Water and Lipid MRI of the Xenopus Oocyte Jonathan V. Sehy, 1 Joseph J.H. Ackerman, 2,3,4 and Jeffrey J. Neil, 3,5 * Oocytes of Xenopus laevis are large, single cells that provide a promising model system for the exploration of the MR biophys- ics fundamental to more complex living systems. Previous stud- ies have generally employed 2D spin-echo sequences with an image slice thickness greater than the thickness of the cellular volumes of interest. Also, the large cytoplasmic lipid signal has typically been ignored. This study describes separate, high- resolution 3D measurements of the water and lipid spin densi- ties, T 1 and T 2 relaxation time constants, and the water appar- ent diffusion rate constant (ADC) in the Xenopus oocyte without significant partial volume artifacts. The lipid spin-density and values for water MR properties varied monotonically from the vegetal to animal poles, indicating that the border between the poles is not sharply demarcated. Regional water MR property values correlated with lipid signal intensity. Lipid-specific im- aging is shown for which water suppression is achieved via high diffusion weighting in the imaging sequence. Magn Reson Med 46:900 –906, 2001. © 2001 Wiley-Liss, Inc. Key words: Xenopus laevis oocyte; lipid imaging; relaxation; apparent diffusion coefficient The Xenopus laevis oocyte is a well-established cell model used in many branches of modern experimental biology. This versatile cell provides a promising model platform upon which to explore fundamental aspects of the MR biophysics underlying more complex living systems. In particular, the ability to spatially resolve the intracellular compartment allows the direct testing of hypotheses re- garding various MR properties of the cytoplasm and nu- cleus. Recent efforts to integrate confocal microscopy and MR imaging will likely augment the capability of such experiments (1). Aguayo et al. (2) reported the first MR images of the Xenopus oocyte in 1986. Along with intriguing contrast between intracellular regions, the authors described a strong chemical shift artifact from cytoplasmic lipid. Posse and Aue (3) further characterized this lipid signal using a 4D spectroscopic imaging technique. Both studies local- ized the lipid signal to the cytoplasm, with little or no lipid signal arising from the nucleus. Since these initial studies, several investigators have calculated spin densi- ties and T 2 and/or T 1 relaxation time constants for the oocyte nucleus, animal pole, and vegetal pole in control or altered cells (4 – 6). The effect of the unknown, spatially varying lipid spin-density on these calculations, if any, has been largely ignored. Further, these studies have rou- tinely employed 2D spin-echo sequences with slice pro- files thicker than the cellular volumes of interest (VOIs), raising concerns regarding partial volume effects. This shortcoming is especially detrimental to studies of the nucleus, which has a diameter of only 300 m. Herein we present separate, high-resolution 3D mea- surements of the water and lipid spin densities, T 1 and T 2 relaxation time constants, and the water apparent diffu- sion coefficient (ADC) in the Xenopus oocyte without sig- nificant partial volume artifacts. We challenge the conven- tional binary segmentation of the oocyte cytoplasm into vegetal and animal poles and recognize cylindrical sym- metry about the vegetal–animal (V–A) axis. Lipid-specific imaging is demonstrated for which water suppression is achieved via high diffusion weighting in the imaging se- quence. We correlate the lipid signal intensity with the water MR parameters. MATERIALS AND METHODS Oocyte Preparation Portions of ovary were isolated from mature female Xeno- pus laevis (Xenopus Express) by conventional methods (7). Stage V oocytes were chemically defolliculated by 4 h of digestion with 0.2 mg/ml type I collagenase (Sigma, St. Louis, MO) in media (96 mM NaCl, 2.0 mM KCl, 1.8 mM CaCl 2 , 1.0 mM MgCl 2 , 5.0 mM HEPES pH 7.5, 2.5 mM sodium pyruvate, 50 units/ml penicillin, 0.05 mg/ml streptomycin) and stored in media at room temperature not more than 3 days prior to use. Healthy oocytes were selected under a dissecting microscope by morphology and pigmentation. MR Imaging All images and spectra were collected at room temperature (23°C) in a 4.7T Oxford Instruments magnet equipped with a 600 mT/m Magnex Scientific gradient set and a Varian (San Fernando, CA) UNITY INOVA console. A 2.0-mm in- side-diameter (ID) solenoid wrapped around a 1.6-mm ID capillary tube provided RF excitation and detection. Each oocyte was positioned in the coil by gravity settling against a polyurethane stopper. A syringe pump (Harvard Appa- ratus, Dover, MA) provided media perfusion at a flow rate of 0.3 ml/h, which is an average velocity of 0.010 m/ms. Water-specific images were collected using a 3D spin- echo sequence modified for microscopy (Fig. 1). A fre- quency-selective pulse at the beginning of the TE period was used to excite water spins while avoiding excitation of the large cellular lipid signal. Crusher gradients enclose the hard (broadband)  pulse. The phase-encode and read- out refocus gradients are placed just before the acquisition period to minimize unwanted diffusion weighting. Com- mon implementations of the 3D spin-echo sequence place 1 Program in Molecular Cell Biology, St. Louis, Missouri. 2 Department of Chemistry, Washington University, St. Louis, Missouri. 3 Department of Radiology, Washington University, St. Louis, Missouri. 4 Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri. 5 Department of Neurology, Division of Pediatric Neurology, St. Louis Chil- dren’s Hospital, St. Louis, Missouri. Grant sponsor: NIH; Grant number: NS35912. *Correspondence to: Jeffrey J. Neil, M.D., Ph.D., Biomedical MR Laboratory, Washington University School of Medicine, 4525 Scott Avenue, Room 2313, St. Louis, MO 63110. E-mail: neil@wuchem.wustl.edu Received 3 January 2001; revised 19 April 2001; accepted 18 May 2001. Magnetic Resonance in Medicine 46:900 –906 (2001) © 2001 Wiley-Liss, Inc. 900