here were carried out with a nanoparticle-ICG formulation for one such application in a mouse model (Vide infra). We have used sterically shielded liposome (SSL) (37) based nanoparticulates as a delivery vehicle to carry our fluorescent probes to their target location. The reason for using a delivery vehicle is twofold; first, it increases the bioavailability of the probe in in ViVo conditions, second, it increases the targeting efficiency of this nanoparticle toward leaky vasculature (11, 38), thus reducing the signal/noise ratio. Since all the components of the formula-

10.1021/bc900311h  2010 American Chemical Society Published on Web 12/03/2009

94 Bioconjugate Chem., Vol. 21, No. 1, 2010

tion are already used in clinics, we do not anticipate any toxicity from the delivery vehicle. Apart from in ViVo validation of the probe, we have also validated the performance of the probe at the cellular level.

EXPERIMENTAL PROCEDURES Reagents. Egg phosphaditylcholine, cholesterol, 1,2-disteroylsn-glycero-3-phosphoethanolamine-N-methoxy(poly(ethylene glycol))-2000) were purchased from Avanti Polar Lipids, Inc. Cyanine 5.5 and indocyanine green (ICG) were purchased from Fluka. Hoechst, Phalloidin, and CM-DiI were purchased from Molecular Probes. RPMI medium and fetal bovine serum were purchased from Invitrogen. Synthesis and Characterization of Fluorescent Nanoprobes. Liposomes were prepared at defined molar ratio of individual lipids to give a predetermined total lipid concentration of 7.6 mg/mL. Egg phosphaditylcholine, cholesterol, 1,2-disteroyl-sn-glycero-3-phosphoethanolamine-N-methoxy(poly(ethylene glycol))-2000) (2:1:0.2 molar ratio) and fluorescent probe (Cy5.5 or ICG) were dissolved in 5 mL chloroform, and the resulting mixture was evaporated on a rotary evaporator under room temperature. The resulting lipid film was again subjected to high vacuum to remove residual solvent. Finally, the lipid film was hydrated at 50 °C with 5 mL 10 mM PBS buffer (pH 7.4) to give liposome solution. The hydrated mixture was frozen at liquid nitrogen temperature and heated to 50 °C. This treatment was repeated 3 times for the preferential encapsulation of Cy5.5 or ICG inside the nanoprobe. After this, the solution was passed through a 0.2 µM filter (3 times) to get rid of larger particles. The unbound probe molecules present in the bulk solution were separated by passing solution through a Sephadex G-25 M column. Percentage encapsulation efficiency of Cy5.5 and ICG was 5% and 50%, respectively. Liposomes were lysed by the addition of methanol, and the encapsulation efficiency was calculated from the amount of released dye, which is determined from the UV-vis calibration curve. Size of nanoprobes was characterized by dynamic light scattering using a Zetasizer Nano ZS over a time period of 70 days at 4 °C. Spectroscopic Measurements. The UV-vis absorption spectra were recorded on a PerkinElmer spectrophotometer using plastic cells. Fluorescence spectra were recorded on a Varian fluorimeter. The spectra were recoded using a 1 mL disposable plastic cell. Excitation and emission bandwidth was kept at 2.5 and 5 nm, respectively, for Cy5.5 nanoprobe and for ICG nanoprobe bandwidth was kept at 10 and 20 nm, respectively. Cy5.5 nanoprobe was excited at 660 nm and the emission was recorded from 670 to 900 nm. For ICG-nanoprobe, the excitation wavelength was 770 nm and emission was recorded from 770 to 900 nm. Cell Studies and Imaging with IN Cell Analyzer 3000. Prostate cancer cells (PC3, passage number 12) were cultured at 37 °C under 5% CO2 in RPMI supplemented with 10% heat-inactivated fetal bovine serum, 2% glutamine, and 0.1% penicillin-streptomycin solution. For passaging, cells were detached with 0.25% trypsin/1 mM EDTA, counted with nucleocounter (ChemoMetec, Allerød, Denmark), and the required number of cells were seeded in T75 cell culture flasks. For imaging plates, Cells were seeded on black Corning 384-well plates using multidrop, 10 000 cells/well in 100 µL, and incubated for 24 h. After 24 h, 16.6 µL nanoprobes (PEGylated nanoprobe 1: size ) 73 nm, Cy5.5 concentration ) 50 µg/mL. PEGylated nanoprobe 2: size ) 120 nm, Cy5.5 concentration ) 43 µg/mL. Naked nanoprobe: size ) 106 nm, Cy5.5 concentration ) 24 µg/mL. Control nanoprobe: size ) 106 nm, Cy5.5) were added and incubated

Sandanaraj et al.

for required time. The cells were fixed by adding 100 µL 8% PFA which contained 50 ng/well Hoechst (a nuclear stain). Thus, the end concentration of PFA was 4% and that of Hoechst was 25 ng/well. Cells were incubated with PFA and Hoechst for 15 min at RT. Then, the PFA-Hoechst solution was aspirated. The cells were washed once with PBS. The cells were washed twice with 20 µL PBS, which was aspirated, and the wells were filled with 20 µL PBS. The plates were sealed with silverseal aluminum tape and kept at 4 °C in the dark until imaging with the IN Cell Analyzer 3000. Images were acquired using the IN Cell Analyzer 3000 (GE Biosciences, Piscataway, NJ) automated cell imager. Flat field correction was done using a plate with various mixtures of Oregon Green (Molecular Probes, Carlsbad, CA), Cy5 calibration reagent (GE Biosciences), and Alexa Fluor 350 carboxylic acid (Molecular Probes). Images for the nuclear stain Hoechst and Cy5.5 were taken on 2 different channels with 2 CCD cameras simultaneously. Cy5.5 was excited with the 647 nm krypton laser, and a 695/BP55 emission filter was used to collect the emitted light. Hoechst was excited with 350 nm line of the argon laser, and a 450/BP50 emission filter was used to collect the emitted light. Exposure time for both the channels was 1.7 ms. Intravital Imaging and Preparation of DiI mice. Two weeks before the experiment, 2 × 106 PC3 prostate cancer cells were labeled in Vitro with DiI and implanted subcutaneously into nude mice. Twenty-four hours before imaging, the Cy5.5-labeled nanoprobe (size ) 120 nm, Cy5.5 concentration ) 43 µg/mL) was injected via tail vein injection 5 min before imaging, FITC-dextran (MW ) 500 000; 5 µL of 1 mg/mL stock diluted into 95 µL NaCl 0.9%) was applied via tail vein injection as well, and then mice were anesthetized with Ketamin/Xylasol and a skin-flap was performed in order to get access to the subcutaneously implanted tumor. Images were taken using the IV100 intravital microscope from Olympus using 4× and 20× objectives as well as a 27× stick lense for highest resolution. A stock solution of 1 mg/mL CM-DiI (Molecular Probes) was prepared in dimethyl sulfoxide (DMSO) and stored at -20 °C. For labeling of cells, the stock solution was diluted in cell culture medium to achieve a concentration of 10 µg/mL. One million cells were suspended per milliliter and incubated for 30 min at 37 °C. Cells were gently resuspended every 5 min. Afterwards, cells were washed twice in PBS (centrifugation for 10 min at 400 g at room temperature) and resuspended in cell culture medium at 750 000 cells per 100 µL. Mouse Tumor Model and NIRF Imaging. 3 × 106 PC-3 human prostate adenocarcinoma (ATCC: CRL-1435) cells in 100 µL HBSS (1 mg/mL Matrigel) were injected under Forene inhalation narcosis on the right flank of 6-8 week old female athymic nude mice for generation of subcutaneous tumors. The Siemens BonSAI Imager was used (Siemens AG, Medical Solutions, and Erlangen, Germany). NIRF imaging started at day 14 after inoculation of the tumor cells, and NIRF images were recorded 6, 24, 48, 72, and 96 h after i.v. administration of the 100 µL Cy5.5 nanoprobe (size ) 73 nm, Cy5.5 concentration ) 50 µg/mL). For NIRF measurements, the animals were anaesthetized using 1.5% Isoflurane (Abbott, Cham) in nitrous oxide/oxygen 2:1 administered via a face mask. The duration of the anesthesia was 3 min. All experiments have been carried out in adherence to the Swiss laws of animal protection. For fluorescence excitation, three laser diodes at 660 nm with a power of 10 mW/cm2 were used. The fluorescent light emitted from the sample (mouse) was detected by a charge-coupled device (CCD) camera (Hamamatsu) equipped with a focusing lens system (macro lens 60 mm, 1:2.8, Nikon). The CCD features low noise and

Fluorescent Nanoprobes as Biomarker

Bioconjugate Chem., Vol. 21, No. 1, 2010 95

Figure 1. (A) Chemical constituents of fluorescent nanoprobe. (B) Schematic representation of formulation procedure.

low dark signal enabling low light level detection as well as long integration times. The matrix size of the images is 532 × 256 pixels. A hard filter was used for detection wavelength selection (700 nm). Data acquisition (i.e., integration) times ranged from 0.5 to 2.0 s depending on the intensity of fluorescence. The experiment was controlled by a PC using the Siemens SYNGO software. NIRF images have been analyzed quantitatively on a ROI basis using the same SYNGO software. Mouse Antigen Induced Arthritis Model and NIRF Imaging. Ten female OF-1 mice, 12-14 weeks of age, were used. They were housed under standard conditions with a 12 h light/dark cycle, and water was provided ad libitum. During the 14 days before imaging experiments, the animals were fed a low-pheophorbide diet to reduce autofluorescence. Arthritis was induced in all 10 mice. The phase of systemic immunization proceeded as follows. Mice were sensitised intradermally on the back at two sites to methylated bovine serum albumin (mBSA) homogenised 1:1 with complete Freund’s adjuvant on days -21 and -14 (0.1 ml containing 1 mg/ml mBSA). On day 0, the mice were anaesthetised using a 5% isoflurane/air mixture and maintained using isoflurane in a face mask for the intra-articular injections. The right knee received 10 µl of 10 mg/ml mBSA in 5% glucose solution (antigen injected knee), while the left knee received 10 µl of 5% glucose solution alone (vehicle injected knee). From day 0 to 7, the animals were treated with dexamethasone (NVP-AFC572-AB-2; 1 mg/kg, p.o.) or vehicle (A. dist). Arthritis development was monitored by measuring the lateral joint diameter with an Oditest vernier caliper on days 0, 2, and 4. Using a GE eXplore Optix small animal imager (GE Healthcare, London, England), NIRF images for arthritis studies were recorded 6 h after i.v. injection of the ICG nanoprobe (size) 120 nm, ICG concentration ) 151 µg/mL). For NIRF measurements, the animals were anaesthetized using 1.5% Isoflurane (Abbott, Cham) in nitrous oxide/oxygen 2:1 administered via a face mask. The duration of the anesthesia was 3 min, the time needed to scan 1 animal. All experiments have been carried out in adherence to the Swiss laws of animal protection. A pulsed laser at 670 nm with a power of 4.4 mW at 80 MHz repetitive rate and pulse durations of 60 ps was used for fluorescence excitation. The fluorescent light emitted from the sample (mouse) was detected by a photomultiplier tube (TD-

Figure 2. (A) UV-vis spectra of Cy 5.5 or ICG inside nanoprobe. (B) Fluorescence spectra of Cy 5.5 or ICG inside nanoprobe. Cy5.5 nanoprobe was excited at 660 nm, and the emission was recorded from 670 to 900 nm. For ICG-nanoprobe, the excitation wavelength was 770 nm and emission was recorded from 770 to 900 nm. (C) Stability studies of nanoprobe formulation: average size and polydispersity index of liposomes with respect to time.

detector) at 700 nm. The experiment was controlled by a PC using the GE eXplore Optix acquisition software. NIRF images have been analyzed quantitatively on a ROI basis using the same software tool.


Sandanaraj et al.

Figure 3. NIRF imaging started at day 14 after inoculation of the tumor cells, and NIRF images were recorded 6, 24, 48, 72, and 96 h after i.v. administration of the 100 µL Cy5.5 nanoprobe (size ) 73 nm, Cy5.5 concentration ) 50 µg/mL). (A) A representative NIRF whole body image at 24 h time point. (B) Fluorescence signal from tumor and nontumor area. (C) Ratio of tumor to nontumor signal ratio. Each data point represents the mean ( SEM of 3 animals.

Statistical Analysis. All results are expressed as mean ( SEM (n ) number of animals).

RESULTS AND DISCUSSION Formulation and Characterization of Fluorescent Nanoprobes. Formulation of sterically shielded liposomes (SSL) was achieved by thin film hydration method. Egg phosphatidylcholine, cholesterol, 1,2-disteroyl-sn-glycero-3-phosphoethanolamine-N-methoxy(poly(ethylene glycol))-2000) (2:1:0.2 molar ratio) and Cy5.5 or ICG were mixed together and subjected to repeated freeze-thaw cycles and filtration through polycarbonate membranes and finally passed through a Sephadex column (Figure 1A). The size and polydispersity index (PDI) of the synthesized nanoprobes were estimated using dynamic light scattering (DLS) technique. Scanning electron microscopy studies showed that the particles are spherical and about the size of 100-150 nm, which is consistent with the results from DLS studies (data not shown). The liposome particles are decorated with poly(ethylene glycol) (PEG) arms in order to

improve the pharmacokinetics/pharmacodynamics in ViVo and avoid particle phagocytosis by macrophages (39). As the probe molecules can be hydrophilic or hydrophobic depending on the nature of chemical functionality, it is crucial that delivery vehicles can encapsulate all types of compounds irrespective of their hydrophilic/hydrophobic index. The advantage of using liposomes is twofold: (i) they have an aqueous interior where hydrophilic compounds can be dissolved in the nanoscale dimension; (ii) second, they also have lipid bilayers made up of phosphatidylcholine and cholesterol that can be exploited to encapsulate hydrophobic probe molecules. Therefore, we consider that liposome-based nanoparticulates are ideal delivery vehicles for in ViVo and cell-based applications. In order to demonstrate that our delivery vehicles can indeed encapsulate all types of probe molecules, we have tested the encapsulation ability of these particles with both hydrophobic (indocyanine green, ICG) and hydrophilic (Cy5.5) fluorescent probes. As expected, the liposomes can dissolve both hydrophilic and hydrophobic probe molecules (Figure 2). We also carried out



Figure 4. Cy5.5-labeled nanoprobe (size ) 120 nm, Cy5.5 concentration ) 43 µg/mL) was injected via tail vein injection 5 min before imaging, FITC-dextran (MW ) 500 000; 5 µL of 1 mg/mL stock diluted into 95 µL NaCl 0.9%) was applied via tail vein injection. (A) 30 min; (B) 6 h; (C) 24 h.

fluorescence measurements to determine that the dye inside the nanoscale compartment of liposomes resembles a dye in bulk solution. This is important because it has been shown that increased local concentration of a fluorophore inside the aqueous interior of the liposome might result in fluorescence quenching, thus complicating any analysis of the results. We have also performed stability studies of liposome formulation for 70 days and found that the particles were exceptionally stable. There was no problem of particle aggregation with long-term storage (Figure 2C). In ViWo Studies of Fluorescent Nanoprobes in Tumor Model. In order to test the passive targeting efficiency of these particles toward leaky vasculature, we chose tumor and arthritis disease models in mice. These two diseases share a common feature, which is the leaky vasculature nature of tumor/arthritic tissue (11) where the fluorescent liposome particles can be trapped and retained for longer time periods through the process called enhanced permeation and retention (EPR) effect (11, 38). The first in ViVo studies of fluorescent nanoprobes were carried out in mice bearing PC3 tumors. PC3 tumor cells were subcutaneously implanted in nude mice and allowed to grow for 2 weeks until tumors achieved size of 150 mm3. The nanoprobe containing Cy5.5 dye was injected i.v. via the tail

vein of the mice and the particles tracked at different times. The 6 h time point revealed that the particles were circulating uniformly throughout the body. There was no preferential accumulation of particles in the tumor tissue/region at this time point, and the nanoprobe was distributed uniformly throughout the body. However, at 24 h, we could differentiate between tumor area and other parts of the body because of the enhanced fluorescence signal from the tumor region. We continued the studies to the next day, and the third reading was taken at the 48 h time point. The fluorescence signal intensity from the tumor area further increased, but the fluorescence signals from other parts of the body remained the same as previous time points. The studies were continued for another two days; the fluorescence signal at the 72 h time point started to decrease gradually and attained the base level at the 96 h time point (Figure 3a). Another interesting feature is that the fluorescence signal from a nontumor area remained the same during the entire study period (Figure 3b). In addition to that, we did not observe any adverse side effects in the animals due to nanoparticle toxicity and they were healthy throughout the study period. The above studies reiterate the point that these nanoprobes have good bioavailability and there are no toxicity issues associated with these delivery vehicles.


Figure 5. Intracellular uptake studies of nanoprobes in PC3 cell line: (A) PEGylated nanoprobe 1, size ∼120 nm; (B) PEGylated nanoprobe 2, size ∼73 nm; (C) naked nanoprobe (without PEG coating), size ∼106 nm; (D) control nanoprobe (empty liposomes), size ∼106 nm.

Intravital Microscopy Studies. The results from in ViVo studies gave us only the macroscopic picture of the targeting nature of our nanoprobe toward tumor tissue but did not reveal information about the localization of our probe inside tumor cells. In order to look more closely at the cellular level, we have used intravital microscopy (IVM) for our studies. IVM has the ability to provide insight into various molecular and cellular processes in ViVo with high spatial and temporal

Sandanaraj et al.

resolution down to the subcellular level (40). More recently, IVM was used to study the architecture and physiology of tumor neovasculature (41). For our studies, nude mice were implanted with DiI-labeled (green channel) PC-3 cells 2 weeks before nanoprobe infusion. The study involved 6 mice total, 2 mice/ group, 3 time points (20 min, 6 h, and 24 h). The blood vessels were stained with FITC-labeled dextran (blue channel). Cy5.5labeled nanoprobe (red channel) was injected in the tail vein and measurements were taken at time points similar to tumor studies in order to compare the results. At the 20 min time point, the fluorescent nanoprobe did not extravasate from intact blood vessels to the tumor tissue. After 6 h, Cy5.5-labeled nanoprobes were taken up by some of the tumor cells, which is in contrast to the in ViVo imaging study results where fluorescence signal from the tumor area remained the same as the background signal (Figure 4). The inherent difference between fluorescence signal intensities from these two methods is attributed to the high sensitivity of IVM. However, the IVM results at the 24 h time point resembled in ViVo studies in which fluorescent particles are retained in the tumor region for extended period. The colocalization of the nanoprobes and tumor cells at the 24 h time point suggests that the nanoprobe might be internalized by the tumor cells. In summary, the biodistribution and pharmacokinetics trends of liposome particles in in ViVo studies are supported by the intravital microscopy results. Intracellular Uptake of Nanoprobes in Tumor Cells. The results from intravital microscopy results suggest that nanoprobes might be internalized by tumor cells, but it is not definitive evidence. Therefore, we have focused our attention on in Vitro cellular characterization of the nanoprobes through fluorescence microscopy. The uptake of custom-designed nanoparticles or microparticles inside a living cell depend on size, shape, and the particle surface potential (42, 43). In a classical example, DeSimone et al. showed that positively charged microparticles were taken up by the live cells more readily compared to negatively charged particles. In our studies, we

Figure 6. (A) NIRF images for arthritis studies were recorded 6 h after i.v. injection of the ICG nanoprobe (size ∼120 nm, ICG concentration ) 151 µg/mL). (B) Effect of drug treatment. Each data point represents the mean ( SEM of 10 animals.


have used PEG-coated nanoprobes to increase the circulation time and bioavailability of the fluorescent cargo. In order to demonstrate that PEG-coated nanoprobes have the same ability to get into the cells as nanoprobes without PEG coating, we incubated PEG-coated nanoprobes and naked nanoprobes with PC3 cells, and the particle internalization was analyzed. Results from fluorescence microscopy show that both particles were indeed internalized by the PC3 cells irrespective of their surface charge. We also studied the effect of particle size on the internalization process; however, it does not have any effect. The reason for the above phenomenon may be the small difference in the nanoparticle size (73 nm vs 120 nm) that we tested in this study (Figure 5). Inflammation Imaging. Encouraged by the results in the tumor model, we wanted to extend this methodology to another disease model. We have chosen an arthritis model, antigeninduced arthritis (AIA), because of the presence of leaky vasculature in arthritic tissue (44, 45). We encapsulated indocyanine green (ICG), a clinically approved fluorescent dye (36), inside the nanoprobe. Arthritis was induced in the right knee of mice by the injection of mBSA into the joint in mice previously sensitized to mBSA systemically. The nanoprobes were injected i.v. via the tail vein and the NIRF measurements taken at different time points (Figure 6). There was no preferential accumulation of particles in the arthritis joints at initial time points. However, at the 24 h time point, we saw 2-fold increases in fluorescence signal from arthritic joint compared to the nonarthritic joint. These studies indicate that the nanoparticles can passively target leaky vasculature and can be trapped in arthritic joints for a longer time. The extension of this methodology in an arthritic disease model has shown the versatile nature of this methodology. The effects of drugs on vascular leakiness were subsequently investigated using fluorescent nanoparticles. Animals were treated orally with vehicle or dexamethasone at a dose of 0.3 mg/kg. Dexamethasone is an anti-inflammatory steroid from a class of drugs frequently used to treat human rheumatoid arthritis (RA) (46–48). Nanoprobes were injected i.v. after the administration of dexamethasone or vehicle, and the fluorescence signal was measured for these two groups of animals. The fluorescence signal from arthritic joint of the dexamethasone-treated animals was similar to the signal from nonarthritic joints. In contrast, the signals from the arthritic joints of vehicle-treated animals were 2-fold higher than the nonarthritic joints. This result indicates that treatment with dexamethasone can inhibit the vascular leakiness associated with the inflammation in the arthritic joints. The above results indicate that this method could be used as a tool to screen the efficacy of anti-inflammatory compounds in preclinical settings. Preclinical and Clinical Implications of the Present Findings. The immediate question that one might ask is, how is this biomarker different from existing optical biomarkers for the detection of vascular leakiness? How could it be used to identify different stages of cancer or arthritis? Is it possible to translate this biomarker from preclinical to clinical use? Use of near-infrared fluorescent probes for the detection of arthritis has previously been reported in the literature (29, 49). However, these small molecule fluorescent probes suffer from severe limitations that arise due to poor pharmacokinetics and biodistribution properties (50). In addition, these probes nonspecifically bind to various blood proteins (51). Fluorophorelabeled antibodies were also explored for these kinds of applications (49). However, no specific biomarker for arthritis is known. The significance of our finding is the elegant combination of delivery vehicle (34) (PEGylated liposomes) and the fluorescent probe (ICG), as both are clinically approved for use in humans for different applications. By encapsulating ICG


inside liposomes, the half-life time of this probe is increased from minutes (50) to several hours. Second, the size range of liposome particles (100-200 nm) favors the targeting efficiency of these particles toward leaky vasculature. Out of all, the most important finding of this report is the use of an optical method to screen anti-inflammatory drugs in in ViVo conditions in preclinical settings. The utility of this method goes beyond the present findings. This tool also could be used to detect at least two stages of antigen-induced arthritis in mouse model, since the vascular leakiness drastically differs between acute and chronic phase of inflammation (52). The extravasation of fluorescent liposomes during acute-phase inflammation is expected to be higher compared to chronic phase, and therefore, the NIRF signal during acute phase should be higher compared to chronic phase. With respect to tumor imaging, the limitation of this probe is the detection of tumors with sizes less than 1-2 mm3, as neovascularization starts only after tumor reaches size of about ∼2 mm3, which results in leaky vasculature of tumor tissue (53–56). On the positive side, this method could be used as a tool to screen the efficacy of various antiangiogenic agents in various tumor models based on the notion that the treatment with antiangiogenic compounds should result in normalization of abnormal blood vessels that should result in decreased extravasation of fluorescent liposomes to tumor tissue (57).

CONCLUSION We have shown the design, synthesis, and characterization of fluorescent nanoprobes. Stability studies of nanoprobe solutions show that these particles are stable for long periods of time. These nanoparticles are decorated with poly(ethylene glycol) for the enhanced circulation and passive targeting of tumor/arthritis tissue. Apart from the validation of probes with near-infrared fluorescence imaging studies, the probes have also been validated in intravital and fluorescence microscopy studies. Results from our studies show that optical imaging combined with appropriate nanoparticles could be used as functional biomarkers for the presence of increased vascular leakiness in pathological conditions. The concept behind the application of this new technology is demonstrated in two different disease models.

ACKNOWLEDGMENT We thank Mr. Fritz Wenger and Ms. Alexandra Suter for preparing the animals for the in ViVo imaging studies. We also thank members of Gabriel Lab for their support and assistance. This work was funded by Novartis Institutes for Biomedical Research, Education Office, and Global Imaging Group. Supporting Information Available: Fluorescence spectra of Cy5.5 in liposome and PBS. This material is available free of charge via the Internet at http://pubs.acs.org.

LITERATURE CITED (1) Severino, M. E., Dubose, R. F., and Patterson, S. D. (2006) A strategic view on the use of pharmacodynamic biomarkers in early clinical drug development. IDrugs. 9, 849–853. (2) Kuhlmann, J., and Wensing, G. (2006) The applications of biomarkers in early clinical drug development to improve decision-making processes. Curr. Clin. Pharmacol. 1, 185–191. (3) Bakhtiar, R. (2008) Biomarkers in drug discovery and development. J. Pharmacol. Toxicol. Methods 57, 85–91.

100 Bioconjugate Chem., Vol. 21, No. 1, 2010 (4) Frank, R., and Hargreaves, R. (2003) Clinical biomarkers in drug discovery and development. Nat. ReV. Drug DiscoVery 2, 566–580. (5) Colburn, W. A. (2003) Biomarkers in drug discovery and development: from target identification through drug marketing. J. Clin. Pharmacol. 43, 329–341. (6) Colburn, W. A. (2000) Optimizing the use of biomarkers, surrogate endpoints, and clinical endpoints for more efficient drug development. J. Clin. Pharmacol. 40, 1419–1427. (7) Hargreaves, R. J. (2008) The role of molecular imaging in drug discovery and development. Clin. Pharmacol. Ther. 83, 349– 353. (8) Rudin, M. (2009) Noninvasive structural, functional, and molecular imaging in drug development. Curr. Opin. Chem. Biol. 13, 360–371. (9) Rudin, M. (2007) Imaging readouts as biomarkers or surrogate parameters for the assessment of therapeutic interventions. Eur. Radiol. 17, 2441–2457. (10) Rudin, M., and Weissleder, R. (2003) Molecular imaging in drug discovery and development. Nat. ReV. Drug DiscoVery 2, 123–131. (11) Maeda, H., Fang, J., Inutsuka, T., and Kitamoto, Y. (2003) Vascular permeability enhancement in solid tumor: various factors, mechanisms involved and its implications. Int. Immunopharmacol. 3, 319–328. (12) Andersson, S. E., Lexmuller, K., and Ekstrom, G. M. (1998) Physiological characterization of mBSA antigen induced arthritis in the rat. I. Vascular leakiness and pannus growth. J. Rheumatol. 25, 1772–1777. (13) Andersson, S. E., and Ekstrom, G. M. (1997) Changes in vascular porosity and joint blood flow during development of collagen induced arthritis in the rat. Modulation by indomethacin and L-NAME. J. Rheumatol. 24, 2188–2195. (14) Medarova, Z., Castillo, G., Dai, G., Bolotin, E., Bogdanov, A., and Moore, A. (2007) Noninvasive magnetic resonance imaging of microvascular changes in type 1 diabetes. Diabetes 56, 2677–2682. (15) De Paepe, M. E., Corriveau, M., Tannous, W. N., Seemayer, T. A., and Colle, E. (1992) Increased vascular permeability in pancreas of diabetic rats: detection with high resolution protein A-gold cytochemistry. Diabetologia 35, 1118–1124. (16) Muruganandham, M., Lupu, M., Dyke, J. P., Matei, C., Linn, M., Packman, K., Kolinsky, K., Higgins, B., and Koutcher, J. A. (2006) Preclinical evaluation of tumor microvascular response to a novel antiangiogenic/antitumor agent RO0281501 by dynamic contrast-enhanced MRI at 1.5 T. Mol. Cancer Ther. 5, 1950–1957. (17) Demsar, F., van Dijke, C. F., Kirk, B. A., Kapila, S., Peterfy, C. G., Roberts, T. P., Shames, D. M., Tomazic, S., Mann, J., and Brasch, R. C. (1996) Mapping abnormal synovial vascular permeability in temporomandibular joint arthritis in the rabbit using MRI. Br. J. Rheumatol. 35, 23–25, Suppl 3. (18) Kaplan, J. D., Trulock, E. P., Anderson, D. J., and Schuster, D. P. (1992) Pulmonary vascular permeability in interstitial lung disease. A positron emission tomographic study. Am. ReV. Respir. Dis. 145, 1495–1498. (19) Kruger, E. F., Puchalski, S. M., Pollard, R. E., Galuppo, L. D., Hornof, W. J., and Wisner, E. R. (2008) Measurement of equine laminar blood flow and vascular permeability by use of dynamic contrast-enhanced computed tomography. Am. J. Vet. Res. 69, 371–377. (20) Roberts, H. C., Roberts, T. P., Lee, T. Y., and Dillon, W. P. (2002) Dynamic, contrast-enhanced CT of human brain tumors: quantitative assessment of blood volume, blood flow, and microvascular permeability: report of two cases. Am. J. Neuroradiol. 23, 828–832. (21) Terada, T., Nambu, K., Hyotani, G., Miyamoto, K., Tsuura, M., Nakamura, Y., Nishiguchi, T., Itakura, T., Hayashi, S., and Komai, N. (1992) A method for quantitative measurement of cerebral vascular permeability using X-ray CT and iodinated contrast medium. Neuroradiology 34, 290–296.

Sandanaraj et al. (22) Heppner, P., and Lindner, J. R. (2005) Contrast ultrasound assessment of angiogenesis by perfusion and molecular imaging. Expert ReV. Mol. Diagn. 5, 447–455. (23) Jaffer, F. A., Kim, D. E., Quinti, L., Tung, C. H., Aikawa, E., Pande, A. N., Kohler, R. H., Shi, G. P., Libby, P., and Weissleder, R. (2007) Optical visualization of cathepsin K activity in atherosclerosis with a novel, protease-activatable fluorescence sensor. Circulation 115, 2292–2298. (24) Pande, A. N., Kohler, R. H., Aikawa, E., Weissleder, R., and Jaffer, F. A. (2006) Detection of macrophage activity in atherosclerosis in vivo using multichannel, high-resolution laser scanning fluorescence microscopy. J. Biomed. Opt. 11, 021009. (25) Wunder, A., Tung, C. H., Muller-Ladner, U., Weissleder, R., and Mahmood, U. (2004) In vivo imaging of protease activity in arthritis: a novel approach for monitoring treatment response. Arthritis Rheum. 50, 2459–2465. (26) Weissleder, R., and Ntziachristos, V. (2003) Shedding light onto live molecular targets. Nat. Med. 9, 123–128. (27) Petrovsky, A., Schellenberger, E., Josephson, L., Weissleder, R., and Bogdanov, A., Jr. (2003) Near-infrared fluorescent imaging of tumor apoptosis. Cancer Res. 63, 1936–1942. (28) Weissleder, R., Tung, C. H., Mahmood, U., and Bogdanov, A., Jr. (1999) In vivo imaging of tumors with protease-activated near-infrared fluorescent probes. Nat. Biotechnol. 17, 375–378. (29) Hansch, A., Frey, O., Hilger, I., Sauner, D., Haas, M., Schmidt, D., Kurrat, C., Gajda, M., Malich, A., Brauer, R., and Kaiser, W. A. (2004) Diagnosis of arthritis using near-infrared fluorochrome Cy5.5. InVest. Radiol. 39, 626–632. (30) Hickson, J. (2009) In vivo optical imaging: preclinical applications and considerations. Urol. Oncol. 27, 295–297. (31) Wong, D. F. (2008) Imaging in drug discovery, preclinical, and early clinical development. J. Nucl. Med. 49, 26N–28N. (32) Kiesslich, R., Gossner, L., Goetz, M., Dahlmann, A., Vieth, M., Stolte, M., Hoffman, A., Jung, M., Nafe, B., Galle, P. R., and Neurath, M. F. (2006) In vivo histology of Barrett’s esophagus and associated neoplasia by confocal laser endomicroscopy. Clin. Gastroenterol. Hepatol. 4, 979–987. (33) Kiesslich, R., Burg, J., Vieth, M., Gnaendiger, J., Enders, M., Delaney, P., Polglase, A., McLaren, W., Janell, D., Thomas, S., Nafe, B., Galle, P. R., and Neurath, M. F. (2004) Confocal laser endoscopy for diagnosing intraepithelial neoplasias and colorectal cancer in vivo. Gastroenterology 127, 706–713. (34) Torchilin, V. P. (2005) Recent advances with liposomes as pharmaceutical carriers. Nat. ReV. Drug DiscoVery 4, 145–160. (35) Zhu, L., Zheng, Y., von Kerczek, C. H., Topoleski, L. D., and Flower, R. W. (2006) Feasibility of extracting velocity distribution in choriocapillaris in human eyes from ICG dye angiograms. J. Biomech. Eng. 128, 203–209. (36) Ishigami, Y., Masuzawa, M., Miyoshi, E., Kato, M., Tamura, K., Kanda, M., Awazu, K., Taniguchi, K., Kurita, M., and Hayashi, N. (1993) Clinical applications of ICG Finger Monitor in patients with liver disease. J. Hepatol. 19, 232–240. (37) Klibanov, A. L., Maruyama, K., Torchilin, V. P., and Huang, L. (1990) Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Lett. 268, 235–237. (38) Maeda, H., Sawa, T., and Konno, T. (2001) Mechanism of tumor-targeted delivery of macromolecular drugs, including the EPR effect in solid tumor and clinical overview of the prototype polymeric drug SMANCS. J. Controlled Release 74, 47–61. (39) Harasym, T. O., Tardi, P., Longman, S. A., Ansell, S. M., Bally, M. B., Cullis, P. R., and Choi, L. S. (1995) Poly(ethylene glycol)-modified phospholipids prevent aggregation during covalent conjugation of proteins to liposomes. Bioconjugate Chem. 6, 187–194. (40) Jain, R. K., Schlenger, K., Hockel, M., and Yuan, F. (1997) Quantitative angiogenesis assays: progress and problems. Nat. Med. 3, 1203–1208. (41) Brown, E. B., Campbell, R. B., Tsuzuki, Y., Xu, L., Carmeliet, P., Fukumura, D., and Jain, R. K. (2001) In vivo measurement of gene expression, angiogenesis and physiological function in

Fluorescent Nanoprobes as Biomarker tumors using multiphoton laser scanning microscopy. Nat. Med. 7, 864–868. (42) Gratton, S. E., Napier, M. E., Ropp, P. A., Tian, S., and DeSimone, J. M. (2008) Microfabricated particles for engineered drug therapies: elucidation into the mechanisms of cellular internalization of PRINT particles. Pharm. Res. 25, 2845–2852. (43) Gratton, S. E., Ropp, P. A., Pohlhaus, P. D., Luft, J. C., Madden, V. J., Napier, M. E., and DeSimone, J. M. (2008) The effect of particle design on cellular internalization pathways. Proc. Natl. Acad. Sci. U.S.A. 105, 11613–11618. (44) Levick, J. R. (1981) Permeability of rheumatoid and normal human synovium to specific plasma proteins. Arthritis Rheum. 24, 1550–1560. (45) Kushner, I., and Somerville, J. A. (1971) Permeability of human synovial membrane to plasma proteins. Relationship to molecular size and inflammation. Arthritis Rheum. 14, 560–570. (46) Yamaki, K., Uchida, H., Li, X., Yanagisawa, R., Takano, H., Hayashi, H., Mori, Y., and Yoshino, S. (2005) Effect of varying types of anti-arthritic drugs on Th1 and Th2 immune responses in mice. Int. J. Immunopathol. Pharmacol. 18, 133–144. (47) Takeshita, M., Sugita, T., and Takata, I. (1997) Pathological evaluation of effect of anti-rheumatic drugs on type II collageninduced arthritis in Lewis rats. Exp. Anim. 46, 165–169. (48) Ueno, M., Imaizumi, K., Sugita, T., Takata, I., and Takeshita, M. (1995) Effect of a novel anti-rheumatic drug, TA-383, on type II collagen-induced arthritis. Int. J. Immunopharmacol. 17, 597–603.

Bioconjugate Chem., Vol. 21, No. 1, 2010 101 (49) Hansch, A., Frey, O., Sauner, D., Hilger, I., Haas, M., Malich, A., Brauer, R., and Kaiser, W. A. (2004) In vivo imaging of experimental arthritis with near-infrared fluorescence. Arthritis Rheum. 50, 961–967. (50) Desmettre, T., Devoisselle, J. M., and Mordon, S. (2000) Fluorescence properties and metabolic features of indocyanine green (ICG) as related to angiography. SurV. Ophthalmol. 45, 15–27. (51) Mordon, S., Devoisselle, J. M., Soulie-Begu, S., and Desmettre, T. (1998) Indocyanine green: physicochemical factors affecting its fluorescence in vivo. MicroVasc. Res. 55, 146–152. (52) Hansch, A., Sauner, D., Hilger, I., Frey, O., Haas, M., Malich, A., Brauer, R., and Kaiser, W. A. (2003) Noninvasive diagnosis of arthritis by autofluorescence. InVest. Radiol. 38, 578–583. (53) Naumov, G. N., Folkman, J., and Straume, O. (2009) Tumor dormancy due to failure of angiogenesis: role of the microenvironment. Clin. Exp. Metastasis 26, 51–60. (54) Naumov, G. N., Folkman, J., Straume, O., and Akslen, L. A. (2008) Tumor-vascular interactions and tumor dormancy. APMIS 116, 569–585. (55) Naumov, G. N., Akslen, L. A., and Folkman, J. (2006) Role of angiogenesis in human tumor dormancy: animal models of the angiogenic switch. Cell Cycle 5, 1779–1787. (56) Folkman, J. (2002) Role of angiogenesis in tumor growth and metastasis. Semin. Oncol. 29, 15–18. (57) Jain, R. K. (2005) Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 307, 58–62. BC900311H

Fluorescent Nanoprobes as a Biomarker for Increased Vascular

Recommend Documents