The in vivo assessment of mouse colorectal anatomy and function is challenging for all imaging modalities but could potentially provide new insights into disease progression and aid in the evaluation of new therapeutics for clinical use. Computed tomography (CT) colonography has shown great promise in the detection of colorectal polyps in both humans and mice and serves as an accurate high-resolution anatomical screening tool but provides no functional information relevant to the underlying tumor physiology (1,2). Conversely, optical and positron emission tomography (PET) imaging can interrogate the underlying biochemical processes of tissue function with the use of targeted contrast agents but have a limited spatial resolution (3-5). MRI has only recently been used to study the mouse colon but could potentially bridge the gap between these imaging paradigms by providing high-resolution anatomical and dynamic contrast-enhanced functional images of the colon. In contrast to conventional imaging-based colonography, which yields a morphological assessment of the colon, a “functional colonography” could provide a physiological characterization of the colon tissue. The use of dynamic contrast-enhanced (DCE) MRI methods and standard gadolinium (Gd)-based agents, such as Gd-DTPA, has the potential to measure colonic tissue contrast agent kinetic parameters that have been shown to differentiate malignant from benign tumors (6), aid in tumor staging (7), and monitor treatment response (8) in other disease states. With targeted Gd-based agents this approach could potentially report on gene activity, cell receptor expression, and/or other biochemical processes in colon tissue (9-11). Functional MRI colonography has great potential for the physiological assessment of murine models of colon cancer. One such model, the Min (multiple intestinal neoplasia) mouse, is an appropriate model for the study of earlystage colon tumor progression as a result of an APC mutation similar to the human Familial Adenomatous Polyposis Coli (FAP) syndrome that predisposes to colon cancer (12). These mice contain a truncating mutation of the Apc tumor suppressor gene and spontaneously develop adenomas in both the small and large intestine. This murine model is increasingly utilized to evaluate the efficacy of colorectal cancer treatments (13,14). Treatment of Min mice with the colitis-inducing agent dextran sodium sulfate (DSS) results in a greater number specifically of colonic as opposed to small intestinal polyps (15). Additionally, the DSS-elicited polyps are generally larger and more advanced than colonic polyps in non-DSS treated mice. The noninvasive, serial, and functional characterization of the microenvironment of these polyps could offer new insights into their progression and potentially provide a means to identify polyps that will undergo malignant transformation. In both humans and animals the detection of colon polyps using MR colonography requires clear delineation of the colon wall from bowel lumen and the capability to differentiate between polyps and artifacts of residual stool or air bubbles. With T1-weighted imaging, high contrast between the colon wall and the lumen can be achieved by enhancing either the colon lumen using Gd-doped enemas (bright lumen MRI) or the colon wall by intravenous administration of paramagnetic contrast agents and water enemas (dark lumen MRI). With bright lumen MRI, colonic polyps appear as dark filling defects within the bright colon lumen but differentiating these masses from air bubbles or residual stool is difficult and often requires multiple scans with the patient in different positions (16). To overcome this limitation while maintaining high contrast between the colon wall and lumen, dark lumen colonography was proposed and has been shown to be highly accurate for the detection of colorectal masses (17,18). Dark lumen MRI relies on minimizing the signal intensity of the intestinal lumen and its contents while enhancing the signal from the lumen wall and any masses using intravenous injections of paramagnetic contrast agents. Clinically, two dark lumen strategies have been developed to enhance the contrast between the colon wall and the lumen on T1-weighted images, including backfilling the colon with water or air following an enema (17,19,20) or alternatively adding contrast agents (typically barium sulfate) to the food that reduces the MRI signal from feces (fecal tagging) (21). Currently, two studies have adapted these methods for use in mouse models. Using the fecal tagging approach with barium sulfate, Larsson et al. (22) were able to differentiate between healthy and inflamed colon tissue in a mouse model of colitis. Hensley et al. (23) identified colon polyps in a Min mouse model as small as 1.5 mm in diameter using a 7T animal scanner by first cleansing the bowels and then inserting a polyurethane tube filled with Gd-DTPA fully into the colon, which assisted in the identification of the colon on the MRI images. Despite the success of these early studies, these methods are not optimal for functional colonography. To serially assess tumor growth and evaluate treatment response, it is important to identify lesions and changes in their size on a scale smaller than 1.5 mm. Also, given that the location and number of polyps is unknown prior to a study it would be advantageous to clearly and repeatedly delineate all portions of the colon. Without distending the colon this is a challenging problem given that mucosal folds and redundancies of the colon can be unpredictably flattened, hindering both the detection of polyps and the accurate determination of their dimensions. We propose that these issues can be minimized if the colon is fully distended using a filling agent prior to each scan, similar to the original dark lumen MRI methods that require patients to retain 2-2.5 L of enema solution during the examination (19). Furthermore, although water enemas do reduce the lumen signal intensity, we propose that the residual signal seen using water enemas can be effectively eliminated with the use of nonhydrogenous perfluorinated oil enemas. Therefore, the goals of this study were to demonstrate: 1) the potential of perfluorinated oils as colon-filling agents for dark lumen MRI mouse studies, 2) the feasibility of dark lumen MRI to serially assess polyp growth, and 3) the feasibility of dark lumen DCE-MRI to characterize the physiology of colon polyps and the colon wall.