Physiological roles of CIDEs CIDEA deficient mice exhibit lean phenotype and are obesity resistant 5

Chapter Chapter Phenotype analysis of Cidea null mice 71 Chapter Cidea is initially identified as a member of the CIDE family proteins which share homology with the N-terminal region of the DNA fragmentation factor (DFF40/CAD and DFF45/ICAD (Inohara et al., 1998). It was previously reported that over expression of Cidea induced caspase independent cell death (Chen et al., 2000), and Cidea induced cell death when overexpressed had been observed in various mammalian cell lines in our laboratory (Toh, S.Y. and Li, P., personel communication, unpublished data). In spite of this, the mechanism and importance of Cidea in development is not clear. The gene targeting strategy provided us an effective tool to dissect and analyse Cidea’s in vivo physiological roles. 4.1 Cidea expression is restricted to brown adipose tissue (BAT) To determine the tissue expression profile of Cidea, mouse tissue northern blot analysis was performed using the mouse Cidea cDNA as probe. Surprisingly high levels of Cidea expression were detected in brown adipose tissue (BAT) (Figure16A). More interestingly, Cidea mRNAs were detected only in differentiated brown adipocytes but not in undifferentiated preadipocytes (Figure 16B). The expression profile of Cidea closely resembles that of the BAT specific uncoupling protein (UCP1). On the other hand, Cideb is enriched in liver and kidney, and FSP27 levels are highest in WAT, with moderate levels in BAT and skeletal muscle. 72 Chapter A Cidea Cideb FSP27 UCP1 β-Actin B Cidea UCP1 β-Actin Figure 16 High levels of Cidea expression in BAT. (A) Tissue distribution of the Cidea, Cideb and FSP27 transcripts in adult mice analyzed by Northern blot analysis. Sample loading: lane1: BAT, lane2: liver, lane3: Kidney, lane4: skeletal muscle, lane5: heart, lane6: WAT, Lane7: Lung. 30µg of total RNA was loaded in each lane. β-Actin was probed as a loading control. UCP1 was used as a BAT marker. (B) RT-PCR analysis of Cidea expression in undifferentiated preadipocytes (lane 1) or differentiated brown adipocytes (lane 2). β-Actin was amplified as a template loading control. All three CIDE family members are therefore expressed in a tissue specific manner and in the tissues or organs that are important for energy metabolism. These expression patterns suggest that the physiological roles of the CIDE family members could be related to the regulation of fat metabolism. BAT is present in almost all mammals and is essential for the maintenance of body temperature in small mammals. It is a major site for cold-induced thermogenesis in 73 Chapter rodents and also contributes to diet-induced thermogenesis. This tissue is well developed around birth in large mammals, and throughout the lifespan in rodents. BAT is found in characteristic deposits scattered in specific areas of the body. The major deposits are at the interscapular, axillar, perirenal, thoracic, and between the neck muscles. The specific expression of UCP1 in brown adipocytes has been confirmed in several laboratories (Desautels and Dulos, 1988; Florez-Duquet et al., 1998; Monemdjou et al., 1999; Nedergaard et al., 2001; Nicholls, 2001). UCP1 was therefore chosen as a marker of BAT for comparative in situ hybridization and northern analysis of Cidea. A D E15 B E15 E E18 C E18 F 1W Cidea 1W UCP1 Figure 17 In situ hybridization analysis of Cidea expression. E15, E18 and 1W represent sections obtained from embryonic day 15, day 18 and postnatal 1-week mice. Sections A, B and C were probed with an anti-sense Cidea RNA probe and sections D, E, and F were probed with an anti-sense UCP1 RNA 35 probe. Probes were labeled with α- S-UTP. 74 Chapter The detail spatial and temporal expression profile of Cidea was further investigated by in situ hybridization analysis using mouse embryo sections. As shown in Figure 17A, Cidea expression was detected as early as embryonic day15 (E15) in the interscapular region of the embryo. These regions were also UCP1 positive (Figure 17D), suggesting that this region contains developing BAT. The expression of both Cidea and UCP1 remained prominent in the interscapular region of day 18 embryo (E18, Figure 17B & 17E) and postnatal 1-week mice (1W, Figure 17C & 17F). Like UCP1, Cidea expression is therefore highly tissue specific and restricted to BAT. 4.2 Cidea expression increases in aged mice The expression of Cidea in BAT was further examined by northern blot analysis using RNA isolated from BAT of wild type mice at different ages and mice subjected to cold treatment and fasting / feeding with a high-fat diet. While UCP1 was expressed at very high level in the BAT of one-month old mice, its expression level decreased with increasing age (Figure 18A). The level of UCP1 mRNA in BAT of nine-month old mice was four-fold lower than that found in one-month-old mice (Figure 18B). These data suggest that uncoupling activity and energy expenditure decreased with increasing in age, consistent with increased lipid accumulation in the BAT from aged animals (Figure 27C). This result is in agreement with previous research results (McDonald and Horwitz, 1999; Scarpace and Matheny, 1996; Scarpace et al., 1994). Interestingly, Cidea mRNA levels, on the contrary, were maintained at high levels throughout adulthood and actually increased with age. Amongst the ages examined, the highest expression was detected in BAT of nine months old mice (Figure 18A). Cidea mRNA level from BAT of nine-month-old mice was 1.5 fold (Figure 18B) higher than that from one-month-old mice. Increased UCP1 75 Chapter mRNA levels were observed in BAT of mice subjected to cold treatment, overnight fasting or feeding with a high-fat diet (Figure 18C). However, Cidea mRNA levels in BAT of these treated mice were maintained at roughly similar levels compared to mice of the same age. The contrasting changes in Cidea levels with age compared to that of UCP1 suggest that it may play a role in influencing the metabolic rate in aged mice which have lower basal metabolic rate and higher tendency to become obesity. 76 Chapter Figure 18. Increased levels of Cidea expression in adult and aged mice. (A) Northern blot analysis of RNA isolated from BAT of wild type C57/BL6J mice. Right panel: Total RNA was isolated from month old, 24 hrs cold exposed (cold), month old, 24 hrs fasted (fast), month, months, months, months and months old mice. Left panel: Total RNA was extracted from months old mice fed with normal chow (5% fat) or a high fat diet (37.5% fat). The blot was hybridized with Cidea or UCP1 specific cDNA probes. β-Actin was probed as an RNA loading control.(B-C) Northern blots were exposed to a Kodak phosphoimager screen and bands were quantified by a Bio-rad FX Imager. RNA levels were normalized against that of βactin and presented as fold difference for mice at different ages (B) or cold exposed, fasted and high fat diet treated (C). Although BAT and WAT exert opposite functions in energy expenditure, with BAT consuming energy and WAT storing energy (see Introduction), most of the genes 77 Chapter involved in adipocyte differentiation and lipid metabolism are expressed in both tissues. UCP1 is the only gene identified thus far to be specifically expressed in BAT. Using Northern blotting and in situ hybridization analyses, Cidea is thus identified as another BAT specific gene, showing a highly restricted expression pattern in BAT but not in WAT. Cidea is expressed as early as embryonic day 15 in the interscapular region of the embryo, consistent with the initial appearance of BAT during development (Barak et al., 1999). The BAT specific expression of Cidea suggests that it may have a role in the tissue. The increased levels of UCP1 expression detected in young animals as well as animals subjected to cold treatment, fasting and feeding with high fat diet are interesting. These represent situations in animals related to higher metabolic rates and energy expenditure. In contrast, higher levels of Cidea expression were associated with aging; a state characterized by lower metabolic rates and increased risk of obesity. A hypothesis that emerges at this stage is that Cidea may have a role in the down regulation of metabolic activity in BAT. The Cidea-/- or null mice we generated provided an excellent tool to test this hypothesis. 78 Chapter 4.3 Cidea null mice not exhibit a difference in differentiation and cell death compared to wild type mice. Figure 19 Differentiation markers RNA levels in BAT and WAT. Left panel: Northern blots analysis of total RNAs extracted from BAT of weeks old Cidea+/+ and Cidea-/- mice. LPL and PPARγ represent lipoprotein lipase and peroxisome proliferator-activated receptor-γ respectively. Right panel: RT-PCR analysis of genes expressed in WAT, ALBP (adipocyte lipid-binding protein) and Adipsin is an adipose tissue specific protein. Each sample is a pool of RNAs extracted from tissues of eight mice. In characterizing the Cidea-/- mice, we first asked if BAT cells from these mice show any significant difference from that of wild type mice. As shown in Figure 19, Cidea-null mice exhibited higher levels of mRNA for several adipocytes markers, including UCP1 (45% higher), FSP27 (138% higher), lipoprotein lipase (LPL, 24% higher) and peroxisome proliferator-activated receptor γ (PPARγ, 33% higher) (Figure 19, left panel). Expression levels of several fat cell specific markers (as analysed by RT-PCR) in WAT were largely similar between wild type and Cidea-null mice (Figure 19, right panel). 79 Chapter Figure 20 BAT protein expression levels. Western blot analysis of UCP1, COX IV and Cytochrome C was carried out using total cell lysate of BAT while that for Cidea was carried out using heavy membrane preparation of BAT isolated from three, six, twelve and fifteen-months old Cidea+/+ and Cidea-/- mice. Each protein sample is extracted from a pool of eight mice BAT. As shown in Figure 20, protein levels for UCP1, cytochrome c, COX IV and βActin were also similar in BAT of wild type and Cidea-null mice at different ages. These data suggest that BAT and WAT of Cidea-null mice had no discernible alterations in terms of the expression of development or differentiation specific markers. Similarly, the brain, liver and skeletal muscles of Cidea-null mice had no obvious anatomical and morphological defects. Since liver is the most important organ for lipid metabolism, the liver morphology of high fat diet treated mice was investigated. As shown in Figure 21, no significant histological difference was observed between the liver sections of wild type and Cidea-null mice. 80 Chapter RT A Recover C +/+ B D -/- Figure 29A Images of H&E stained BAT. Bright-field micrographs (H&E staining) of wild type (A&C) and Cidea null BAT (B&D) housed either at room temperature (RT) (A&B) or after recovery at 22ºC for one week from eight days of exposure at 4ºC (C&D). BAT is processed for paraffin sectioning as described in section 2.4.1. 90 Chapter Figure 29B Images of H&E stained BAT. Bright-field micrographs (H&E staining) of paraffin sections of BAT from Cidea+/+ (WT), Cidea+/-(Hez), Cidea-/-(Mu) housed at room temperature (No), one day at 4ºC (1D), days at 4ºC (3D) or days at 4ºC (8D). 91 Chapter The lipid content decreased faster in Cidea-null compared to wild type mice was further substantiated by the morphological changes of BAT under different time period of cold exposure shown in Figure 29B. To confirm that lipolysis in BAT is enhanced in Cidea-null mice, the amounts of glycerol and FFA released from explants of BAT and WAT that were maintained in vitro was determined. The amount of glycerol released from BAT of Cidea-null mice was 67% higher than that of wild-type mice (0.5830 ±0.01580 versus 0.9743 ±0.03228 µmol g-1 h-1) at basal level and 198% higher than that of wild-type mice (1.824 ±0.05541 versus 5.440 ±0.1473 µmol g-1 h-1) in the presence of isoproterenol (β3 agonist, Figure 30A), a clear indication of enhanced lipolysis. The amount of non-estherified fatty acids (NEFA) released from BAT of Cidea-null mice was lower in both basal and isoproterenol-treated tissues compared to that of wild-type mice (Figure 30A). As the complete lipolysis of mol of triacylglycerides yields mol of glycerol and three mols of fatty acids, the lower levels of fatty acid released from BAT suggest that Cidea-null mice may have increased fatty acid recycling or fatty acid oxidation activity. On the contrary, no difference in the release of glycerol or NEFA was observed in WAT from wild type and Cidea-null mice (Figure 30B). 92 Chapter Figure 30 Increased lipolysis in BAT of Cidea null mice. Each genotype seven 5-6 weeks old male mice were used in the test. BAT or WAT were dissected out and used for in vitro lipolysis assay. The details are described in Chapter 2, section 2.7.6. The results shown thus far indicated that Cidea null mice possessed a higher thermogenic activity and were able to maintain a higher body temperature after cold exposure. Increased body temperature that results from hyper-thermogenic activity was also seen in mutant mice harboring a deletion of the RII β subunit of protein kinase A, in which the mRNA and protein levels of UCP1 are drastically up-regulated (Cummings et al., 1996). Histologically, brown adipocytes of wild type and mutant mice appeared to be very different. While wild type brown adipocytes contained large lipid droplets, Cidea 93 Chapter null brown adipocytes (at 40C) showed almost no lipid accumulation, suggesting that the tissue is hyperactive. The Cidea null animals were also more active in lipid metabolism because the overall fat content of their BAT was markedly less than wild type mice in aged animals or after animals were exposed to cold. This was further confirmed by the in vitro experiment using explants of BAT showing that the BAT of Cidea null mice was more sensitive to the β3- agonist isoproterenol (Figure 30). 4.6 Less adiposity and lean phenotype in Cidea-null mice Results presented thus far indicate that the basal metabolism and adaptive thermogenesis of Cidea null mice were significantly higher than that of the wild type mice. In order to further investigate the role of Cidea in regulating whole body weight and adiposity, the body weights of mice given normal chow and high fat diet treatment was examined. There was no significant difference in body weight between wild type and Cidea-null mice up to months of age when fed with either normal chow or a highfat diet (Table 1). The wet weight of other tissues, including liver, kidney, spleen and heart, was also similar in Cidea-null and wild-type mice (Table 1). However, Cidea-null mice had markedly less white adipose tissue (Figure 31) . The adiposity index (ratio of total body fat to body weight) was 9.05% and 3.47% (P < 0.001) for wild type and Cidea-null male mice, respectively, corresponding to a 62% reduction in Cidea-null mice compared to wild type (Figure 31A). 94 Chapter +/+ Male B A Normal Chow Male 0.4 Female 0.1 P[...]... Cidea- null mice Wild type and Cidea- null mice had similar levels of triglycerides when fed, but levels of triglycerides and NEFA were about 60% lower (P < 0.001) in Cidea- null mice than in wild-type mice when fasted (Figure 33C&D) This suggests that Cidea- null mice have a higher rate of lipid metabolism Notably, levels of plasma glycerol in both fed and fasted states were similar in wild type and Cidea- null... that of wild-type mice (0 .58 30 ±0.0 158 0 versus 0.9743 ±0.03228 µmol g-1 h-1) at basal level and 198% higher than that of wild-type mice (1.824 ±0. 055 41 versus 5. 440 ±0.1473 µmol g-1 h-1) in the presence of isoproterenol (β3 agonist, Figure 30A), a clear indication of enhanced lipolysis The amount of non-estherified fatty acids (NEFA) released from BAT of Cidea- null mice was lower in both basal and isoproterenol-treated... Time(mins) Figure 25 Indirect calorimetry methods for measuring mice energy expenditure Mice are kept in individual chambers seperately (A) in which the input and output of air are controlled and measured by gas analysers (B), the data are collected by computer(C) and reported as a curve shown in (D) 0. 75 * ** 50 0.70 RER VO2 (ml/kg/min) 75 25 0 0. 65 +/+ -/- 0.60 +/+ -/- Figure 26 Cidea null mice had increased... (Figure 30) 4.6 Less adiposity and lean phenotype in Cidea- null mice Results presented thus far indicate that the basal metabolism and adaptive thermogenesis of Cidea null mice were significantly higher than that of the wild type mice In order to further investigate the role of Cidea in regulating whole body weight and adiposity, the body weights of mice given normal chow and high fat diet treatment was... Cidea- null compared to wild type mice was further substantiated by the morphological changes of BAT under different time period of cold exposure shown in Figure 29B To confirm that lipolysis in BAT is enhanced in Cidea- null mice, the amounts of glycerol and FFA released from explants of BAT and WAT that were maintained in vitro was determined The amount of glycerol released from BAT of Cidea- null mice. .. wild type and Cidea- null mice up to 9 months of age when fed with either normal chow or a highfat diet (Table 1) The wet weight of other tissues, including liver, kidney, spleen and heart, was also similar in Cidea- null and wild-type mice (Table 1) However, Cidea- null mice had markedly less white adipose tissue (Figure 31) The adiposity index (ratio of total body fat to body weight) was 9. 05% and 3.47%... months, wildtype mice became obese, but Cidea- null mice had an adiposity index similar to that of wild-type mice fed with normal chow This represents a 36% difference (P < 0.001; Figure 31B) in body fat, suggesting that Cidea- null mice are protected against obesity induced by a high-fat diet Consistent with lower amount of WAT, Cidea- null mice had 50 % less plasma leptin than wild-type mice when fed with... al., 1995a) Alternatively, the less dramatic phenotype may be due to the compensating effect of FSP27, another CIDE family member, as it is also expressed in BAT Cidea knockout mice derived from pure C57BL/6J inbred background or mice contained Cidea and FSP27 double mutations can be generated in the future to address these concerns 4.7 Levels of glucose, FFA and triglycerides in Cidea- null mice Obesity. .. Chapter 4 Male +/+ Male -/- o Core Body Temp ( C) 40.0 37 .5 35. 0 32 .5 Female +/+ Female -/- 37 .5 o Core Body Temp ( C) 30.0 35. 0 32 .5 30.0 -1 0 1 2 3 4 5 6 7 8 9 10 Time (hrs) Figure 24 Core body temperatures before and after placement at 4 ºC from 22 ºC 8 four weeks old male and female mice of Cidea+ /+ or Cidea- /- were used for the measurement Mice kept at room temperature (22 ºC) were fasted for 12... phenotype in Cidea null mice Adiposity index (AI) of mice fed with either a normal chow (A) or a high fat diet (B) Number of mice examined in each group is indicated accordingly in the respective bars Upper pannel: photo comparing the gonadial fat pad of +/+ and -/- male mice White adipocytes were also smaller and more compact in the WAT of Cidea- null mice relative to wild-type mice (Figure 32, upper . sections of wild type and Cidea- null mice. Chapter 4 81 Figure 21 Light micrograph of liver sections of 9-month-old Cidea +/+ and Cidea -/- mice. The mice were fed with standard chow. the Cidea -/- mice, we first asked if BAT cells from these mice show any significant difference from that of wild type mice. As shown in Figure 19, Cidea- null mice exhibited higher levels of. hybridization and northern analysis of Cidea. Figure 17 In situ hybridization analysis of Cidea expression. E 15, E18 and 1W represent sections obtained from embryonic day 15, day 18 and postnatal