GASTROENTEROLOGY 2002;122:406-414 Detection of Dysplastic Intestinal Adenomas Using EnzymeSensing Molecular Beacons in Mice KATHARINA MARTEN,* CHRISTOPH BREMER,* KHASHAYARSHA KHAZAIE,~ MANSOUREH SAMENI,§ BONNIE SLOANE,§ CHING-HSUAN TUNG,* and RALPH WEISSLEDER* *Center for Molecular Imaging Research, Massachusetts General Hospital and Harvard Medical School, Boston; ~Dana-Farber Cancer Institute, Cancer Immunology & AIDS, Harvard Medical School, Boston, Massachusetts; §Department of Pharmacology, Wayne State University School of Medicine, Detroit, Michigan See editorial on page 571. Background&Aims: Proteases play key roles in the pathogenesis of tumor growth and invasion. This study assesses the expression of cathepsin B in dysplastic adenomatous polyps. Methods: Aged Apc M~n/+ mice served as an experimental model for familial adenomatous polyposis. The 4 experimental groups consisted of (a) animals injected with a novel activatable, cathepsin B sensing near infrared fluorescence (NIRF) imaging probe; (b) animals injected with a nonspeciflc NIRF; (c) uninjected control animals; and (d) non-APCM~n/+ mice injected with the cathepsin B probe. Lesions were analyzed by immunohistochemistry, Western blotting, reverse transcription-polymerase chain reaction, and optical imaging. Results: Cathepsin B was consistently overexpressed in adenomatous polyps. When mice were injected intravenously with the cathepsin reporter probe, intestinal adenomas became highly fluorescent indicative of high cathepsin B enzyme activity. Even microscopic adenomas were readily detectable by fluorescence, but not light, imaging. The smallest lesion detectable measured 50 ~m in diameter. Adenomas in the indocyanine green and/or noninjected group were only barely detectable above the background. Conclusions: The current experimental study shows that cathepsin B is up-regulated in a mouse model of adenomatous polyposis. Cathepsin B activity can be used as a biomarker to readily identify such lesions, particularly when contrasted against normal adjacent mucosa. This detection technology can be adapted to endoscopy or tomographic optical imaging methods for screening of suspicious lesions and potentially for molecular profiling in vivo. dentification and removal of colonic adenomatous polyps during endoscopy has been shown to reduce the incidence of colorectal cancer. ~ Adenomatous polyps represent up to half of all colonic polyps and are particularly worrisome if they are large (>1 cm) or multiple, have I extensive villous components, 2,3 and/or are highly dysplastic. 1 The latter feature in particular has been associated with the successive development of oncogenes and the loss of tumor-suppressor genes 4 and ultimately leads to carcinoma formation. However, apart from the size criterion, it is currently clinically difficult to ascertain the extent of dysplastic features 5 during colonoscopy in vivo. The availability of techniques that allow in vivo identification of small highly dysplastic adenomas from innocuous lesions would be helpful to guide selective removal of polyps. The exact molecular events occurring during colorectal tumorigenesis are slowly emerging. The genesis of familial adenomatous polyposis (FAP), as well as sporadic colorectal neoplasms, is closely linked with genetic defects that result in carboxy-terminal truncations of the adenomatous polyposis coli (APC) protein. Colorectal tumors with intact APC genes were found to contain activating mutations of ~-catenin. 6,v APC and [3-catenin are key components of the W n t signaling pathway (see reviewsS-U)). It is generally agreed that most colonic cancers develop from adenomatous polyps. Common to most tumors, several generic features become altered during multistage tumor progression, most importantly, the control of proliferation, the balance between cell survival, and programmed cell death (apoptosis), interactions with host cells and extracellular matrix, angiogenesis, and metastatic dissemination. H'12 Proteolytic enzymes have been shown to play an essential role in many of these processes, in particular high cell turnover, invasion, and angiogenesis, t3 Some of these proteases include matrix-metalloproteases (e.g., MMP), serine proAbbreviations used in this paper: APC, adenomatous polyposis coli; ICG, indocyanine green; FAP, familial adenomatous polyposis; MMP, matrix-metalloproteases; NIR, near-infrared; NIRF, near infrared fluorescence; SI, signal intensities; TBC, target to background contrast. © 2002 by the American Gastroenterological Association 0016-5085/02/$35.00 doi:10.1053/gast.2002.30990 February 2002 FLUORESCENCEIMAGING OF INTESTINALADENOMAS 407 teases (e.g., urokinase-type plasminogen activator), aspartic proteases (e.g., cathepsin D), and cysteine proteases (e.g., cathepsin 8). 13,14 Metalloproteinase has been reported to be direct targets of the W n t signaling pathway. t5,[6 Cathepsin B in particular has been shown to be up-regulated in areas of inflammation, necrosis, angiogenesis, t3 and focal invasion iv of colorectal carcinomas and in dysplastic adenomas. 18,19 O u r laboratory has been interested in developing protease-specific sensor molecules that can be used for the noninvasive in vivo m o n i t o r i n g of enzyme activities. 2° 22 W e have developed optically based, protease activatable fluorescent sensors that operate in the near-infrared ( N I R ) region for m a x i m u m light tissue penetration. 2°,23 The goal of the current study was to determine whether cathepsin B protease activity could be identified within dysplastic adenomatous polyps and whether targeting of this enzyme could be used for the detection of small dysplastic adenomatous lesions. W e chose to investigate adenomatous polyps in the A P C Min/+ mouse which is heterozygous for a germ-line m u t a t i o n in the mouse homologue of the h u m a n A P C gene. 24 These animals develop multiple adenomas in the small and large bowel which simulate dysplastic adenomatous polyps found in h u m a n disease. 24 26 Materials and Methods Mouse Model A P e i~[in/+ mice (n = 24, age 7 - 28 weeks) were obtained from the Jackson Laboratories (Bar Harbor, ME). Mice were randomly divided into 3 experimental groups: (1) animals receiving an intravenous injection of the cathepsin B-sensitive NIRF probe (n = 10; 142 adenomas; mean age, 15.8 weeks; 2 nmol Cy5.5); (2) animals receiving an intravenous injection with indocyanine green (ICG), a nonactivatable fluorochrome (n = 4 mice, 44 adenomas; mean age, 16 weeks; 200 btg); and (3) noninjected control mice (n = 2 mice; 18 adenomas; mean age, 18 weeks). In addition, we also included a cohort of strain matched non-APC mice (C57BL6, Jackson Laboratories, Bar Harbor, ME, group d) and injected them with the cathepsin B-sensitive NIRF probe (n = 3; no true adenomas; mean age, 12 weeks; 2 nmol Cy5.5). The animal protocol was approved by the Institutional Review Board. Mice were anesthetized (90 mg/kg ketamine and 10 mg/kg xylazine intraperitoneally) for intravenous injections and killed 24 hours later under halothane inhalation, and the entire bowel was removed for light and NIR fluorescence imaging and for correlative studies. Adenomas were identified after trypan blue staining using a dissection microscope. With this technique it is easy to distinguish a lymph node (Peyer's patch) from an adenoma; the former has a smooth surface and occurs often in pairs, whereas the latter has typical pedunculated/tubular or sessile/villous morphology. A total of 142 polyps (group a) and 30 "polyp-like" lesions (group [a] and [d]) were imaged (Table 1), whereas lesions in these groups were identified by dissecting microscopy. An additional, 6 animals were used for in vitro analyses and 2 animals (n = 30 polyps) for hematoxylin and eosin staining of polyps. Histology Bowel tissue of a subset of animals (30 adenomas) was fixed in phosphate-buffered formalin for 24 hours, paraffinembedded, sectioned into 6-Ftm thin slices and stained with hematoxylin and eosin. Immunohistochemistry for cathepsin B expression was performed on frozen sections using a primary polyclonal anti-cathepsin B antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Binding of the primary antibody was revealed using an alkaline phosphatase-labeled secondary antibody. NBT/BCIP substrate (Boehringer-Mannheim, IN) was used to visualize specific alkaline phosphatase activity. Sections were counterstained with nuclear fast red. Control sections were processed identically but using only the secondary antibody. Fluorescence confocal microscopy was also performed on normal and APC Min/+ mouse intestine. Sections of the paraffinembedded tissues were placed on glass slides, subjected to deparaffination in xylene and ethanol, and microwaved twice for 2-5 minutes each. 2v Sections were blocked with 10% normal donkey serum for 1 hour at room temperature. Sections were incubated with a 1:400 dilution of primary antibody (polyclonal anti-cathepsin B immunoglobulin [Ig]G, developed, and characterized for specificity by us) overnight at room temperature. After washing 5 × in phosphate-buffered saline (PBS), they were reacted for 1 hour at room temperature with a 1:100 dilution of secondary antibody (fluorescein-conjugated donkey anti-rabbit IgG) plus 5% normal donkey serum and washed 5 × with PBS. Immunofluorescence caused by binding of the primary antibody to cathepsin B was observed on a Zeiss LSM 310 confocal microscope (Zeiss, Thornwood, NY). Controis were run in the absence of primary antibody. Reverse Transcription-Polymerase Chain Reaction Mice were killed, and the bowel was removed, flushed with cold phosphate-buffered saline, opened longitudinally, and examined under a stereomicroscope. Adenomas were ex- Table 1. Sensitivity and Specificity Visible light imaging with magnification NIR fluorescence Lesions Positive Negative Positive Negative 142 polyps 30 polyp-like lesions 70 18 72 12 136 2 6 28 NOTE. Results are based on 142 polyps in APCM~n/+ mice and 30 polyp-like lesions (APCM'n/÷ mice and C57BL6 mice). The polyps were identified by trypan blue staining of mucosa and visualizing bowel specimen under a dissection microscope. 408 MARTENET AL. cised and snap frozen in liquid nitrogen. Normal intestinal mucosa from the vicinity of the adenomas was collected in a similar fashion. Reverse transcription-polymerase chain reaction (RT-PCR) was performed using an ABI Prism 7700 Sequence BioDetector (PE Biosystems, Foster City, CA) with SYBR-green fluorescence detection according to the manufacturer's instructions (Perkin Elmer, Foster City, CA). Total RNA was extracted using Trizol (Gibco BRL) and reverse transcribed using Superscript IIRT (Gibco BRL) and oligo(dT)15 priming. Cathepsin B primers were intron-spanning and included (5' to 3') AGG TTC GGT CAG AAA TGG CTT, and (5' to 3') ATC CTT CTT TCT TGC CTG CTG. Beta-Actin-1 primers were (5' to 3') TGG AAT CCT GTG GCA TCC ATG AAA C, and (5' to 3') TAA AAC GCA GCT CAG TAA CAG TCC G. Cycling conditions were as follows: 5 minutes at 95°C, 2 cycles (45 seconds) at 94°C, 45 seconds at 60°C, 1 minute at 72°C, 2 cycles (45 seconds) at 94°C, 45 seconds at 58°C, 1 minute at 72°C, 2 cycles (45 seconds) at 94°C, 45 seconds at 56°C, 1 minute at 72°C, and 34 cycles (45 seconds) at 94°C, 45 seconds at 54°C, 1 minute at 72°C, followed by 10 minutes at 72°C. Actin and cathepsin B reactions were analyzed on the same plate. Relative starting quantities of complementary DNA (cDNA) for each tissue sample were determined using standard curves made from six 1:3 serial dilutions of wild-type Balb/c total spleen cDNA. Standard curves were plotted as dilution factor versus threshold cycle. Values for both actin and cathepsin B expression in adenoma and normal control tissue samples were determined by putting the experimentally determined threshold cycle values into the standard curve formula. Raw data were normalized for relative amount of total cDNA and tabulated. GASTROENTEROLOGYVol. 122, No. 2 probe (Figure 1) and a nonspecific fluorochrome (indocyanine green), in clinical use for retinal angiography28 The NIRF probe contained Cy5.5 monofunctional dye (Amersham Pharmacia Biotech, UK) reporters adjacent to . . K - K . . . cleavage sites on a macromolecular assembly, described in more detail elsewhere, e° Activatability by cathepsin B and quality control was performed for each batch synthesized. The assembly consisted of a synthetic graft copolymer containing partially pegylated (5 kilodaltons) poly-L-lysine (35 kilodaltons), similar to what had been used clinically, e9 The injection dose (2 nmol Cy5.5 per animal and time of imaging after injection) had previously been optimized in non-APC M'"/+ mice bearing other pathologies. 2°m ICG (Akorn, IL, 2 mg/mL) was freshly prepared and used for intravenous injections (200 b~g per animal). ICG was used as a control to test the hypothesis that the fluorescence signal is mainly caused by activation of the NIRF probe rather than nonspecific accumulation. MPEG NH 2 MPEG L!s--Lys--Lys--Lys--L!s Western Blotting Adenomas and healthy mucosa of APC M'n/+ mice were homogenized in lysis buffer (50 mmol/L Tris-HC1 pH 7.4, 100 mmol/L NaC1, 10 mmol/L CaCI2 containing 0.25% Triton X-100 and "complete" protease inhibitor cocktail; Boehringer Mannheim, Germany). Total protein content was determined using a bicinchoninic acid protein assay (Pierce, Rockford, IL). Equal amounts of adenoma and mucosa extracts (4.5-30 I~g) were loaded onto a 10% sodium dodecyl sulfate polyacrylamide gel. After separation proteins were transferred to a polyvinylidine difluoride membrane (Bio-Rad Laboratories, Hercules, CA). After blocking for 1 hour (3% bovine serum albumin in phosphate-buffered saline [PBS]), membranes were incubated with the primary anti-cathepsin B polyclonal antibody (Bio-Rad Laboratories, Hercules, CA) and a secondary antibody conjugated with alkaline phosphatase (Sigma, St. Louis, MO). Alkaline phosphatase activity was revealed with NBT/BCIP substrate (Boehringer Mannheim, Germany). Membranes were scanned and lanes were analyzed digitally (Kodak Digital Science 1D software, Rochester, NY). NIR Fluorochrome Probes Two different near-infrared fluorescence (NIRF) probes were used in this study: an activatable cathepsin B sensing g athepsin B activation MPEG NH2 Lys~Lys-- LIs y ~ MPEG NH 2 I HOOC ~ Ly s ~ Lys I NH -O- NH h 'ox -@- Figure 1. Schematic diagram of the cathepsin B probe and its activation. The imaging probe consists of a cleavable indocyanine containing graft copolymer. The fluorochromes are essentially nonfluorescent in their native state caused by energy resonance transfer among fluorochromes. On enzymatic cleavage the agent becomes fluorescent in the near-infrared ( h v e x = excitation wavelength, h v e m = emission wavelength). February 2002 FLUORESCENCEIMAGINGOF INTESTINALADENOMAS 409 Imaging and Image Analysis Bowel samples were imaged immediately after excision using a custom built NIRF reflectance imaging system. 23 The system consisted of a light-tight chamber equipped with a 130-watt halogen white light source and an excitation filter system (610-650 nm, Omega Optical, Brattleboro, VT). Light diffusers were used to homogeneously distribute light over the field of view. A 12-bit monochrome CCD camera (Kodak, Rochester, NY) equipped with a f/1.2 12.5-75 mm zoom lens and an emission bandpass filter (Omega Optical, Brattleboro, VT) was used to detect fluorescence (Cy5.5 680720 nm and ICG 780-820 nm emission). Images were acquired over 3 minutes and analyzed using commercially available software (Kodak Digital Science 1D software, Rochester, NY). Regions of interests were obtained from the entirety of each adenoma and from adjacent size matched intestinal mucosa. Mean signal intensities (SI) were recorded. The target (adenoma) to background (mucosa) contrast (TBC) was calculatedas follows: TBC(%) = ([SI.d ..... -SI ....... ]/SI ...... ) × 100 (i.e., a value of 100 will represent a 100% higher fluorescence signal of the adenoma compared with mucosa). All results are presented as mean + standard error of the mean (SEM). Statistical analysis of the 2 groups was conducted using a 2-tailed Student t test for unpaired samples. A P value <0.05 was considered to be significant. Results Adenomas (n = 204) ranged in size from 50 btm to 6 m m in diameter (mean 2 mm). It has previously been shown that the majority of cells found in tumors of APC Min/+ mice do not display any of the differentiation markers found in the various cell types of the intestinal epithelium and therefore represent undifferentiated cells. 26 Analysis of histologic sections showed the characteristic architecture of dysplastic crypts located at the luminal surface of the mucosa, with underlying, wellspaced, nondysplastic crypts. In some lesions, high-grade dysplasia, based on the identification of full-thickness crypt epithelial nuclear stratification and loss of cytoplasmic mucinous differentiation, was present. Dysplasia in general was characterized by tall, hyperchromatic disorderly cells with cigar-shaped nuclei and concomitant crypt budding (Figure 2). Immunohistochemistry and fluorescence confocal microscopy was positive for cathepsin B expression throughout the adenoma in epithelial and stroma cells (Figure 2). W i t h progressive stages of dysplasia, cathepsin B expression levels were successively higher. By Western blotting, adenomas had a 36% -+ 8.6% higher cathepsin B protein level when compared with adjacent normal mucosa. RT-PCR showed cathep- sin B m R N A to be about 35% -+ 8.3% higher compared with adjacent mucosa on average. In the noninjected animals, N I R imaging revealed a similar signal intensity of adenomas compared with that of adjacent mucosa (Figure 3). However, adenomas in animals injected with the cathepsin B probe showed a remarkably higher TBC (Figure 3). On average, contrast was highest in large adenomas (220% + 97%), presumably caused by the higher amount of converting enzyme per lesion. To prove that fluorescence signal intensity was caused by enzyme activation in adenomas and not only nonspecific accumulation, a cohort of animals was injected with fluorescent ICG. Adenomas (n = 44) in these animals showed significantly lower TBC contrast as those having received the cathepsin B sensing probe (TBC = 34% - 4% vs. 119% + 71%, P < 0.01, Figure 3). Figure 4 summarizes the visible light and the N I R appearance of jejunal bowel segments in the different experimental groups. The N I R images are scaled equally and clearly show the high conspicuity of adenomas using the cathepsin B reporter probe. Figure 5 summarizes the appearance of colonic adenomatous polyps. Interestingly, even minute adenomas, not detectable by light imaging, became easily detectable by NIRF imaging. Table 1 summarizes quantitative data on lesion detectability using either visible light or N I R fluorescence. Sensitivity for NIRF imaging was 96% with a specificity of 93%. The sensitivity of light imaging was 49% and the specificity 40%. Discussion The current results indicate that adenomatous polyps show moderately elevated cathepsin B expression and high-enzyme activity. The enzyme activity within these lesions was ubiquitous and was highest in larger colonic polyps with high degrees of dysplasia. These lesions were highly conspicuous and even adenomas of sizes as small as 50 lzm in diameter could be readily identified. The results from this pilot study have 5 practical implications: (1) cathepsin B protease may play a role in early alterations leading to tumor formation; (2) such proteolytic enzymes and potentially other "biomarkers" can be used for in vivo molecular imaging of suspicious lesions; (3) multiwavelength imaging with different probes may facilitate "typing" of lesions; (4) the strategy can be used to improve the detection of adenomas (particularly partly obscured sessile lesions); and (5) the probe technology could be readily adapted to conventional endoscopy or even external tomographic N I R imaging of bowel. 410 MARTEN ET AL. GASTROENTEROLOGYVol. 122, No. 2 Figure 2. Adenomas in the APCMin/+ model. (A-C) Hematoxylin-eosin stain of small (A) and a large (B, C) colonic adenoma (asterisk). The latter shows the characteristic architecture of dysplastic epithelium with tortuous and branching crypts and hyperchromatic nuclei. (A, B) Objective magnification: 2-fold, (C) 20-fold. (D-l) Fluorescence confocal microscopy. (D) Normal intestine; (E-I) Adenomas representing progressive stages of dysplasia from (E) least to (/) most dysplastic. Note the increasing cathepsin B expression levels. Role of Proteases in Premalignant Lesions The involvement of proteases in cancers is well established; however, recently there has been mounting evidence that proteases are also involved in early alterations leading to tumor formation. 13 For example, increased immunostaining and altered localization of cathepsin B has been observed in late human adenomas 3° and has been associated in particular with high-grade dysplasia) 1 In the latter study, cathepsin B-positive tumor cells were observed in 67% of adenomas but in 100% of adenomas with high-grade dysplasia or adenocarcinomas. Additional evidence for the importance of protease involvement comes from animal studies. When APC Min/+ mice, such as those used in this study, are crossed with matrilysin (matrix metalloprotease) deficient mice, the development of spontaneous intestinal polyps is decreased) 2 Our own data show that cathepsin B was up-regulated in intestinal adenomas in the February 2002 FLUORESCENCE IMAGING OF INTESTINAL ADENOMAS 4110¸ 3-6ram T "~ 300 ¸ <3ram 200 a. T ,00: ,7-, Control ICG [~ ± 411 enomas were observed at the site of infection within 1 week after inoculation of the virus and loss of both APC alleles. However, in conflict with a simple dominantnegative mode of action of the truncated APC, over- _k Cathepsin B probe Figure 3. Conspicuity of adenomatous polyps. Data show the TBC. Adenomas investigated with the cathepsin B imaging probe show a TBC between 100% and 350%, at which adenomas appear as "light bulbs" against dark mueosa (Figure 4). A P C Min/+ m o u s e both at the m R N A and protein level. The exact mechanism by which cathepsin B is up-regulated remains to be investigated. One working hypothesis is that it could be up-regulated through the W n t signaling pathway similar to the up-regulation of matrix metalloproteases. Alternative possibilities include an indirect effect of epithelial hyperproliferation or activation, mesenchyme response to signals from the transformed epithelial cells, or caused by the inflammatory response and recruitment of activated host immune cells. In any case, it is unlikely that cathepsin activation is caused by a secondary genetic/oncogenic event other than loss of APC, because cathepsin activation can be already detected in very early adenomas. Studies with transgenic knock-out mice have confirmed that loss of the wild-type APC allele is the critical secondary genetic event determining the transition from normal epithelium ro polyps. 33-36 These studies have been confirmed by conditional knock-out of both alleles of APC in adult m i c e ) v In this case exon 14 of the mouse APC gene was flanked with loxP sequences, and the Cre recombinase was provided to the adult mice through intrarectal infection with recombinant adenoviral vectors encoding Cre. Ad- Figure 4. Light and NiR fluorescence imaging of jejunal adenomas. (A, B) Light images and (C-D) NIRF images of (A, C) a noninjected 6-month-old APCMrn/+ mouse and (B, D) a 6-month-old APCMin/+ mouse injected with the NIRF probe. (A) In the noninjected mouse, multiple adenomas can be seen on the light image, (C) but are essentially not fluorescent. (D) With injection of the NIRF probe, the adenomas are clearly identified by NIRF imaging (Scale bar, 5 ram). Figure 5. Colonic adenomas in a 7-month-old APCMin/+ mouse 24 hours after injection of the NIRF probe. (A) The light image shows 3 adenomas at a size range of 2-5 mm in diameter. Multiple small adenomas are readily detectable by NIRF imaging (B, arrows), but can barely be seen on the light image. (C) Correlative transillumination microscopy revealing the small adenoma (shown in B, right arrow) to be approximately 50 i~m in size (arrow). 412 MARTEN ETAL. GASTROENTEROLOGYVol. 122, No. 2 expression of truncated APC genes in transgenic mice did not lead to polyposis. 35 Another area of investigations lies in determining the mechanism by which cathepsin B may increase cell proliferation for example by activating growth factors (e.g., fibroblast growth factor, epidermal growth factor, transforming growth factor-B, vascular endothelial growth factor) or by liberating them from the extracellular matrix. 38 Irrespective of these areas of ongoing investigation, it is clear through the current data and that of other investigators, 13 that proteases may play important roles in premalignant lesions. Reporter Probe The recent development of targeted NIRF, 39 activatable NIR fluorochromes, 2o red-shifted fluorescent proteins, 4° and bioluminescent probes 4~ is slowly opening the road toward in vivo molecular imaging. The cathepsin B probe used in the current study was a quenched NIR probe previously used to detect malignant lesions. 2° One significant advantage of fluorescent over other reporters (e.g., isotopes, iodinated agents for radiograph) is that they can be "silenced" and "activated," enabling the design of molecular switches or "beacons" (Figure 1). The probe used in this study and probes in other studies 2°-22 are nonfluorescent in their native (quenched) state and become highly fluorescent after target interaction, resulting in signal amplification of up to several hundred-fold, depending on the specific design. The dosing and timing of imaging of these probes has previously been optimized and was adapted in the current study. Using activatable probes in particular has several major advantages over single fluorochromes attached to affinity molecules (such as antibodies). Most important, quenching results in reduction of background "noise" by several orders of magnitude and a single enzyme can cleave multiple fluorochromes resulting in efficient signal amplification. This advantage is best exemplified by our results (Figure 3) in which we show that the TBC is an order of magnitude higher with the cathepsin B probe than when nonspecific fluorochromes are used. By choosing the appropriate substrate spacer, a series of very specific enzyme probes can be developed, and multiple probes can potentially be used for multispectrum imaging. Detection Technology Inherently linked to the development of the previously described reporter probes2°-= and the molecular analysis of cancers and precancerous lesions, 13,17,3°is the need to develop detection technology that can accurately quantitate NIR fluorochrome concentrations and fluorescence activation in vivo. The current study used CCD technology to examine the bowel specimen in reflectance model, 23 similar to the technology used during fiberoptic endoscopy. The current detection technology could thus be easily adapted t o real-time endoscopy or even multiwavelength channel endoscopy. The adaptation would essentially require a separate NIR light source, appropriate filters, and a sensitive CCD camera for detection. NIR light has been shown to travel up to 7-10 cm through tissue using Food and Drug Administration class I-3 laser sources. With advanced technology for single photon counting or very low-noise detection systems, it is also feasible to potentially record the molecular signatures from outside the abdomen. Recently, optical tomography with NIR light has been described42 and being facilitated by rigorous mathematical modeling of light propagation in tissue and technological advancements in photon sources and detection techniques. It is clear that the newly described probe armamentarium and novel NIR photon detection technology stand a good chance to significantly impact on our capability of imaging molecular targets in vivo. Clinical Implications The current study has several clinical implications. NIRF endoscopy using enzyme-sensing imaging probes may become a complementary tool (molecular endoscopy) to other types of endoscopy such as chromoe n d o s c o p y 43,44 or light-induced fluorescence spectroscopy.45-47 We believe that the required NIRF imaging technology could be easily integrated into existing endoscopic systems, to provide high-resolution, real-time imaging of larger intestinal surface areas. A second clinical application may be in externally usable tomographic NIRF imaging, by which the underlying colon could be imaged from outside the abdominal surface. Based on the current probe, a number of different enzyme or targetsensing molecules could be designed for in vivo sensing of broader ranges of biomarkers. Finally, such reporter probes may become useful in monitoring of therapy with protease inhibitors using other forms of endoscopy or laparoscopy. 22 It is also conceivable that the technology could be used to identify other pathologies, e.g., active inflammation in ulcerative colitis. In conclusion, NIRF imaging using protease-activatable imaging probes may have a significant impact on diagnosis of a very early stage of intestinal disease. With further advances in technology and chemistry, we are likely to see significant advances in optical imaging in vivo and sensing. References 1. Winawer SJ, Fletcher RH, Miller L, Godlee F, Stolar MH, Mulrow CD, Wootf SH, Glick SN, Ganiats TG, Bond JH, Rosen L, Zapka JG, Olsen SJ, Giardiello FM, Sisk JE, Van Antwerp R, Brown-Davis C, February 2002 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Marciniak DA, Mayer RJ. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology 1997;112:594-642. O'Brien MJ, Winawer SJ, Zauber AG, Gottlieb LS, Sternberg SS, Diaz B, Dickersin GR, Ewing S, Geller S, Kasimian D, et al. The National Polyp Study. Patient and polyp characteristics associated with high-grade dysplasia in colorectal adenomas. Gastroenterology 1990;98:371-379. Atkin WS, Morson BC, Cuzick J. Long-term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med 1992;326:658-662. Christofori G, Hanahan D. Molecular dissection of multi-stage tumorigenesis in transgenic mice. Semin Cancer Biol 1994;5: 3-12. Pfau PR, Sivak MV. Endoscopic diagnostics. Gastroenterology 2001;120:763-781. Morin PJ, Sparks AB, Korinek V, Barker N, Clevers H, Vogelstein B, Kinzler KW. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science 1997;275: 1787-1790. Sparks AB, Morin PJ, Vogelstein B, Kinzler KW. Mutational analysis of the APC/beta-catenin/Tcf pathway in colorectal cancer. Cancer Res 1998;58:1130-1134. van Es JH, Giles RH, Clevers HC. The many faces of the tumor suppressor gene APC. Exp Cell Res 2001;264:126-134. Fearon ER, Dang CV. Cancer genetics: tumor suppressor meets oncogene. Curr Biol 1999;9:R62-R65. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell 1996;87:159-170. Compagni A, Christofori G. Recent advances in research on multistage tumorigenesis. Br J Cancer 2000;83:1-5. Gryfe R, Swallow C, Bapat B, Redston M, Gailinger S, Couture J. Molecular biology of colorectal cancer. Curr Probl Cancer 1997; 21:233-300. Koblinski JE, Ahram M, Sloane BF. Unraveling the role of proteases in cancer. Clin Chim Acta 2000;291:113-135. Kountouras J, Boura P, Lygidakis NJ. New concepts of molecular biology for colon carcinogenesis. Hepatogastroenterology 2000; 47:1291-1297. Brabletz T, Jung A, Dag S, Hlubek F, Kirchner T. Beta-catenin regulates the expression of the matrix metalloproteinase-7 in human colorectal cancer. Am J Pathol 1999;155:1033-1038. Crawford HC, Fingleton BM, Rudolph-Owen LA, Goss KJ, Rubinfeld B, Polakis P, Matrisian LM. The metalloproteinase matrilysin is a target of beta-catenin transactivation in intestinal tumors. Oncogene 1999;18:2883-2891. Emmert-Buck MR, Roth MJ, Zhuang Z, Campo E, Rozhin J, Sloane BF, Liotta LA, Stetler-Stevenson WG. Increased getatinase A (MMP-2) and cathepsin B activity in invasive tumor regions of human colon cancer samples. Am J Pathol 1994;145:12851290. Hazen LG, Bleeker FE, Lauritzen B, Bahns S, Song J, Jonker A, Van Driel BE, Lyon H, Hansen U, Kohler A, Van Noorden CJ. Comparative localization of cathepsin B protein and activity in colorectal cancer. J Histochem Cytochem 2000;48:1421-1430. Herszenyi L, Plebani M, Carraro P, De Paoli M, Roveroni G, Cardin R, Tulassay Z, Naccarato R, Farinati F. The role of cysteine and serine proteases in co~orectal carcinoma. Cancer 1999;86: 1135-1142. Weissleder R, Tung CH, Mahmood U, Bogdanov A Jr. In vivo imaging of tumors with protease-activated near-infrared fluorescent probes. Nat Biotechnol 1999;17:375-378. Tung C, Mahmood U, Bredow S, Weissleder R. In vivo imaging of proteolytic enzyme activity using a novel molecular reporter. Cancer Res 2000;2000:4953-4958. Bremer C, Tung CH, Weissleder R. In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med 2001; 7:743-748. FLUORESCENCE IMAGING OF INTESTINAL ADENOMAS 413 23. Mahmood U, Tung C, Bogdanov A, Weissleder R. Near infrared optical imaging system to detect tumor protease activity. Radiology 1999;213:866-870. 24. Moser AR, Pitot HC, Dove WF. A dominant mutation that predisposes to multiple intestinal neoplasia in the mouse. Science 1992;247:322-324. 25. Merritt AJ, Gould KA, Dove WF. Polyclonal structure of intestinal adenomas in Apc M'n / + mice with concomitant loss of Apc+ from all tumor lineages. Proc Natl Acad Sci U S A 1997;94: 13927-13931. 26. Shoemaker AR, Gould KA, Luongo C, Moser AR, Dove WF. Studies of neoplasia in the min mouse. Biochim Biophys Acta 1997; 1332:F25-F48. 27. Moin K, Day NA, Sameni M, Hasnain S, Hirama T, Sloane B. Human tumour cathepsin B. Comparison with normal liver cathepsin B. Biochem J 1992; 285(Pt 2):427-434. 28. Slakter JS, Yannuzzi LA, Guyer DR, Sorenson JA, Orlock DA. Indocyanine-green angiography. Curt Opin Ophthalmol 1995;6: 25-32. 29. Callahan RJ, Bogdanov A Jr, Fischman AJ, Brady TJ, Weissleder R. Preclinical evaluation and phase I clinical trial of a 99mTc-labeled synthetic polymer used in blood pool imaging. AJR Am J Roentgenol 1998;171:137-143. 30. Campo E, Munoz J, Miquel R, Palacin A, Cardesa A, Sloane BF, Emmert-Buck MR. Cathepsin B expression in colorectal carcinomas correlates with tumor progression and shortened patient survival. Am J Pathol 1994;145:301-309. 31. Khan A, Krishna M, Baker SP, Banner BF. Cathepsin B and tumor-associated laminin expression in the progression of colorectal adenoma to carcinoma. Mod Pathol 1998;11:704-708. 32. Su LK, Kinzler KW, Vogelstein B, Preisinger AC, Moser AR, Luongo C, Gould KA, Dove WF. Multiple intestinal neoplasia caused by a mutation in the murine homolog of the APC gene. Science 1992;256:668-670. 33. Levy DB, Smith KJ, Beazer-Barclay Y, Hamilton SR, Vogelstein B, Kinzler KW. Inactivation of both APC alleles in human and mouse tumors. Cancer Res 1994;54:5953-5958. 34. Luongo C, Moser AR, Gledhill S, Dove WF. Loss of Apc+ in intestinal adenomas from Min mice. Cancer Res 1994;54:59475952. 35. Oshima M, Dinchuk JE, Kargman SL, Oshima H, Hancock B, Kwong E, Trzaskos JM, Evans JF, Taketo MM. Suppression of intestinal poiyposis in Apc delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell 1996;87:803-809. 36. Shoemaker AR, Moser AR, Dove WF. N-ethyI-N-nitrosourea treatment of muttiple intestinal neoplasia (Min) mice: age-related effects on the formation of intestinal adenomas, cystic crypts, and epidermoid cysts. Cancer Res 1995;55:4479-4485. 37. Shibata H, Toyama K, Shioya H, Ito M, Hirota M, Hasegawa S, Matsumoto H, Takano H, Akiyama T, Toyoshima K, Kanamaru R, Kanegae Y, Saito I, Nakamura Y, Shiba K, Noda T. Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 1997;278:120-123. 38. Taipale J, Keski-Oja J. Growth factors in the extracellular matrix. FASEB J 1997;11:51-59. 39. Becker A, Hessenius C, Licha K, Semmler W, Wiedenmann B, Groetzinger C. Receptor targeted optical imaging of tumors with near infrared fluorescent ligands. Nat Biotech 2001;19:327331. 40. Gross LA, Baird GS, Hoffman RC, Baldridge KK, Tsien RY. The structure of the chromophore within DsRed, a red fluorescent protein from coral. Proc Natl Acad Sci U S A 2000;97:1199011995. 41. Contag P, Olomu I, Stevenson D, Contag C. Bioluminescent indicators in living mammals. Nat Med 1998;4:245-247 42. Ntziachristos V, Yodh AG, Schnall M, Chance B. Concurrent MRI 414 43. 44. 45. 46. 47. MARTEN ETAL. and diffuse optical tomography of breast after indocyanine green enhancement. Proc Natl Acad Sci U S A 2 0 0 0 ; 9 7 : 2 7 6 7 - 2 7 7 2 . Acosta MM, Boyce HW. Chromoendoscopy--where is it useful? J Clin Gastroenterol 1998;27:13-20. Axelrad AM, Fleischer DE, Geller AJ, Nguyen CC, Lewis JH, AIKawas FH, Avigan MI, Montgomery EA, Benjamin SB. High-resolution chromoendoscopy for the diagnosis of diminutive colon polyps: implications for colon cancer screening. Gastroenterology 1996;110:1253-1258. Schomacker KT, Frisoli JK, Compton CC, Flotte TJ, Richter JM, Deutsch TF, Nishioka NS. Ultraviolet laser-induced fluorescence of colonic polyps. Gastroenterology 1 9 9 2 ; 1 0 2 : 1 1 5 5 - 1 1 6 0 . Wang TD, Van Dam J, Crawford JM, Preisinger EA, Wang Y, Feld MS. Fluorescence endoscopic imaging of human colonic adenomas. Gastroenterology 1 9 9 6 ; 1 1 1 : 1 1 8 2 - 1 1 9 1 . Wang TD, Crawford JM, Feld MS, Wang Y, Itzkan I, Van Dam J. In vivo identification of colonic dysplasia using fluorescence endoscopic imaging. Gastrointest Endosc 1 9 9 9 ; 4 9 : 4 4 7 - 4 5 5 . GASTROENTEROLOGY Vol. 122, No. 2 Received June 27, 2001. Accepted October S, 2001. Address requests for reprints to: Ralph Weissleder, M.D., Ph.D, Center for Molecular Imaging Research, Massachusetts General Hospital, Building 149, 13th Street, 5403 Charlestown, Massachusetts 02129. e-mail: weissleder@helix.mgh.harvard.edu; fax: (617) 726-5708. Supported by National Institute of Health grants P50 CA86355, R33 CA88365, and N01-C097065; the German Research Foundation (K.M. and C.B.); and a DFCI grant "National Colorectal Cancer Research" (K.K.). Drs. Marten and Bremer contributed equally to this article. We thank Anja Siermann for her excellent technical assistance in performing the animal dissections and the RT-PCR, and Colin Martin for valuable advice on performing the RT-PCR. We would like to acknowledge the kind gift of some APC Min mice from Dr. Christoph Peters, University of Freiburg, Germany, and Dr. Umar Mahmood for providing the optical imaging system as well as advice in data acquisition and interpretation.