The APC emphasizes quality control in all aspects of its functions. All data is examined and quality controls performed in a manner that is specific for the multiple core methodologies. Aspects of these procedures are discussed below in the context of provided core services.
I. Whole-body composition analysis
- Chemical carcass analysis (CCA). CCA is still the gold standard for the determination of body composition. The APC is well versed in this method and still uses CCA as the standard for validating new instruments and techniques. In addition, this method is useful when an investigator's animals have been killed and frozen prior to analysis. Methods: Mouse carcasses are dried to constant mass at 65ºC for the determination of body water content. The dried carcass is then homogenized, and samples (2–5g; analyzed in duplicate) are extracted with petroleum ether in a Soxhlet apparatus. Defatted, dry samples are then burned in a muffle furnace (600°C for 4 hours) to determine ash weight. Total body water is equal to the difference in weight of the carcass before and after drying. Fat weight is calculated as the percent of sample extracted times the carcass weight. Fat-free mass is calculated as the percent of sample not extracted times the carcass mass. Rat carcass analysis is conducted according to the procedure of Harris and Martin.
- Dual-energy X-ray absorptiometry (DXA). DXA has revolutionized the measurement of body composition, not only in humans but also in small animals. The APC was the first to successfully validate the GE-Lunar PIXImus and the Norland pDEXA Sabre. The technique allows for the rapid determination of fat, soft-lean tissue, and bone (bone mineral content and density) in vivo in small rodents, allowing for repeated measures of the same animal. Methods: After a 3-hour fast, animals are anesthetized using isoflurane and placed on the image-positioning tray in a prostrate fashion with front and back legs extended away from the body. Image acquisition is obtained in slightly less than 5 minutes. Calibration of the instrument is conducted as suggested by the manufacturer. An aluminum/Lucite phantom (TBMD = 0.0594 g/cm2, percent fat = 12.4%) is placed on the specimen tray and measured 25 times without repositioning. Thereafter, the phantom is analyzed daily before animal testing for quality control purposes. The APC validates new versions of the software as they become available. Similar methods are used for the determination of body composition of rats using the GE-Lunar Prodigy DXA and small animal software as we have previously validated.
- Quantitative magnetic resonance (QMR). With the acquisition of the Echo Medical 3-in-1 system in the spring of 2006, the laboratory has begun to offer the determination of fat and lean tissue in vivo in animals as small as 10mg up to 100gm (see validations below). The organism of interest (from 10 fruit flies up to a 100g mouse) is placed into one of the three differently-sized holders (biopsy 0–20mg, tissue 0.2–10.0g, and mouse holders 15–100g) and inserted into the instrument. The organism does not need to be anesthetized. Measurements take from 80 seconds for mouse-sized organisms to 9 minutes for fruit flies (10mg). We have recently published two abstracts on the validation of the instrument for fruit-flies and excised liver samples. The use of QMR for mouse studies has been validated and published. A second QMR instrument has recently been added for animals ranging in body weight from 100–900g.
- In vivo and ex vivo imaging of small rodents using micro-computed tomography (µCT). Unlike DXA and QMR, µCT can determine both regional fat distribution and the different components of lean tissue. However, unlike DXA and QMR, the quantitative analysis of µCT data can be labor intensive and much developmental work remains. Quantitatively, the core currently can determine fat pad volume, density, and hence mass of mouse and rat fat pads in vivo. However, the analysis is quite labor intensive and quantifying even a few fat pads requires up to ½ day of technician time. Volumes and masses of other structures such as tumors, and organ and muscle mass can also be determined. However because of the time commitment, the core currently does not offer the analysis as a service, instead we train one of the investigators staff or trainees to conduct the analysis. (The core conducts the actual scans.) The lab is currently working on software that will automate the process and speed up analysis time (see developmental work below). Methods: All scans are performed using the MicroCAT II computed tomography scanner (Imtek Inc., Knoxville, TN) and the manufacturer's phantom for quality control. Mice are anesthetized using sodium pentobarbital and placed on the scanning bed in a prostrate position. Each scan takes approximately 20 minutes. Files from the CT are reconstructed using the RVA reconstruction software, which creates a series of slices for the Amira advanced visualization, data analysis and geometry reconstruction software. Fat and lean tissue show up as different shades of grey (attenuation) due to their differing densities. Thresholds are determined for the fat and then the different fat pads are outlined in all the slices. A three-dimensional reconstruction is then performed on these outlined fat pads, which gives the volume and average density, from which fat mass can be obtained.
- Measurement of in vivo liver fat using µCT. We have validated and published the technique of determining in vivo liver fat using µCT (18) thereby allowing serial determinations within an animal. Because this quantitative analysis is not as labor intensive as the fat pad analysis, the APC staff conduct the analysis for the investigator. Methods: Scans are conducted as above. For the analysis, a liver section five slices distal to the lung is outlined and the average attenuation value is obtained. That value is then converted into % liver fat based on our previous validation studies.
II. Comprehensive assessments of metabolic rate (indirect calorimetry), food intake, fecal output, and activity
The core has a complete mouse metabolic phenotyping setup that allows for the determination of oxygen consumption, carbon dioxide production, locomotor activity, and food intake.
III. Glucose homeostasis
- Insulin tolerance test (ITT). After a 4-hour fast, mice or rats are weighed and placed in a clear mouse (or rat) restraint tube, which allows easy access to the tail for blood sampling. Blood glucose is analyzed using the Hemocue glucose 201 analyzer and cuvettes, which need 8–10 ul per sample. Insulin is injected i.p. (0.75–1.5 U/kg), and blood glucose is measured at multiple time points at baseline and post injection:
- Glucose tolerance test (GTT). After the fast, mice or rats are weighed and placed in clear restraint tubes, which allows easy access to the tail for blood sampling. Glucose is injected i.p. (1–2 g/kg) after the baseline glucose measurement, and blood glucose is measured at multiple time points post injection. Blood samples can be taken during the GTT for the determination of insulin concentration by radio-immunoassay.
IV. Cardiovascular assessments
The DRC core will provide state-of-the-art methodology in blood pressure and myocardial function to accommodate a substantial number of DRC members interested in diabetic cardiomyopathy or in the vascular effects of various perturbations, animal models, insulin resistance, and transgenes. For in vivo measurements of LV function, animals will undergo echocardiographic examinations (Figure 2) and hemodynamic assessments to obtain LV dimensions and LV end-systolic stress-volume and LV end-diastolic pressure-volume relations using the impedance catheter. Mouse studies are performed with the Visualsonics Echo/Doppler system with a 40mHz transducer and the rat studies are performed with the Agilent 5500 Echo/Doppler.
- Systolic function. In addition to fractional shortening, we measure the relationship of heart rate corrected velocity of shortening (VCFr) to end-systolic stress and the relationship between end-systolic stress and end-systolic dimension. For example, in order to asses these relationships we generated a regression line with 95% confidence using 33 age-matched sham animals and compared this data to ACF rats at 4–21 weeks of aorto-caval fistula (ACF). In shams there was a highly linear inverse relationship between VCFr and end-systolic circumferential stress. As can be seen in Figure 3, there is a spectrum of systolic function at the 4 and 8 week time points. VCFr is inappropriately low for the level of end-systolic stress in some of ACF animals, in spite of normal fractional shortening. In a similar fashion, there is an inappropriate end-systolic dimension for the level of end-systolic circumferential stress in some 4- and 8-week ACF rats. The stress/shortening relationship represents a single point analysis that assumes a zero intercept and is problematic especially in the volume overloaded heart. To better define in-vivo LV contractile function, we will use the impedance catheter combined with inferior vena cava occlusion to achieve multiple loading conditions as demonstrated in Figure 4 in a control and and a diabetic rat.
- Diastolic function. To define changes in LV chamber stiffness during the progression of VO and PO, we calculate diastolic pressure-volume relations from the impedance catheter. Chamber stiffness is defined by analyzing the curvilinear diastolic pressure-volume relation and fitting data to the equation p = Ae bv + C, where b is the exponential chamber stiffness constant. Diastolic pressure and volume can be measured throughout diastole from a single cardiac cycle. However, chamber stiffness is more accurately defined in the remodeled heart by multiple end-diastolic points using the impedance catheter combined with inferior vena cava occlusion to obtain an end-diastolic pressure-volume relationship over a range of end-diastolic volumes. This beat-by-beat analysis cannot be adequately obtained using echocardiography to assess volume. Myocardial stiffness will be calculated from conductance catheter derived volumes combined with echocardiographic-derived mass. Because LV mass is constant throughout the cardiac cycle and over acute changes in loading conditions, LV dimension and wall thickness can be calculated at each conductance catheter volume by measuring LV mass, dimension, and thickness at one LV volume and dimension and thickness at each new volume produced by IVC occlusion. These techniques will be used to calculate myocardial stress and strain at multiple end-diastolic values and fit these values to the exponential equation s = A e a e + C, where s = stress, e = strain, and a is the exponential myocardial stiffness constant. Thus, we have applied state-of-the-art methodology to follow LV chamber and myocardial stiffness in the diabetic hearts.
- Vascular compliance. Vascular compliance is determined from the pressure-volume data set.
- Blood pressure. The cardiac imaging component also is set up for chronic telemetry for conscious blood pressure monitoring in both mice and rats. In addition, this allows for an indwelling, high-fidelity aortic pressure that can be obtained with echocardiography in a serial fashion in mice and rats. The peak systolic pressure and aortic dicrotic notch pressure (equivalent to LV end-systolic pressure) can be combined with simultaneously obtained echo dimensions to derive LV end-systolic wall stress shortening relations in defining systolic function in mice and rats in a serial fashion.
V. Optical Imaging
- Bioluminescence and fluorescence. Bioluminescence imaging will be applied to serially evaluate luciferase-positive cells, including immune cells, and will be used to assess dynamic molecular signaling in vivo in real time (25). An example is shown in Figures 4 and 5 below for activation of growth hormone signaling in the liver (collaboration between Stuart Frank and Kurt Zinn). A luciferase reporter gene construct containing a GH response element was expressed in the river of nude mice via an adenoviral vector (Ad-GHRE-Luc) (Fig 4). Figure 5 demonstrates dynamic dose-responsive activation of ligand-mediated activation of the GH pathway in the liver of intact mice by bioluminescence imaging. In vitro imaging of cells growing in cell culture plates will also be supported, especially to validate in vivo reagents. Fluorescence imaging will be applied in a similar manner to detect cells or receptors. For example, Fig 6 shows pancreatic islets, transfected ex vivo with Ad-Luc, transplanted into the liver of a rat. Bioluminescence can be used to track the location and viability of the transplanted islets over time. The fluorescent stereomicroscope will provide high spatial resolution, especially for surface measurements of fluorescence and during necropsy. The time domain fluorescence instrument will provide depth and concentration information for the fluorescent-labeled probes, as well as fluorescence lifetime and intensity measurements. Two IVIS-100 systems (Xenogen, Inc.) for bioluminescence imaging with upgraded capability for fluorescence imaging are available. The eXplore Optix system (ART/GE) with two pulse lasers (488 nm, 632 nm) was installed in 2005. A custom-designed fluorescent stereomicroscope was also built for live animal imaging. Techniques previously developed at UAB will be used to image live mice (26;27). Both the stereomicroscope and eXplore Optix are functional in the near infrared. Multiple objectives allow for a wide range of imaging, from the whole animal to monitoring individual cells. Appropriate filter modules in a turret allow quick change for different fluorophores.
Calibration standards will be imaged in established positions to insure a constant detection;signal under each condition. For fluorescence, commercially available fluorophore-loaded beads are used. Consistency of the illumination source will be monitored. For bioluminescence, a certified light source for absolute calibration will be applied. For all light-based methods, the intensity of signal per pixel will be determined using region-of-interest analyses.
Fig. 7. SPECT/CT scan (1mm slice) using omnipaque, 0.5 ml iv, and Tc-99m-labeled antibody (Tx3.833) targeting lung caveolae (in collaboration with Dr. Jan Schnitzer). The top panel of images from the enhanced CT scan shows the pulmonary vasculature and bronchial tree. The SPECT gamma images (middle panel) demonstrate localization of the labeled antibody. Merged images (bottom panel) show the anatomic localization of the antibody, with binding at high levels in the vascular tree. This minimally invasive technology can be readily applied to the liver and pancreas to localize labeled proteins, antibodies, and cells over time.
- Gamma-ray imaging, including SPECT. Radiolabeling of peptides, proteins, and viral vectors will be provided for in vivo imaging of receptors. Planar (static and dynamic protocols), SPECT, and SPECT/CT imaging studies of these radiolabeled materials will be conducted in appropriate animal models. Dr. Zinn will supervise radiolabeling of proteins and peptides (28-32). Proteins and peptides are modified with succinimidyl 6‑hydrazinonicotinate (HYNIC). The HYNIC-modified constructs are radiolabeled with Tc-99m using tricine as the transfer ligand and purified from non‑bound Tc-99m by G‑25 Sephadex size exclusion chromatography. I-125 labeling will be accomplished as needed with the Iodogen method. For radiolabeled proteins, care is taken (Scatchard Analyses, plate imaging assays) to validate that attachment of the radioisotope does not change receptor affinity. This technology has been applied to label peptide, antibodies, and soluble receptors. Core instruments include three gamma cameras for planar imaging; two are equipped with PC-based Numa acquisition systems. In addition, a SPECT/CT camera (X-SPECT, GammaMedica, Inc.) for 3-D rodent imaging was installed in 2004. As a research component, Dr. Zinn will develop the SPECT/CT for imaging pancreatic islets in collaboration with Dr. Marron (and the new T1DM research leader and $10 million dollar program in beta cell biology/immunology-see Administrative Component). Figure 7 above demonstrates some capabilities of this technology imaging labeled antibody in lung vasculature.
VI. Transgenic Animal/Embryonic Stem Cell Resource
Expertise related to transgenics. Transgenic animal modeling has become increasingly important for elucidating disease mechanisms and investigating new treatment approaches. With completion of the sequences of the human and mouse genomes, and the new resources available for genetic manipulation of the mouse genome to develop novel mouse models (e.g., tissue specific gene ablation, regulated expression of transgenes, etc.), the UAB Transgenic Mouse Facility (TMF) plays a pivotal role in supporting and enhancing research efforts by DRTC investigators. The overall goal of the TMF is to provide expert services to all UAB investigators for the production and storage of transgenic mice. The generation of transgenic mice is technically challenging and cost prohibitive for individual laboratories due to the trained staff required. For gene-specific manipulation in the mouse, several months to years of training are required to master the technical skills necessary to successfully produce ES cell lines that have both the desired gene-specific targeted alleles and also the capacity to produce a germ-line chimeric mouse for the perpetuation of the targeted allele in mice. The TMF is a fully functional transgenic mouse core that has supported 48 different federally-funded investigators at UAB in the last 5 years and which during the recent site visit by NIH, received an assessment of Outstanding Merit.
Services Offered by the UAB Transgenic Mouse Facility
- DNA microinjection (pronuclear)
- Gene Targeting of ES cells (with & without screening)
- ES cell microinjection (blastocysts)
- In vitro fertilization (IVF)
- Embryo cryopreservation
- Sperm cryopreservation
- Long-term storage of cryopreserved transgenic lines
- Assisted reproduction / rederivations
- Consultation & training
While previously the onus for all non-technical aspects of physically creating transgenic mice was primarily the responsibility of UAB investigators' individual laboratories, the TMF has expanded services offered in recognition of investigator stated needs. To this end, the TMF will provide DRC researchers (especially those with limited transgenic experience), with the following services: 1) assistance with construct preparation, 2) advice on husbandry and colony management, 3) genotyping of genetically manipulated mice, and 4) generation of transgenic/knock-out mice. Dr. Kesterson will meet with DRC investigators to help design the most effective transgene constructs, strategies for molecular analyses of transgenic mice, and breeding strategies required for perpetuation of transgenic mouse lines. Dr. Kesterson will also provide individual training related to database searches of gene trapping resources, mouse mutagenesis, and phenotyping resources. DRC investigators will be encouraged to use C57BL/6 mice as a reference strain to conduct studies on a defined genetic background most commonly utilized for many diabetes and obesity related projects. The UAB TMF has well-established procedures for developing C57BL/6 transgenic mouse models. Currently, approximately 70% of all pronuclear injections use C57BL/6 eggs (with founder generation rates of 16-18% of all pups born), and the TMF has obtained and tested several C57BL/6 ES cell lines. Primogenix C57BL/6 ES cells have been utilized to produce a novel type II activin receptor B (ActRIIB) knockout mouse in less than 6 months (Kesterson, unpublished), as well as several other knock-in mice. Importantly, in-house screening services to identify ES cells that have undergone correct homologous recombination at the target loci will be provided to DRC investigators.