- Email: [email protected]
Targeted optical imaging of cancer cells using lectin-binding BODIPY conjugated avidin
BBRC Biochemical and Biophysical Research Communications 348 (2006) 807–813 www.elsevier.com/locate/ybbrc
Targeted optical imaging of cancer cells using lectin-binding BODIPY conjugated avidin Yukihiro Hama a, Yasuteru Urano b, Yoshinori Koyama a, Peter L. Choyke a, Hisataka Kobayashi a,* a
Molecular Imaging Program, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, MD 20892-1088, USA b Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Received 19 June 2006 Available online 4 August 2006
Abstract Lectins (asialo-receptor family) are expressed on a number of tumors that develop peritoneal metastases. To demonstrate that fluorescence imaging based on lectin binding is applicable for a variety of tumors, we conjugated BODIPY to avidin (avidin–BODIPY), and studied the efficacy of tumor targeting in 9 cancer cell lines in vitro and an ovarian cancer cell line in vivo using a murine peritoneal cancer model. All 9 cell lines showed specific intracellular accumulation with avidin–BODIPY on fluorescence microscopy and flow cytometry. In vivo spectral molecular imaging clearly visualized the peritoneal tumor foci with avidin–BODIPY, whereas, deglycosylated avidin– BODIPY (neutravidin–BODIPY) showed only minimal fluorescence from the tumor foci and was accompanied by higher background signals. These results suggest the lectin-targeted molecular imaging technique using a targeted green fluorescence probe is potentially useful in a wide variety of cancers with a proclivity for dissemination in the peritoneal space. 2006 Elsevier Inc. All rights reserved. Keywords:
D-Galactose
receptor; Molecular imaging; Neoplasms; Peritoneal metastasis; Surgery; Flow cytometry; Optical imaging; Endoscopy
Peritoneal dissemination of cancer is difficult to treat and is often fatal. During surgery, it is difficult to detect small imbedded tumors in the peritoneum due to its complex anatomy and the poor visual contrast between the tumor and normal tissue using white light imaging. To detect small peritoneal implants with higher accuracy, several targeted and non-targeted detection methods have been proposed which utilize either radiolabeled probes [1,2] or optical probes [3,4]. Targeted detection methods based on radiolabeled probes have been hampered by poor image resolution, the necessity of gamma cameras in the operating room and ionizing radiation exposure to patients and operators [1,2]. Recently, a lectin (asialo-receptor family)-targeted in vivo optical imaging technique has been proposed which utilizes a fluorescein-avidin conjugate
*
Corresponding author. E-mail address: [email protected] (H. Kobayashi).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.07.169
which enables the visualization of submillimeter peritoneal implants in a mouse ovarian cancer model [4]. Avidin is a 68 kDa tetrameric glycoprotein and binds with high capacity to lectins which are expressed on the surface of many cancer cells [1,2,4,5]. Lectins are proteins that specifically bind to carbohydrate groups on glycolipids and glycoproteins, such as avidin and galactosylated proteins [1,2,4,5]. Since avidin binds to lectin or other lectinlike negatively charged molecules at physiological pH and bound avidin is subsequently internalized into cancer cells it is an excellent targeting moiety for optical imaging [4]. In addition, when avidin is administered into the peritoneal cavity, unbound avidin is absorbed through the peritoneal membrane, enters the circulation, and is quickly trapped by the liver removing it from the imaging field [1,4,6]. Thus, avidin conjugated with optical probe is capable of targeted molecular imaging with high target-to-background ratios as well as high spatial resolution [4]. However, this targeting method for peritoneal cancer implants has been studied
808
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
in only two cancer cell lines without satisfactory in vitro cellular basis or in vivo imaging analyses for individual cells [1,2,4,7]. Moreover, the previously utilized fluorophore, fluorescein, while possessing some advantages in cost and availability, is not as stable as other fluorophores in the intracellular environment. A fluorophore 4,4-difluoro5-phenyl-4-bora-3a,4a-diaza-s-indacene-3-propionic acid, succinimidyl ester (BODIPY-R6G), which is a variant of BODIPY dyes with a little longer wavelength emission than 4,4-difluoro-5,7-dimethyl-4-bora-3a,4a- diaza-s-indacene-3-propionic acid, succinimidyl ester (BODIPY-FL), is stable in acidic environments and demonstrates an excellent quantum yield even after internalization into cancer cells [8] (Supplementary Fig. 1). Thus, the purpose of this study was to investigate the availability of lectin-binding as a target for a variety of cancer cell lines, all of which are associated with peritoneal dissemination using an avidin–fluorophore conjugate, Avidin–BODIPY. Materials and methods Synthesis of avidin–BODIPY and neutravidin–BODIPY conjugates. Avidin and neutravidin were purchased from Pierce Biochemical Inc. (Milwaukee, WI, USA). BODIPY-R6G was purchased from Molecular Probes Inc. (Eugene, OR, USA). We made BODIPY-conjugated avidin (avidin–BODIPY) and BODIPY-conjugated neutravidin (neutravidin– BODIPY). At room temperature, 500 lg (7.3 nmol for avidin and 8.3 nmol for neutravidin) of avidin or neutravidin in 200 lL of 0.1 M
Na2HPO4 was incubated with 8.7 lg (20 nmol) of BODIPY in DMSO for 15 min. The mixture was purified with Sephadex G50 (PD-10, GE Healthcare, Milwaukee, WI, USA). The protein concentration of each sample was determined with Coomassie Plus protein assay kit (Pierce Chem Co., Rockford, IL, USA) by measuring the absorption at 595 nm with a UV-Vis system (8453 Value UV-Bis system, Agilent Technologies, Palo Alto, CA, USA). Then, the BODIPY concentration was measured by the absorption at 534 nm with a UV–vis system (8453 Value UV-Bis system, Agilent Technologies, Palo Alto, CA, USA) to confirm the number of BODIPY molecules conjugated with each avidin or neutravidin molecule. For flow cytometry and fluorescence microscopy, the number of BODIPY molecules per avidin was 2 and the number of BODIPY molecules per neutravidin was 2. For in vivo study, the concentration of avidin or neutravidin was changed to produce the same fluorescent intensity from avidin–BODIPY and neutravidin–BODIPY. The number of BODIPY conjugated with avidin was 1.4 and that with neutravidin 1.7. Cell culture. An established human ovarian cancer cell line SHIN3 [7] as well as eight commercially available cancer cell lines (OVCAR-3, SKOV-3, PANC-1, Capan-2, MCF7, PC-3, HT-29, N87: ATCC, Manassas, VA USA) were used. SHIN3, OVCAR-3, and SKOV-3 were from human ovarian cancer, PANC-1 and Capan-2 from human pancreatic cancer, HT-29 from human colon cancer, N87 from human gastric cancer, MCF7 from human breast cancer, and PC-3 from human prostate cancer. SHIN3, OVCAR-3, MCF7, PC-3, HT-29, and N87 were grown in RPMI 1640 medium (Gibco, Gaithersburg, MD, USA), PANC-1 in DMEM (Gibco, Gaithersburg, MD, USA), Capan-2 and SKOV-3 in McCOY’s 5A medium (Quality Biological Inc., Gaithersburg, MD, USA) containing 10% fetal bovine serum (FBS) (Gibco, Gaithersburg, MD, USA), 0.03% L-glutamine at 37 C, 100 U/mL penicillin and 100 lg/mL streptomycin in 5% CO2. Fluorescence microscopy. The 9 cancer cells (5 · 105) were plated on a coverglass bottom culture well and incubated for 24 h. Avidin–BODIPY
Fig. 1. In vitro fluorescence microscopy (left) and differential interference contrast (right) images of 9 cancer cell lines. Although there are some differences in number, size, and intensity of fluorescent dots, these intracellular fluorescent dots are detected in all 9 cell lines after 4 h incubation with 30 lg/ml avidin–BODIPY.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813 was added to the medium (30 lg/ml) and the cells were incubated for another 4 h. Cells were washed one time with PBS. Fluorescence microscopy was performed using an Olympus BX51 microscope (Olympus America Inc., Melville, NY, USA) equipped with the following filters: excitation wavelength 470–490 nm, emission wavelength 515 nm long pass. Transmitted light differential interference contrast (DIC) images were also acquired. Flow cytometry. To test whether the avidin–BODIPY binds specifically to lectins on cancer cells through the glycosylation of avidin, one-color flow cytometry of avidin–BODIPY and deglycosylated avidin–BODIPY (neutravidin–BODIPY) was performed. Cells were incubated for 16 h and treated with 10 lg/ml avidin–BODIPY or neutravidin–BODIPY for 96 h. The cells were washed once with PBS, trypsinized, and flow cytometry was performed using a FACScan cytometer (Becton–Dickinson, Franklin Lakes, NJ, USA). The argon ion 488 nm laser was employed for excitation. Signals from cells were collected using a 530/30 nm band-pass filter. Control cells used in the flow cytometry were the same cell line that had not been incubated with avidin–BODIPY or neutravidin–BODIPY. Fluorescence histograms (in FL-1 channel) were created and the percentage of positive cells was analyzed using Cell Quest software (Becton–Dickinson, Franklin Lakes, NJ, USA). The percentage of fluorescence-gated positive cells which corresponds to an increase of fluorescence (M1) is indicated for each experimental condition.
809
Tumor model. All animal experiments were carried out in compliance with the Guide for the Care and Use of Laboratory Animal Resources (1996), National Research Council, and approved by the National Cancer Institute Animal Care and Use Committee. The intraperitoneal tumor xenografts were established by intraperitoneal injection of 2 · 106 OVCAR-3 cells suspended in 200 lL PBS in female nude mice (National Cancer Institute Animal Production Facility, Frederick, MD, USA). Experiments with tumor-bearing mice were performed at 28 days for OVCAR3 after injection of the cells. In vivo fluorescence imaging. Three hours after intraperitoneal injection of 50 lg avidin–BODIPY or 50 lg neutravidin–BODIPY diluted in 300 lL PBS into tumor-bearing mice, the mice were sacrificed with CO2 gas. The abdominal walls were removed and spectral fluorescence images were obtained using the Maestro In Vivo Imaging System (CRi Inc., Woburn, MA, USA). For direct comparison of fluorescence intensity between avidin–BODIPY and neutravidin–BODIPY, the number of BODIPY molecules conjugated to avidin or neutravidin was adjusted to obtain the same fluorescence intensity prior to the intraperitoneal administration. In addition, two mice or two extracted tumors either with avidin–BODIPY or with neutravidin–BODIPY were placed side-by-side to directly compare the fluorescence intensity of peritoneal cancer implants. Semiquantitative ex vivo comparison of fluorescence intensity of two tumors was performed. A region of interest (ROI) as large as each
Fig. 2. Flow cytometry of 9 cancer cell lines incubated with avidin–BODIPY (green) or neutravidin–BODIPY (red). Cells were incubated for 16 h followed by treatment with avidin–BODIPY or neutravidin–BODIPY for 96 h. Then the cells were washed, trypsinized, and analyzed for flow cytometry. The horizontal axis represents the fluorescence intensities of each cell line. The vertical axis represents the number of cells. The percentage of avidin– BODIPY- or neutravidin–BODIPY-accumulated cells which corresponds to an increase of fluorescence (M1) is indicated for each experimental condition. Control cells (black) were the same cell line that had not been incubated with avidin–BODIPY or neutravidin–BODIPY.
810
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
tumor was drawn to determine the fluorescence intensity as well as the histogram using ImageJ software (http://rsb.info.nih.gov/ij/plugins/ mri-analysis.html). The dynamic range of the fluorescent intensity in arbitrary unit (au) was split into equal-sized 256 bins (0–255). A band pass filter from 480 to 520 nm and a long pass filter over 550 nm were used for excitation and emission light, respectively. The tunable filter was automatically stepped in 10 nm increments from 500 to 800 nm while the camera captured images at each wavelength interval with constant exposure. Spectral unmixing algorithms were applied to create the unmixed images of BODIPY and autofluorescence. Three mice were studied in each group and all images were analyzed using commercial Maestro software (Nuance ver. 1.4, CRi Inc.).
Results Avidin-BODIPY internalized in all 9 cancer cell lines To confirm that avidin-BODIPY binds to cancer cells and internalizes into them, in vitro fluorescence microscopy as well as DIC images were obtained. Fluorescence microscopy of cultured cells clearly showed a large number of fluorescent dots within the cytoplasm (Fig. 1), indicating that the avidin–BODIPY internalized and located within the endoplasmic vesicles or lysosomes and continued to remain fluorescent. These findings are compatible to the previous studies performed by fluorescein-conjugated avidin for SHIN3 cancer cells [4], although the fluorescence appears more intense and of longer duration using the avidin–BODIPY conjugate.
of OVCAR-3 ovarian cancer was performed. This cell line was selected because OVCAR3 showed the maximal rightward shift in respective cancers after treatment with avidin–BODIPY in the flow cytometry analysis (Fig. 2). Before comparing the in vivo fluorescence intensity of the peritoneal implants produced by avidin–BODIPY and neutravidin–BODIPY, pre-injection fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY probes were adjusted to be the same by changing the number of BODIPY molecules per avidin or neutravidin molecule. The number of BODIPY molecules conjugated with avidin and neutravidin were 1.4 and 1.7, respectively. Signal intensities as well as the emission spectra of 8 lg avidin–BODIPY and 8 lg neutravidin–BODIPY in 390 lL PBS were comparable (Fig. 3). Using these two probes, in vivo spectral fluorescence imaging was performed 3 h after intraperitoneal injection of 50 lg avidin–BODIPY or 50 lg neutravidin–BODIPY. Since the fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY prior to the injection were the same, side-by-side comparison of the imaging was possible. The fluorescence intensity of the tumor nodules on the spectral fluorescence image was
Avidin–BODIPY is specifically bound and internalized into cancer cells via glycosylation To study whether the avidin–BODIPY is bound specifically to lectins on cancer cells through glycosylation, one-color flow cytometry of avidin–BODIPY and deglycosylated avidin–BODIPY (neutravidin–BODIPY) was performed. The percentage of fluorescence-gated cells which corresponds to internalization of avidin–BODIPY or neutravidin–BODIPY is shown in Fig. 2. For avidin–BODIPY, the percentages of SHIN3, OVCAR-3, SKOV-3, PANC-1, Capan-2, HT-29, N87 MCF7, and PC-3, increased from 2.3%, 3.7%, 0.8%, 2.3%, 3.0%, 1.4%, 1.2%, 1.7%, and 1.2% to 99.9%, 99.8%, 99.7%, 74.2%, 96.7%, 99.5%, 98.4%, 99.9%, and 99.9%, respectively. Whereas, when treated with neutravidin–BODIPY, the percentages of SHIN3, OVCAR-3, SKOV-3, PANC-1, Capan-2, HT-29, N87, MCF7, and PC-3 were 3.5%, 4.4%, 11.0%, 3.5%, 5.3%, 6.1%, 1.5%, 4.8%, and 2.1%, respectively. These results indicate that the avidin–BODIPY will bind specifically to lectins on a wide variety of cancer cells through the glycosylation of avidin. Avidin–BODIPY specifically targeted the peritoneal cancer foci in vivo To investigate the efficacy and the applicability of avidin–BODIPY to target the peritoneal cancer implants, in vivo spectral fluorescence imaging of peritoneal implants
Fig. 3. Optical characteristics of avidin–BODIPY and neutravidin– BODIPY. Eight micrograms of avidin–BODIPY and neutravidin– BODIPY in 390 lL PBS were placed in a nonfluorescent 96-well plate and spectral fluorescence image was obtained. The numbers of BODIPY molecules conjugated with avidin and neutravidin molecule were adjusted to 1.4 and 1.7, respectively, to achieve equal fluorescence of both molecules. (Upper) Fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY were 242 and 243 in arbitrary units (au). (Lower) Emission spectra. These two dyes had the same emission peak at a wavelength of 570 nm and the almost identical fluorescence intensity when concentration was adjusted.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
811
substantially higher with avidin–BODIPY than that with neutravidin–BODIPY (Fig. 4). Conversely, the peri-tumoral background signals with neutravidin–BODIPY were substantially higher than the background signals with avidin–BODIPY due to the unbound free neutravidin–BODIPY reagent within the peritoneal cavity and the absorption of unbound free avidin–BODIPY across the peritoneal membrane. To avoid the normal autofluorescence from the intestine, aggregated tumor nodules were extracted, placed side-by-side on a non-fluorescent plate, and spectral fluorescence imaging was performed. Ex vivo spectral fluorescence image demonstrated that the fluorescence intensity of the tumors treated with avidin–BODIPY was higher than that with neutravidin–BODIPY (Fig. 5A).
Fig. 5. Ex vivo spectral fluorescence image of aggregated tumor nodules instilled with avidin–BODIPY or neutravidin–BODIPY. (A) An ROI was drawn as large as each tumor to compare the fluorescence intensity between avidin–BODIPY and neutravidin–BODIPY. (B) Histogram of fluorescent intensity of an ROI drawn on each of the tumor instilled with avidin–BODIPY or neutravidin–BODIPY. The dynamic range of the fluorescent intensity in arbitrary unit (au) was split into equal-sized 256 bins (0–255). Then for each bin (horizontal axis), the number of pixels from the data set that fall into each bin (vertical axis) is counted. The avidin–BODIPY has a larger proportion of counts for pixels in the higher fluorescent signal intensity range than neutravidin–BODIPY, indicating the fluorescence intensity of the tumor treated with avidin–BODIPY is higher than that of neutravidin–BODIPY.
The histogram of avidin–BODIPY has a larger proportion of counts for pixels in the higher fluorescent intensity range than neutravidin–BODIPY (Fig. 5B). Discussion
Fig. 4. In vivo spectral fluorescence images of tumor-bearing mice 3 h after intraperitoneal injection of avidin–BODIPY and neutravidin–BODIPY. (Upper) Autofluorescence image. (Middle) BODIPY fluorescence image. (Lower) Composite image (red: autofluorescence, green: BODIPY fluorescence). Spectral fluorescence images of the peritoneal cavities clearly visualized the aggregated peritoneal tumor foci (arrows) in mice injected with avidin–BODIPY, whereas, neutravidin–BODIPY showed minimal fluorescence from the tumor foci with higher background signals (arrowheads) due to the free unbound neutravidin–BODIPY reagent in the peritoneal cavity.
These results indicate that avidin–BODIPY binds specifically to cancer cells through lectin–avidin interaction, but unbound free avidin–BODIPY is absorbed through the peritoneal membrane and cleared away from the peritoneal cavity, ultimately resulting in a high signal-to-background ratio. Avidin–BODIPY, successfully targeted a wide variety of cancer cells, all of which have the potential for intraperitoneal dissemination. These results were compatible with the previously reported optical imaging study using fluorescein-conjugated avidin [4], but differ in that the uni-
812
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
versal availability of lectin-targeted optical imaging method was investigated and verified, and because an acid-stable fluorophore, BODIPY, was newly used instead of fluorescein. There are several reasons why we chose BODIPY instead of fluorescein for targeting cancer foci both in vivo and in vitro. First, fluorescein is susceptible to photobleaching. Considering the clinical application of this targeting method in which the agent would constantly be exposed to surgical or endoscopic illumination might easily bleach the fluorescence of fluorescein, and ultimately decrease the signal intensity from the tumor foci. BODIPY is more resistant to photobleaching and relatively stable under visible light exposure. Moreover, BODIPY is stable under acidic conditions, whereas fluorescein is particularly vulnerable to acidic conditions. Even though fluorescein conjugates are internalized into cancer cells and accumulate within the endosomes or lysosomes, the fluorescing capability is reduced by the acidic conditions in endosomes or lysosomes whereas BODIPY conjugates are more resistant to the acidic environment and will continue to fluoresce. Finally, BODIPY can be synthesized as a potent photosensitizer (high extinction coefficient, high photostability, and insensitivity to solvent environment) [9]. By changing the structure of the BODIPY core, this imaging method might be applicable both for detection and for photodynamic therapy of disseminated peritoneal cancers. Malignant peritoneal dissemination has a high rate of clinical recurrence and is often lethal. However, recently there have been increasing number of reports describing a comprehensive multimodal treatment approach that involves cytoreductive surgery and intraperitoneal chemotherapy or photodynamic therapy [10–13]. Cytoreductive surgery aims to maximally eradicate visible tumor nodules, while intraperitoneal chemotherapy and photodynamic therapy aim to treat the remaining microscopic disease in an attempt to prolong disease free survival or even cure selected patients [10–13]. One of the major limitations of cytoreductive surgery is the completeness of the surgical resection, which is often limited by the poor visibility of cancer foci within the complex anatomy of the peritoneal cavity. At the same time, the non-targeted intraperitoneal chemotherapy and photodynamic therapy have a problem of local and systemic toxicity [3,10–13]. To overcome these problems, a targetspecific agent that could be used for detection as well as photodynamic therapy would be welcome. A probe based on the avidin–BODIPY conjugate could offer both diagnostic and therapeutics in a single agent. It should be noted that avidin is immunogenic and is therefore unlikely to be applicable clinically [14]. However, there are numerous agents with similar binding properties that are currently being investigated as substitutes in order to overcome this limitation. In conclusion, a photostable and acid-resistant targeted molecular imaging technique using avidin–BODIPY opti-
cal probe was shown to bind to a wide variety of cancer cell types, many of which are associated with peritoneal metastases. This technique might become a useful tool in detecting and treating small peritoneal cancer implants and subsequently improving the quality of cytoreductive surgery and adjuvant therapy. Acknowledgment This research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc. 2006.07.169. References [1] Z. Yao, M. Zhang, H. Sakahara, T. Saga, Y. Arano, J. Konishi, Avidin targeting of intraperitoneal tumor xenografts, J. Natl. Cancer. Inst. 90 (1998) 25–29. [2] Z. Yao, M. Zhang, H. Sakahara, T. Saga, Y. Nakamoto, N. Sato, S. Zhao, Y. Arano, J. Konishi, Imaging of intraperitoneal tumors with technetium-99m GSA, Ann. Nucl. Med. 12 (1998) 115–118. [3] M.C. Aalders, N. vd Vange, F.A. Stewart, M.G. Klein, M.J. vd Vijver, H.J. Sterenborg, White-light toxicity, resulting from systemically administered 5-aminolevulinic acid, under normal operating conditions, J. Photochem. Photobiol. B 50 (1999) 88– 93. [4] Y. Hama, Y. Urano, Y. Koyama, M. Kamiya, M. Bernardo, R. Paik, M.C. Krishna, P.L. Choyke, H. Kobayashi, In vivo spectral fluorescence imaging of submillimeter peritoneal cancer implants using a lectin-targeted optical agent, Neoplasia 8 (2006) 607–612. [5] R. Lotan, A. Raz, Lectins in cancer cells, Ann. N.Y. Acad. Sci. 551 (1988) 385–388. [6] H. Kobayashi, H. Sakahara, M. Hosono, Z.S. Yao, S. Toyama, K. Endo, J. Konishi, Improved clearance of radiolabeled biotinylated monoclonal antibody following the infusion of avidin as a ‘‘chase’’ without decreased accumulation in the target tumor, J. Nucl. Med. 35 (1994) 1677–1684. [7] S. Imai, Y. Kiyozuka, H. Maeda, T. Noda, H.L. Hosick, Establishment and characterization of a human ovarian serous cystadenocarcinoma cell line that produces the tumor markers CA-125 and tissue polypeptide antigen, Oncology 47 (1990) 177–184. [8] Y. Hama, Y. Urano, Y. Koyama, M. Bernardo, P.L. Choyke, H. Kobayashi, A Comparison of the emission efficiency of four common green fluorescence dyes after internalization into cancer cells, Bioconjugate Chem (submitted for publictaion). [9] T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa, T. Nagano, Highly efficient and photostable photosensitizer based on BODIPY chromophore, J. Am. Chem. Soc. 127 (2005) 12162–12163. [10] T.D. Yan, O.A. Stuart, D. Yoo, P.H. Sugarbaker, Perioperative intraperitoneal chemotherapy for peritoneal surface malignancy, J. Transl. Med. 4 (2006) 17. [11] D.K. Armstrong, B. Bundy, L. Wenzel, H.Q. Huang, R. Baergen, S. Lele, L.J. Copeland, J.L. Walker, R.A. Burger, Gynecologic Oncology Group, Intraperitoneal cisplatin and paclitaxel in ovarian cancer, N. Engl. J. Med. 354 (2006) 34–43.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813 [12] I.J. Torres, C.L. Litterst, A.M. Guarino, Transport of model compounds across the peritoneal membrane in the rat, Pharmacology 17 (1978) 330–340. [13] T.J. Dougherty, C.J. Gomer, B.W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, Photodynamic therapy, J. Natl. Cancer. Inst. 90 (1998) 889–905.
813
[14] V.P. Hytonen, O.H. Laitinen, A. Grapputo, A. Kettunen, J. Savolainen, N. Kalkkinen, A.T. Marttila, H.R. Nordlund, T.K. Nyholm, G. Paganelli, M.S. Kulomaa, Characterization of poultry egg-white avidins and their potential as a tool in pretargeting cancer treatment, Biochem. J. 372 (2003) 219– 225.
Targeted optical imaging of cancer cells using lectin-binding BODIPY conjugated avidin Yukihiro Hama a, Yasuteru Urano b, Yoshinori Koyama a, Peter L. Choyke a, Hisataka Kobayashi a,* a
Molecular Imaging Program, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, MD 20892-1088, USA b Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Received 19 June 2006 Available online 4 August 2006
Abstract Lectins (asialo-receptor family) are expressed on a number of tumors that develop peritoneal metastases. To demonstrate that fluorescence imaging based on lectin binding is applicable for a variety of tumors, we conjugated BODIPY to avidin (avidin–BODIPY), and studied the efficacy of tumor targeting in 9 cancer cell lines in vitro and an ovarian cancer cell line in vivo using a murine peritoneal cancer model. All 9 cell lines showed specific intracellular accumulation with avidin–BODIPY on fluorescence microscopy and flow cytometry. In vivo spectral molecular imaging clearly visualized the peritoneal tumor foci with avidin–BODIPY, whereas, deglycosylated avidin– BODIPY (neutravidin–BODIPY) showed only minimal fluorescence from the tumor foci and was accompanied by higher background signals. These results suggest the lectin-targeted molecular imaging technique using a targeted green fluorescence probe is potentially useful in a wide variety of cancers with a proclivity for dissemination in the peritoneal space. 2006 Elsevier Inc. All rights reserved. Keywords:
D-Galactose
receptor; Molecular imaging; Neoplasms; Peritoneal metastasis; Surgery; Flow cytometry; Optical imaging; Endoscopy
Peritoneal dissemination of cancer is difficult to treat and is often fatal. During surgery, it is difficult to detect small imbedded tumors in the peritoneum due to its complex anatomy and the poor visual contrast between the tumor and normal tissue using white light imaging. To detect small peritoneal implants with higher accuracy, several targeted and non-targeted detection methods have been proposed which utilize either radiolabeled probes [1,2] or optical probes [3,4]. Targeted detection methods based on radiolabeled probes have been hampered by poor image resolution, the necessity of gamma cameras in the operating room and ionizing radiation exposure to patients and operators [1,2]. Recently, a lectin (asialo-receptor family)-targeted in vivo optical imaging technique has been proposed which utilizes a fluorescein-avidin conjugate
*
Corresponding author. E-mail address: [email protected] (H. Kobayashi).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.07.169
which enables the visualization of submillimeter peritoneal implants in a mouse ovarian cancer model [4]. Avidin is a 68 kDa tetrameric glycoprotein and binds with high capacity to lectins which are expressed on the surface of many cancer cells [1,2,4,5]. Lectins are proteins that specifically bind to carbohydrate groups on glycolipids and glycoproteins, such as avidin and galactosylated proteins [1,2,4,5]. Since avidin binds to lectin or other lectinlike negatively charged molecules at physiological pH and bound avidin is subsequently internalized into cancer cells it is an excellent targeting moiety for optical imaging [4]. In addition, when avidin is administered into the peritoneal cavity, unbound avidin is absorbed through the peritoneal membrane, enters the circulation, and is quickly trapped by the liver removing it from the imaging field [1,4,6]. Thus, avidin conjugated with optical probe is capable of targeted molecular imaging with high target-to-background ratios as well as high spatial resolution [4]. However, this targeting method for peritoneal cancer implants has been studied
808
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
in only two cancer cell lines without satisfactory in vitro cellular basis or in vivo imaging analyses for individual cells [1,2,4,7]. Moreover, the previously utilized fluorophore, fluorescein, while possessing some advantages in cost and availability, is not as stable as other fluorophores in the intracellular environment. A fluorophore 4,4-difluoro5-phenyl-4-bora-3a,4a-diaza-s-indacene-3-propionic acid, succinimidyl ester (BODIPY-R6G), which is a variant of BODIPY dyes with a little longer wavelength emission than 4,4-difluoro-5,7-dimethyl-4-bora-3a,4a- diaza-s-indacene-3-propionic acid, succinimidyl ester (BODIPY-FL), is stable in acidic environments and demonstrates an excellent quantum yield even after internalization into cancer cells [8] (Supplementary Fig. 1). Thus, the purpose of this study was to investigate the availability of lectin-binding as a target for a variety of cancer cell lines, all of which are associated with peritoneal dissemination using an avidin–fluorophore conjugate, Avidin–BODIPY. Materials and methods Synthesis of avidin–BODIPY and neutravidin–BODIPY conjugates. Avidin and neutravidin were purchased from Pierce Biochemical Inc. (Milwaukee, WI, USA). BODIPY-R6G was purchased from Molecular Probes Inc. (Eugene, OR, USA). We made BODIPY-conjugated avidin (avidin–BODIPY) and BODIPY-conjugated neutravidin (neutravidin– BODIPY). At room temperature, 500 lg (7.3 nmol for avidin and 8.3 nmol for neutravidin) of avidin or neutravidin in 200 lL of 0.1 M
Na2HPO4 was incubated with 8.7 lg (20 nmol) of BODIPY in DMSO for 15 min. The mixture was purified with Sephadex G50 (PD-10, GE Healthcare, Milwaukee, WI, USA). The protein concentration of each sample was determined with Coomassie Plus protein assay kit (Pierce Chem Co., Rockford, IL, USA) by measuring the absorption at 595 nm with a UV-Vis system (8453 Value UV-Bis system, Agilent Technologies, Palo Alto, CA, USA). Then, the BODIPY concentration was measured by the absorption at 534 nm with a UV–vis system (8453 Value UV-Bis system, Agilent Technologies, Palo Alto, CA, USA) to confirm the number of BODIPY molecules conjugated with each avidin or neutravidin molecule. For flow cytometry and fluorescence microscopy, the number of BODIPY molecules per avidin was 2 and the number of BODIPY molecules per neutravidin was 2. For in vivo study, the concentration of avidin or neutravidin was changed to produce the same fluorescent intensity from avidin–BODIPY and neutravidin–BODIPY. The number of BODIPY conjugated with avidin was 1.4 and that with neutravidin 1.7. Cell culture. An established human ovarian cancer cell line SHIN3 [7] as well as eight commercially available cancer cell lines (OVCAR-3, SKOV-3, PANC-1, Capan-2, MCF7, PC-3, HT-29, N87: ATCC, Manassas, VA USA) were used. SHIN3, OVCAR-3, and SKOV-3 were from human ovarian cancer, PANC-1 and Capan-2 from human pancreatic cancer, HT-29 from human colon cancer, N87 from human gastric cancer, MCF7 from human breast cancer, and PC-3 from human prostate cancer. SHIN3, OVCAR-3, MCF7, PC-3, HT-29, and N87 were grown in RPMI 1640 medium (Gibco, Gaithersburg, MD, USA), PANC-1 in DMEM (Gibco, Gaithersburg, MD, USA), Capan-2 and SKOV-3 in McCOY’s 5A medium (Quality Biological Inc., Gaithersburg, MD, USA) containing 10% fetal bovine serum (FBS) (Gibco, Gaithersburg, MD, USA), 0.03% L-glutamine at 37 C, 100 U/mL penicillin and 100 lg/mL streptomycin in 5% CO2. Fluorescence microscopy. The 9 cancer cells (5 · 105) were plated on a coverglass bottom culture well and incubated for 24 h. Avidin–BODIPY
Fig. 1. In vitro fluorescence microscopy (left) and differential interference contrast (right) images of 9 cancer cell lines. Although there are some differences in number, size, and intensity of fluorescent dots, these intracellular fluorescent dots are detected in all 9 cell lines after 4 h incubation with 30 lg/ml avidin–BODIPY.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813 was added to the medium (30 lg/ml) and the cells were incubated for another 4 h. Cells were washed one time with PBS. Fluorescence microscopy was performed using an Olympus BX51 microscope (Olympus America Inc., Melville, NY, USA) equipped with the following filters: excitation wavelength 470–490 nm, emission wavelength 515 nm long pass. Transmitted light differential interference contrast (DIC) images were also acquired. Flow cytometry. To test whether the avidin–BODIPY binds specifically to lectins on cancer cells through the glycosylation of avidin, one-color flow cytometry of avidin–BODIPY and deglycosylated avidin–BODIPY (neutravidin–BODIPY) was performed. Cells were incubated for 16 h and treated with 10 lg/ml avidin–BODIPY or neutravidin–BODIPY for 96 h. The cells were washed once with PBS, trypsinized, and flow cytometry was performed using a FACScan cytometer (Becton–Dickinson, Franklin Lakes, NJ, USA). The argon ion 488 nm laser was employed for excitation. Signals from cells were collected using a 530/30 nm band-pass filter. Control cells used in the flow cytometry were the same cell line that had not been incubated with avidin–BODIPY or neutravidin–BODIPY. Fluorescence histograms (in FL-1 channel) were created and the percentage of positive cells was analyzed using Cell Quest software (Becton–Dickinson, Franklin Lakes, NJ, USA). The percentage of fluorescence-gated positive cells which corresponds to an increase of fluorescence (M1) is indicated for each experimental condition.
809
Tumor model. All animal experiments were carried out in compliance with the Guide for the Care and Use of Laboratory Animal Resources (1996), National Research Council, and approved by the National Cancer Institute Animal Care and Use Committee. The intraperitoneal tumor xenografts were established by intraperitoneal injection of 2 · 106 OVCAR-3 cells suspended in 200 lL PBS in female nude mice (National Cancer Institute Animal Production Facility, Frederick, MD, USA). Experiments with tumor-bearing mice were performed at 28 days for OVCAR3 after injection of the cells. In vivo fluorescence imaging. Three hours after intraperitoneal injection of 50 lg avidin–BODIPY or 50 lg neutravidin–BODIPY diluted in 300 lL PBS into tumor-bearing mice, the mice were sacrificed with CO2 gas. The abdominal walls were removed and spectral fluorescence images were obtained using the Maestro In Vivo Imaging System (CRi Inc., Woburn, MA, USA). For direct comparison of fluorescence intensity between avidin–BODIPY and neutravidin–BODIPY, the number of BODIPY molecules conjugated to avidin or neutravidin was adjusted to obtain the same fluorescence intensity prior to the intraperitoneal administration. In addition, two mice or two extracted tumors either with avidin–BODIPY or with neutravidin–BODIPY were placed side-by-side to directly compare the fluorescence intensity of peritoneal cancer implants. Semiquantitative ex vivo comparison of fluorescence intensity of two tumors was performed. A region of interest (ROI) as large as each
Fig. 2. Flow cytometry of 9 cancer cell lines incubated with avidin–BODIPY (green) or neutravidin–BODIPY (red). Cells were incubated for 16 h followed by treatment with avidin–BODIPY or neutravidin–BODIPY for 96 h. Then the cells were washed, trypsinized, and analyzed for flow cytometry. The horizontal axis represents the fluorescence intensities of each cell line. The vertical axis represents the number of cells. The percentage of avidin– BODIPY- or neutravidin–BODIPY-accumulated cells which corresponds to an increase of fluorescence (M1) is indicated for each experimental condition. Control cells (black) were the same cell line that had not been incubated with avidin–BODIPY or neutravidin–BODIPY.
810
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
tumor was drawn to determine the fluorescence intensity as well as the histogram using ImageJ software (http://rsb.info.nih.gov/ij/plugins/ mri-analysis.html). The dynamic range of the fluorescent intensity in arbitrary unit (au) was split into equal-sized 256 bins (0–255). A band pass filter from 480 to 520 nm and a long pass filter over 550 nm were used for excitation and emission light, respectively. The tunable filter was automatically stepped in 10 nm increments from 500 to 800 nm while the camera captured images at each wavelength interval with constant exposure. Spectral unmixing algorithms were applied to create the unmixed images of BODIPY and autofluorescence. Three mice were studied in each group and all images were analyzed using commercial Maestro software (Nuance ver. 1.4, CRi Inc.).
Results Avidin-BODIPY internalized in all 9 cancer cell lines To confirm that avidin-BODIPY binds to cancer cells and internalizes into them, in vitro fluorescence microscopy as well as DIC images were obtained. Fluorescence microscopy of cultured cells clearly showed a large number of fluorescent dots within the cytoplasm (Fig. 1), indicating that the avidin–BODIPY internalized and located within the endoplasmic vesicles or lysosomes and continued to remain fluorescent. These findings are compatible to the previous studies performed by fluorescein-conjugated avidin for SHIN3 cancer cells [4], although the fluorescence appears more intense and of longer duration using the avidin–BODIPY conjugate.
of OVCAR-3 ovarian cancer was performed. This cell line was selected because OVCAR3 showed the maximal rightward shift in respective cancers after treatment with avidin–BODIPY in the flow cytometry analysis (Fig. 2). Before comparing the in vivo fluorescence intensity of the peritoneal implants produced by avidin–BODIPY and neutravidin–BODIPY, pre-injection fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY probes were adjusted to be the same by changing the number of BODIPY molecules per avidin or neutravidin molecule. The number of BODIPY molecules conjugated with avidin and neutravidin were 1.4 and 1.7, respectively. Signal intensities as well as the emission spectra of 8 lg avidin–BODIPY and 8 lg neutravidin–BODIPY in 390 lL PBS were comparable (Fig. 3). Using these two probes, in vivo spectral fluorescence imaging was performed 3 h after intraperitoneal injection of 50 lg avidin–BODIPY or 50 lg neutravidin–BODIPY. Since the fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY prior to the injection were the same, side-by-side comparison of the imaging was possible. The fluorescence intensity of the tumor nodules on the spectral fluorescence image was
Avidin–BODIPY is specifically bound and internalized into cancer cells via glycosylation To study whether the avidin–BODIPY is bound specifically to lectins on cancer cells through glycosylation, one-color flow cytometry of avidin–BODIPY and deglycosylated avidin–BODIPY (neutravidin–BODIPY) was performed. The percentage of fluorescence-gated cells which corresponds to internalization of avidin–BODIPY or neutravidin–BODIPY is shown in Fig. 2. For avidin–BODIPY, the percentages of SHIN3, OVCAR-3, SKOV-3, PANC-1, Capan-2, HT-29, N87 MCF7, and PC-3, increased from 2.3%, 3.7%, 0.8%, 2.3%, 3.0%, 1.4%, 1.2%, 1.7%, and 1.2% to 99.9%, 99.8%, 99.7%, 74.2%, 96.7%, 99.5%, 98.4%, 99.9%, and 99.9%, respectively. Whereas, when treated with neutravidin–BODIPY, the percentages of SHIN3, OVCAR-3, SKOV-3, PANC-1, Capan-2, HT-29, N87, MCF7, and PC-3 were 3.5%, 4.4%, 11.0%, 3.5%, 5.3%, 6.1%, 1.5%, 4.8%, and 2.1%, respectively. These results indicate that the avidin–BODIPY will bind specifically to lectins on a wide variety of cancer cells through the glycosylation of avidin. Avidin–BODIPY specifically targeted the peritoneal cancer foci in vivo To investigate the efficacy and the applicability of avidin–BODIPY to target the peritoneal cancer implants, in vivo spectral fluorescence imaging of peritoneal implants
Fig. 3. Optical characteristics of avidin–BODIPY and neutravidin– BODIPY. Eight micrograms of avidin–BODIPY and neutravidin– BODIPY in 390 lL PBS were placed in a nonfluorescent 96-well plate and spectral fluorescence image was obtained. The numbers of BODIPY molecules conjugated with avidin and neutravidin molecule were adjusted to 1.4 and 1.7, respectively, to achieve equal fluorescence of both molecules. (Upper) Fluorescence intensities of avidin–BODIPY and neutravidin–BODIPY were 242 and 243 in arbitrary units (au). (Lower) Emission spectra. These two dyes had the same emission peak at a wavelength of 570 nm and the almost identical fluorescence intensity when concentration was adjusted.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
811
substantially higher with avidin–BODIPY than that with neutravidin–BODIPY (Fig. 4). Conversely, the peri-tumoral background signals with neutravidin–BODIPY were substantially higher than the background signals with avidin–BODIPY due to the unbound free neutravidin–BODIPY reagent within the peritoneal cavity and the absorption of unbound free avidin–BODIPY across the peritoneal membrane. To avoid the normal autofluorescence from the intestine, aggregated tumor nodules were extracted, placed side-by-side on a non-fluorescent plate, and spectral fluorescence imaging was performed. Ex vivo spectral fluorescence image demonstrated that the fluorescence intensity of the tumors treated with avidin–BODIPY was higher than that with neutravidin–BODIPY (Fig. 5A).
Fig. 5. Ex vivo spectral fluorescence image of aggregated tumor nodules instilled with avidin–BODIPY or neutravidin–BODIPY. (A) An ROI was drawn as large as each tumor to compare the fluorescence intensity between avidin–BODIPY and neutravidin–BODIPY. (B) Histogram of fluorescent intensity of an ROI drawn on each of the tumor instilled with avidin–BODIPY or neutravidin–BODIPY. The dynamic range of the fluorescent intensity in arbitrary unit (au) was split into equal-sized 256 bins (0–255). Then for each bin (horizontal axis), the number of pixels from the data set that fall into each bin (vertical axis) is counted. The avidin–BODIPY has a larger proportion of counts for pixels in the higher fluorescent signal intensity range than neutravidin–BODIPY, indicating the fluorescence intensity of the tumor treated with avidin–BODIPY is higher than that of neutravidin–BODIPY.
The histogram of avidin–BODIPY has a larger proportion of counts for pixels in the higher fluorescent intensity range than neutravidin–BODIPY (Fig. 5B). Discussion
Fig. 4. In vivo spectral fluorescence images of tumor-bearing mice 3 h after intraperitoneal injection of avidin–BODIPY and neutravidin–BODIPY. (Upper) Autofluorescence image. (Middle) BODIPY fluorescence image. (Lower) Composite image (red: autofluorescence, green: BODIPY fluorescence). Spectral fluorescence images of the peritoneal cavities clearly visualized the aggregated peritoneal tumor foci (arrows) in mice injected with avidin–BODIPY, whereas, neutravidin–BODIPY showed minimal fluorescence from the tumor foci with higher background signals (arrowheads) due to the free unbound neutravidin–BODIPY reagent in the peritoneal cavity.
These results indicate that avidin–BODIPY binds specifically to cancer cells through lectin–avidin interaction, but unbound free avidin–BODIPY is absorbed through the peritoneal membrane and cleared away from the peritoneal cavity, ultimately resulting in a high signal-to-background ratio. Avidin–BODIPY, successfully targeted a wide variety of cancer cells, all of which have the potential for intraperitoneal dissemination. These results were compatible with the previously reported optical imaging study using fluorescein-conjugated avidin [4], but differ in that the uni-
812
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813
versal availability of lectin-targeted optical imaging method was investigated and verified, and because an acid-stable fluorophore, BODIPY, was newly used instead of fluorescein. There are several reasons why we chose BODIPY instead of fluorescein for targeting cancer foci both in vivo and in vitro. First, fluorescein is susceptible to photobleaching. Considering the clinical application of this targeting method in which the agent would constantly be exposed to surgical or endoscopic illumination might easily bleach the fluorescence of fluorescein, and ultimately decrease the signal intensity from the tumor foci. BODIPY is more resistant to photobleaching and relatively stable under visible light exposure. Moreover, BODIPY is stable under acidic conditions, whereas fluorescein is particularly vulnerable to acidic conditions. Even though fluorescein conjugates are internalized into cancer cells and accumulate within the endosomes or lysosomes, the fluorescing capability is reduced by the acidic conditions in endosomes or lysosomes whereas BODIPY conjugates are more resistant to the acidic environment and will continue to fluoresce. Finally, BODIPY can be synthesized as a potent photosensitizer (high extinction coefficient, high photostability, and insensitivity to solvent environment) [9]. By changing the structure of the BODIPY core, this imaging method might be applicable both for detection and for photodynamic therapy of disseminated peritoneal cancers. Malignant peritoneal dissemination has a high rate of clinical recurrence and is often lethal. However, recently there have been increasing number of reports describing a comprehensive multimodal treatment approach that involves cytoreductive surgery and intraperitoneal chemotherapy or photodynamic therapy [10–13]. Cytoreductive surgery aims to maximally eradicate visible tumor nodules, while intraperitoneal chemotherapy and photodynamic therapy aim to treat the remaining microscopic disease in an attempt to prolong disease free survival or even cure selected patients [10–13]. One of the major limitations of cytoreductive surgery is the completeness of the surgical resection, which is often limited by the poor visibility of cancer foci within the complex anatomy of the peritoneal cavity. At the same time, the non-targeted intraperitoneal chemotherapy and photodynamic therapy have a problem of local and systemic toxicity [3,10–13]. To overcome these problems, a targetspecific agent that could be used for detection as well as photodynamic therapy would be welcome. A probe based on the avidin–BODIPY conjugate could offer both diagnostic and therapeutics in a single agent. It should be noted that avidin is immunogenic and is therefore unlikely to be applicable clinically [14]. However, there are numerous agents with similar binding properties that are currently being investigated as substitutes in order to overcome this limitation. In conclusion, a photostable and acid-resistant targeted molecular imaging technique using avidin–BODIPY opti-
cal probe was shown to bind to a wide variety of cancer cell types, many of which are associated with peritoneal metastases. This technique might become a useful tool in detecting and treating small peritoneal cancer implants and subsequently improving the quality of cytoreductive surgery and adjuvant therapy. Acknowledgment This research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc. 2006.07.169. References [1] Z. Yao, M. Zhang, H. Sakahara, T. Saga, Y. Arano, J. Konishi, Avidin targeting of intraperitoneal tumor xenografts, J. Natl. Cancer. Inst. 90 (1998) 25–29. [2] Z. Yao, M. Zhang, H. Sakahara, T. Saga, Y. Nakamoto, N. Sato, S. Zhao, Y. Arano, J. Konishi, Imaging of intraperitoneal tumors with technetium-99m GSA, Ann. Nucl. Med. 12 (1998) 115–118. [3] M.C. Aalders, N. vd Vange, F.A. Stewart, M.G. Klein, M.J. vd Vijver, H.J. Sterenborg, White-light toxicity, resulting from systemically administered 5-aminolevulinic acid, under normal operating conditions, J. Photochem. Photobiol. B 50 (1999) 88– 93. [4] Y. Hama, Y. Urano, Y. Koyama, M. Kamiya, M. Bernardo, R. Paik, M.C. Krishna, P.L. Choyke, H. Kobayashi, In vivo spectral fluorescence imaging of submillimeter peritoneal cancer implants using a lectin-targeted optical agent, Neoplasia 8 (2006) 607–612. [5] R. Lotan, A. Raz, Lectins in cancer cells, Ann. N.Y. Acad. Sci. 551 (1988) 385–388. [6] H. Kobayashi, H. Sakahara, M. Hosono, Z.S. Yao, S. Toyama, K. Endo, J. Konishi, Improved clearance of radiolabeled biotinylated monoclonal antibody following the infusion of avidin as a ‘‘chase’’ without decreased accumulation in the target tumor, J. Nucl. Med. 35 (1994) 1677–1684. [7] S. Imai, Y. Kiyozuka, H. Maeda, T. Noda, H.L. Hosick, Establishment and characterization of a human ovarian serous cystadenocarcinoma cell line that produces the tumor markers CA-125 and tissue polypeptide antigen, Oncology 47 (1990) 177–184. [8] Y. Hama, Y. Urano, Y. Koyama, M. Bernardo, P.L. Choyke, H. Kobayashi, A Comparison of the emission efficiency of four common green fluorescence dyes after internalization into cancer cells, Bioconjugate Chem (submitted for publictaion). [9] T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa, T. Nagano, Highly efficient and photostable photosensitizer based on BODIPY chromophore, J. Am. Chem. Soc. 127 (2005) 12162–12163. [10] T.D. Yan, O.A. Stuart, D. Yoo, P.H. Sugarbaker, Perioperative intraperitoneal chemotherapy for peritoneal surface malignancy, J. Transl. Med. 4 (2006) 17. [11] D.K. Armstrong, B. Bundy, L. Wenzel, H.Q. Huang, R. Baergen, S. Lele, L.J. Copeland, J.L. Walker, R.A. Burger, Gynecologic Oncology Group, Intraperitoneal cisplatin and paclitaxel in ovarian cancer, N. Engl. J. Med. 354 (2006) 34–43.
Y. Hama et al. / Biochemical and Biophysical Research Communications 348 (2006) 807–813 [12] I.J. Torres, C.L. Litterst, A.M. Guarino, Transport of model compounds across the peritoneal membrane in the rat, Pharmacology 17 (1978) 330–340. [13] T.J. Dougherty, C.J. Gomer, B.W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, Photodynamic therapy, J. Natl. Cancer. Inst. 90 (1998) 889–905.
813
[14] V.P. Hytonen, O.H. Laitinen, A. Grapputo, A. Kettunen, J. Savolainen, N. Kalkkinen, A.T. Marttila, H.R. Nordlund, T.K. Nyholm, G. Paganelli, M.S. Kulomaa, Characterization of poultry egg-white avidins and their potential as a tool in pretargeting cancer treatment, Biochem. J. 372 (2003) 219– 225.