The Effects of Diet Composition on Body Fat and ...

Comparative Medicine Copyright 2011 by the American Association for Laboratory Animal Science

Vol 61, No 1 February 2011 Pages 31–38

The Effects of Diet Composition on Body Fat and Hepatic Steatosis in an Animal (Peromyscus californicus) Model of the Metabolic Syndrome Lisa Krugner-Higby,1,* Stephen Caldwell,3 Kathryn Coyle,5 Eugene Bush,6 Richard Atkinson,4 and Valerie Joers2 The objective of this research was to determine body composition, total fat content, fat distribution, and serum leptin concentration in hyperlipidemic (high responder, HR) and normolipidemic (low responder, LR) California mice (Peromyscus californicus). In our initial experiments, we sought to determine whether differences in regional fat storage were associated with hyperlipidemia in this species. To further characterize the hepatic steatosis in the mice, we performed 2 additional experiments by using a diet containing 45% of energy as fat. The body fat content of mice fed a low fat-diet (12.3% energy as fat) was higher than that of mice fed a moderate-fat diet (25.8% energy as fat). Total body fat did not differ between HR and LR mice. There was no significant difference between intraabdominal, gonadal, or inguinal fat pad weights. Liver weights of HR mice fed the moderate-fat diet were higher than those of LR mice fed the same diet, and the moderate-fat diet was associated with nonalcoholic fatty liver (NAFL). Mice fed the 45% diet had higher histologic score for steatosis but very little inflammatory response. Chemical analysis indicated increased lipid in the livers of mice fed the high-fat diet compared with those fed the low-fat diet. HR and LR mice had similar serum leptin concentrations. California mice develop NAFL without excess fat accumulation elsewhere. NAFL was influenced by genetic and dietary factors. These mice may be a naturally occuring model of partial lipodystrophy. Abbreviations: DEXA, dual-energy X-ray absorptiometry; HR, high responder; LR, low responder; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.

Obesity is a condition of excessive fat distribution. Lipoatrophy is the inability to store peripheral fat. The consequences of obesity and lipoatrophy are remarkably similar. Both obesity and lipoatrophy produce insulin resistance, fatty liver disease, hypertension, and type II diabetes mellitus, the features of the metabolic syndrome.17,18,28,32 Both obesity and lipoatrophy are considered to be consequences of the inability to expand the peripheral fat mass sufficiently to store excess energy. This concept is referred to as the ‘lipid overflow hypothesis.’32 Lipoatrophic syndromes may be generalized or partial, genetic or acquired. Genes associated with mutations leading to congenital generalized lipoatrophy syndromes include 1-acylglycerol-3-phosphate-O-acyltransferase 2, Berardinelli–Seip congenital lipodystrophy 2, and caveolin (CAV1).32 A total of 4 genetic loci have been identified that are associated with familial partial lipodystrophy: lamin A/C, peroxisome proliferatorsactivated receptor γ, vAKT murine thymoma oncogene homolog 2, and zinc metalloprotease.32 These previously identified genes account for more than 95% of cases of congenital generalized lipoatrophy.32 However, many patients with familial partial lipodystroReceived: 13 Apr 2010. Revision requested: 16 May 2010. Accepted: 03 Oct 2010. 1 The Research Animal Resources Center and the 2Wisconsin National Primate Research Center, University of Wisconsin-Madison, Madison, Wisconsin; 3University of Virginia Department of Medicine, Charlottesville, and 4Obetech, Richmond, Virginia; 5Office of the Provost, University of Kentucky, Lexington, Kentucky; and 6Abbott Global Pharmaceutical Research, Abbott Park, Illinois. * Corresponding author. Email: [email protected]

phy syndromes do not have mutations in the known genes, so additional involved genes remain to be identified and studied.18 The 2 mouse models of congenital generalized lipoatrophy are mice that lack the genes caveolin or 1-acylglycerol-3-phosphateO-acyltransferase 2. Mice null for caveolin resist diet-associated weight gain and have fat depots that are 50% smaller than those of controls but their blood glucose and insulin concentrations are normal.32 Mice that lack 1-acylglycerol-3-phosphate-O-acyltransferase 2 show 80% mortality within 3 wk of birth; surviving pups have diabetes and fatty liver disease.32 However, the high neonatal mortality is not a feature of the human form of the disease. Among the mouse models of familial partial lipodystrophy, mice null for lamin A die within 8 wk of birth with severe muscular dystrophy. These mice are reported to be lean, but the muscular dystrophy component is not a common feature of human lipodystrophy syndromes.32 In contrast, hepatic tissue-targeted ablation of peroxisome proliferators-activated receptor γ function leads to significant amelioration of hepatic steatosis in mice fed high fat, suggesting that expression of this gene is necessary for hepatic fat accumulation.32 In addition, AZIP-F1 mice have been used as a model of lipoatrophic insulin resistance and diabetes.30 Evidence from rodent models of lipodystrophy indicates that treatment with exogenous leptin will ameliorate the metabolic syndrome,11 and leptin therapy may be a promising treatment for

31

cm10000039.indd 31

3/4/2011 1:32:54 PM

Vol 61, No 1 Comparative Medicine February 2011

several types of congenital and acquired lipodystrophy in human beings.11 The California mouse (Peromyscus californicus) variably develops many features of the metabolic syndrome, including hypertriglyceridemia, hyperinsulinemia, islet cell hyperplasia, and (occasionally) type II diabetes in response to a diet containing a moderate amount of fat (26% energy as fat) and in the absence of the gross obesity and infertility expected with mutation of the known leptin receptors or leptin deficiency.20 California mice that are susceptible or resistant to hyperlipidemia after consuming a moderate-fat diet have been identified and can be selectively bred for research.24 California mice can be obtained from the Peromyscus Genetic Stock Center.14 Despite their widespread use in biomedical research and the numbers of genetically modified animals available, laboratory mice do not provide a complete representation of all of the diseases found in people. Other disease models, including other rodent models, are needed to study a broader range of human diseases.31 The current studies were undertaken to characterize the body composition, regional fat storage characteristics, and leptin concentrations of susceptible and resistant California mice in response to diets containing 12% or 26% of energy as fat. Additional experiments used a diet containing 45% energy as fat to reveal the extent of hepatic steatosis.

Materials and Methods

Animals. California mice (Peromyscus californicus) were obtained from the breeding colony in the Department of Psychology at the University of Wisconsin–Madison. All experimental procedures were approved by the Animal Care and Use Committee of the University of Wisconsin–Madison and were conducted in adherence with the Animal Welfare Act and Public Health Service policy. Quarterly sentinel surveillance was done by using Mus musculus. Sentinel mice were negative for: Mycoplasma pulmonis, Sendai virus, mouse hepatitis virus, pneumonia virus of mice, reovirus 3, Theiler virus, ectromelia virus, mouse adenovirus, polyoma virus, lymphocytic choriomeningitis virus, cytomegalovirus, murine rotavirus, murine parvovirus, and cilia-associated respiratory bacillus. No other parasites or bacterial pathogens were identified in sentinel mice. Mice were maintained in solidbottomed cages on hardwood bedding. Low-fat (12.3% of energy from fat or 4.5% by weight; the 12% diet; PMI, St Louis, MO), moderate-fat (25.8% of energy from fat or 11% by weight; the 26% diet; PMI), and high-fat (45% of energy from fat or 21% by weight; the 45% diet; Research Diets, New Brunswick, NJ) commercial diets and water were provided ad libitum (Figure 1). The source of fat for the 12% and 45% diets was beef tallow and a combination of beef tallow and soybean oil for the 26% diet.21 Mice were maintained on a 12:12-h light:dark cycle. High-responder (HR; mean serum triglyceride greater than 1000 mg/dL) and low-responder (LR; mean, mean serum triglyceride less than 300 mg/dL) mice were those comprising the highest and lowest quartiles, respectively, of the original population of 24 mice according to their serum triglyceride concentrations after 6 wk of consuming the moderate-fat diet.22,23 A breeding colony of California mice was established as previously described, and the offspring from this colony were used in body composition experiments in addition to the original population of 24 mice.24 Serum lipid status of the original 24 mice and the mice used for body composition analysis has been reported previously.24 Briefly, selective breeding of California mice based

on their serum triglyceride responses to dietary fat was used to establish the colony. Initial test mating was done on mice after feeding the 26% diet for 6 wk. Blood samples for determination of glucose, insulin, cholesterol and triglyceride concentrations were taken from the mice in the founder colony after a 4-h fast. High-, low-, and intermediate-responding mice were designated based on their triglyceride concentrations (greater than 1000 mg/dL, less than 400 mg/dL, and between 400 and 1000 mg/dL, respectively). Progeny from the original test matings were characterized similarly after they consumed the 26% diet for 12 wk after weaning. A larger colony of selectively bred California mice was established by test-feeding the 26% diet to 50 mice for 12 wk. At the end of the test-feeding period, a blood sample was taken from each mouse for serum triglyceride determination. Mice were divided into low, intermediate, and high serum lipid responders according to cut points similar to those used to categorize the original cohort of mice.24 In addition, serum from selected mice (n = 15) evaluated for hepatic steatosis underwent triglyceride analysis as described following.23 Serum triglyceride determination. Blood was obtained by cardiocentesis from pentobarbital-anesthetized mice, placed on ice, and centrifuged to collect the serum, which then was frozen at −70 °C until assayed. Samples for triglyceride analysis alone were obtained from unfasted mice. A commercial test kit was used to measure serum triglyceride (catalog TR0100-1KT , Sigma Aldrich, St Louis, MO). Measurement of body composition. Anesthesia in California mice was induced by isoflurane gas in a closed container and maintained by using isoflurane gas delivered by nose cone. Mice were placed on the densitometer (Lunar PIXImus, GE Healthcare, Piscataway, NJ) for scanning (software version 1.42.006.010, GE Healthcare). Scans lasted approximately 15 min. Body composition was done by using a densitometry scanner that has been validated for use in laboratory mice.25 Terminal measurement of body composition. At necropsy, the intestinal contents of the mouse carcasses were removed, and then the remainder of the carcass was frozen at −20 °C. The carcasses were autoclaved at 250 °C for 6 h in large, covered beakers. The resulting material was homogenized by using a large-bore homogenizer (PT 6000, Brinkman Instruments, Westbury, NY). The samples were weighed before and after autoclaving. The fat was extracted in methanol:chloroform, and then the carcasses were dried in a vacuum oven at 60 °C for 48 h. Triplicate samples were turned to ash for determination of mineral content.4 Body fat distribution in California mice. HR and LR mice were fed either the 26% or 12% diet for 18 wk. At necropsy, livers were scored in situ for gross evidence of hepatic lipidosis, including extension of the liver beyond the costochondral junction, pallor, and reticular pattern. Scoring was not performed blind to treatment condition, but gross scores were only one measure of hepatic steatosis used in these studies. Livers were removed and weighed at necropsy. The intraabdominal, gonadal, subscapular, and inguinal fat pads were removed and weighed. Histopathology. California mice were euthanized by using pentobarbital (greater than 120 mg/kg IP). Blood samples were collected terminally by cardiocentesis, and the right and left liver lobes were dissected free and fixed in 10% neutral buffered formalin for histopathologic evaluation. Formalin-fixed sections were stained with hematoxylin and eosin and scored by using a detailed scoring system.22 Serial sections were stained with Masson

32

cm10000039.indd 32

3/4/2011 1:32:54 PM

Body composition and fatty liver in California mice

Figure 1. Experimental design. 12% diet, low-fat diet; 26% diet, moderate-fat diet; 45% diet, high-fat diet.

trichrome to better quantify collagen deposition associated with fibrosis. Representative liver sections were stained with oil red O to visualize hepatic lipid. Slides were coded and scored according to standard criteria7,22 for steatosis, apoptosis, inflammation, and fibrosis by a boarded veterinary pathologist (KC), who was blind to treatment condition. Circulating leptin concentration. Blood samples were obtained by retroorbital bleeding or cardiocentesis from mice under deep pentobarbital anesthesia at the time of necropsy. Serum was separated from the cellular elements by centrifugation. The serum was removed, placed in a fresh tube, and frozen at −70 °C until assayed. Serum leptin concentration was assayed by using commercial kits (Murine Leptin ELISA, Crystal Chem, Downers Grove, IL). The results of the initial experiments to quantitate and characterize the amount and distribution of fat in California mice indicated that the hepatic steatosis phenotype observed in the previous experiments might become more extreme and manifest more rapidly if the diet contained more fat. The experiments using the 45% diet focused on hepatic steatosis, because fat is stored in the liver of mice. Groups of HR and LR mice (n = 8) were fed the 45% diet for 6 wk. The mice were euthanized, and the livers were collected, weighed, and fixed in 10% neutral buffered formalin. Sections stained with hematoxylin and eosin were scored for steatosis and inflammation as previously described.22 Additional groups of 4 LR mice each were fed either the 12% or 45% diet for 6 wk. The mice were euthanized, and liver samples were frozen at −80 °C. The livers were extracted and assayed for lipid, protein, and ash content as described following. Liver composition analysis. wet liver samples (1 g) were analyzed for relative lipid, protein, and ash compositions. Dry liver weights were determined by freeze-drying the wet liver samples for 72 h. The lipid portions of dry liver samples were separated from the protein and ash.16 The collected lipid fractions were dried under vacuum centrifugation and weighed to determine the lipid weight of dry liver samples. The weight of the protein and ash fraction of dry liver samples was determined by subtract-

ing the lipid fraction weights from the total dry liver weights. Results were expressed per gram of dry liver. Data analysis and statistics. ANOVA, Student t tests, and correlation analysis were performed on parametric data by using Excel software (Microsoft, Redmond, WA). Nonparametric tests including Spearman rank order correlation with covariance, Mann–Whitney test, and Kruskal–Wallis test were done by using SPSS software (SPSS Institute, Cary, NC). Odds ratios were calculated by using Epi Info (Centers for Disease Control, Atlanta, GA). Significance was inferred at an α level of less than or equal to 0.05. Data are given as mean ± SE.

Results

Antemortem and postmortem analysis of body fat. DEXA scans of adult California mice indicated that the mice fed the moderate(26%) fat diet acquire additional body fat as they age. California mice scanned at 1 y of age had significantly (P < 0.05) less body fat than the same group of mice when they were scanned again at 1.5 y of age (Figure 2 A), 3 wk before the mice were euthanized for terminal body fat analysis. A control group of age-matched (older than 1 y) California mice fed the low- (12%) fat diet were scanned at the same time. California mice fed the 26% diet had significantly (P < 0.05) less body fat than did control mice (Figure 2 B). There were no significant differences between male and female mice at either time point (1 or 1.5 y of age) or according to diet. Three weeks after the last DEXA scan, both the 12% and 26% diet groups were euthanized. Terminal body fat assessments by methanol:chloroform extraction validated the results of the DEXA scans, in that the mice fed the 12% diet had a significantly (P < 0.05) higher percentage of body fat than did those fed the 26% diet (Figure 3). Similar to the results obtained by using DEXA scan, there was no significant difference between male and female mice with respect to postmortem body fat. Correlation analysis comparing the results of the ante- and postmortem body fat analyses indicated much higher correlation between the results of body fat analysis performed by DEXA scan antemortem and chemical analysis of body fat postmortem in the mice fed the 12% diet (r = 0.76) than for the mice fed the 26% diet (r = 0.57). There-

33

cm10000039.indd 33

3/4/2011 1:32:55 PM


Figure 3. Mean percentage body fat (error bar, 1 SD) in California mice fed moderate- (26%; n = 10; male = 4, female = 6) or low- (12%; n = 10; male = 5, female = 5) diet. *, P ≤ 0.05.

Figure 2. Results of DEXA scan for body-fat determination in adult California mice. *, P ≤ 0.05. (A) Mean percentage body fat (error bar, 1 SD) in a cohort of California mice (n = 11; male = 5, female = 6) fed the moderate- (26%) fat diet. Scans were done at 1 and 1.5 y of age. (B) Mean percentage body fat (error bar, 1 SD) in California mice fed the 26% (n = 11; male = 5, female = 6) or 12% (low-fat; n = 12; male = 3, female = 9) diet. Mice were evaluated at approximately 1.5 y of age.

fore, the mice fed the 26% diet appeared to have fat stores that were poorly assessed by DEXA scan and that were not present in the mice fed the 12% diet. Body fat distribution in California mice. The liver was the only site where HR California mice had higher amounts of fat than did LR mice. Fat pad weights for intraabdominal, gonadal, subscapular, and inguinal fat pads were similar between HR and LR male mice on the 26% and 12% diets and for female mice on the 12% diet. The only statistically significant finding was that the subscapular fat pads of HR female mice were heavier (P < 0.05) than those of LR mice (Figure 4 B). In all mice examined, the inguinal fat pad was by far the heaviest (Figure 4 A and B). Liver weight and scoring of fatty liver. HR mice fed the 26% diet had higher (P < 0.05) liver weights than did any other group. This difference was statistically significant (P < 0.05) between female HR and LR mice fed and approached significance (P = 0.06) between male HR and LR mice on the 26% diet. Liver weights did not differ between HR and LR mice fed the 12% diet (Figure 5). When livers were scored for evidence of hepatic steatosis, HR California mice fed the 26% diet had a higher probability of having a fatty liver (Grade 1 or higher; odds ratio = 11.0, P = 0.013)

than did LR mice fed the same diet. HR mice fed the 12% diet did not have an increased risk of having a fatty liver compared with LR mice (P = 0.74). Serum triglyceride concentrations in California mice evaluated for hepatic steatosis at 24 wk followed previously reported trends for dietary and genetic response,24 although diet was more strongly associated (P < 0.05, Mann–Whitney test) with serum triglyceride concentration than were genetics. Histopathology of hepatic lipidosis. In initial experiments comparing the 12% and 26% diets, some California mice had normal or minimal hepatic lipid accumulation (Figure 6 A), whereas others, especially those fed the 26% diet, had very pronounced hepatic fat accumulation (Figure 6 B and C). Both macro- and microvesicular steatosis was evident, but microsteatosis predominated, both in ballooned hepatocytes and lipid-laden hepatocytes with normal cellular architecture (Figure 6 B). Fibrosis, inflammation, Mallory bodies, and apoptosis were uncommon. This finding indicated that the condition in the mice more closely resembled human NAFLD rather than NASH.7,15,22 Fibrosis, when present, was usually lobular and mild to moderate in severity (data not shown). The extent of fatty liver disease in California mice was associated with feeding the 26% diet (Spearman rank-order correlation, P < 0.05). The extent and severity of fatty liver disease was not associated with age, gender, or serum triglyceride response to the 26% diet. Serum triglyceride concentrations were higher (P < 0.05, Student t test) in HR mice fed the 26% diet (mean, 752.6 mg/dL) than the 12% diet (mean, 164.1 mg/dL). Age was a covariant with diet in producing a difference between diet groups (Spearman rank order correlation, P < 0.05). All 8 LR mice survived the 6 wk of being fed the 45% diet, 4 of the 8 HR mice fed that diet died unexpectedly during that time. Necropsy findings included acute, severe hepatic necrosis. Steatosis score was compared among 3 diet groups; because of the few HR mice in the 45% diet group, histologic scores for HR and LR

34

cm10000039.indd 34

3/4/2011 1:32:56 PM


Figure 5. Mean liver weight (error bar, 1 SD) of selectively bred California mice. *, P ≤ 0.05; •, P = 0.06.

Figure 4. Mean weights of fat pads (error bar, 1 SD) in California mice fed the moderate- (26%) fat diet. (A) Male mice (HR = 11, LR = 7). (B) Female mice (HR = 6, LR = 9). *, P ≤ 0.05.

mice were combined within diet groups. The 26% and 45% diets both produced higher (Mann–Whitney test, P < 0.05) steatosis scores than did the 12% diet (Figure 7). Chemical extraction of liver. Among LR California mice, those fed the 45% diet had higher amounts of lipid on a dry-weight basis than did mice fed the 12% diet, but the mice fed the 12% diet had higher amounts of protein and ash than did those fed the 45% diet. Total liver weight reflected the differences in lipid analysis (Student t test, P < 0.05; Figure 8). Circulating leptin concentration. Circulating leptin concentrations did not differ between LR or HR mice fed the 12% or 26% diet (Figure 9).

Discussion

Syndromes that produce lipid overflow, obesity, and lipoatrophy may not be discrete, but rather exist as a continuum. On one end of the obesity spectrum are genetic syndromes that cause very severe disease, such as deficiencies in functional leptin, leptin receptor, and the melanocortin pathway.29 At the other end of the spectrum are pathologic genetic and acquired lipoatrophy syndromes.18,32 Between these extremes are conditions under

which an organism might be evolutionarily selected to manifest aspects of either obesity or lipoatrophy to exploit an adaptive advantage. For example, Pima Indians become obese and diabetic on a typical Western diet.35 Additional human populations have been identified that manifest obesity and type II diabetes under conditions of dietary surfeit but who historically have experienced prolonged periods of famine or disease.28,35 The idea that various populations can adapt to periods of caloric deficiency by storing calories as fat more efficiently is referred to as the ‘thrifty genotype hypothesis,’ an idea that has recently been extensively critiqued and discussed.28,35 Selection for partial lipoatrophy has not been well documented in humans, but recently studies13 in West Bengal have identified people with NAFL, 75% of whom are nonobese. In this population, people whose body mass index was between 25 and 30 and those with lower body mass index but no visceral obesity had a high prevalence of NAFL.13 Hepatic fat storage without obesity is part of the biology of birds. Ducks and geese accumulate fat in their livers, but very little peripheral fat, to provide energy for the long periods of fasting during migration. Humans manipulate this aspect of their biology for foie gras production.17 California mice are not prone to deposit peripheral fat under conditions of caloric excess. They do not have noteworthy visceral obesity, given that the weight of the abdominal fat pad did not differ between HR and LR mice or between the 12% and 26% diet groups (Figure 4), but do deposit fat in the liver. These results are consistent with a partial lipoatrophic syndrome. The partial lipoatrophy in California mice presumably is nonpathogenic when high-calorie food sources are seasonal and limited. However, in the laboratory, there is no compensatory period of caloric deficiency to deplete hepatic fat stores and mice subsequently develop insulin resistance. In the current study, results from DEXA showed higher correlation with body fat content in mice fed the 12% diet than the 26% diet. DEXA can underestimate the fat content of solid organs such as the liver,12 perhaps explaining this discrepancy. The earliest rodent models of NAFLD involved genetically determined changes in leptin signaling, resulting in fatty liver in the setting of obesity.2 Subsequent models have emerged based on dietary manipulations, such as the use of a methionine–choline-

35

cm10000039.indd 35

3/4/2011 1:32:57 PM


Figure 6. Histopathology of hepatic lipidosis in California mice. (A) Normal California mouse liver, a section from an animal fed the low-fat diet. Hematoxylin and eosin stain. (B) Liver from a California mouse fed the moderate-fat diet. Lipid is abundant in hepatocytes, but there is no inflammation or fibrosis. Hematoxylin and eosin stain. (C) Serial section of liver of the California mouse liver in panel B. Oil Red O stain. (D) Liver from a California mouse fed the high-fat diet. Hematoxylin and eosin stain. A size bar for panels A through D is shown in panel A.

deficient diet, and newer rodent models have been created using transgenic technology. All of these models may offer unique insights into pathophysiology but typically differ physiologically from typical human NAFLD.15,25,26,29,33 Virtually all of these models have one or more characteristics of human NAFLD, but the inflammation and fibrosis traditionally associated with NASH has been more difficult to replicate in an animal model without resorting to methods of inducing systemic inflammation, such as endotoxin injection, that are not associated with the pathogenesis of human NASH. However, some investigators do not think that fibrosis is necessary for a diagnosis of NASH and therefore score disease severity on the basis of steatosis, inflammation, and ballooning degeneration.22 Diet-induced steatosis in California mice lacks both definitive features of NASH (inflammation and fibrosis) but has the attributes of NAFLD.1 Even though the California mice had hepatic steatosis, inflammatory change in their livers was minimal (Figure 6 B and C). In addition, hepatic enzymes including AST and ALT can be evaluated to determine the extent of liver damage. Liver damage sufficiently severe to cause increases in liver enzymes is more common and more severe in people with NASH, who have hepatic inflammation, than people with NAFLD.6-10 Rats fed a more extreme high-fat diet (71% energy from fat) than those used in the current study develop

NAFLD but lack elevated AST or ALT concentrations.1 The only pathology in addition to minimal hepatic inflammation in our California mice was present in the HR mice that died when they were fed the 45% diet; at necropsy, these mice showed signs consistent with ketoacidosis. Similar unexpected mortality occurred when these mice were fed the 26% diet during lactation.24 The risk factors that differentiate NASH from NAFLD are not known, but extrahepatic sources of focal infection or inflammation may be involved.3 Leptin is a peptide hormone that was identified through the study of 2 naturally occurring mouse models of obesity and hyperphagia.2 Leptin plays a key role in regulating many homeostatic and developmental functions in mammals.2,5 Leptin concentrations in the blood and central nervous system are involved in the regulation of fat mass and maintenance of body weight, onset of puberty in and subsequent fertility of the organism, and fibrosis and reaction to tissue damage.2,5,20 In addition, leptin acts to prevent fatty infiltration of solid organs.12 Whereas homozygous ob/ob and db/db mice are infertile, California mice are fertile, even when they are hyperlipidemic and insulin-resistant.5,24 However, serum leptin concentrations in California mice are comparable to those of other rodent species, and we observed no significant differences in serum leptin concentration between HR and LR mice or mice on the 26% and 12% diets (Figure 9), despite the fact that

36

cm10000039.indd 36

3/4/2011 1:32:58 PM


Figure 7. Steatosis scores for California mice fed the low-, moderate-, or high-fat (12%, 26%, and 45%, respectively) diets. *, P ≤ 0.05 (Mann– Whitney U test).

Figure 8. Mean lipid, protein and ash, and total weights (error bar, 1 SD) from livers of California mice fed the low- (12%) or high- (45%) fat diet. *, P ≤ 0.05.

mice on the 12% diet had a higher fat mass than did those fed the 26% diet. Obese human beings are considered to be leptinresistant because they have increased concentrations of leptin in their blood but are still hyperphagic.11,32 Conversely, people who are lipoatrophic respond to exogenous administration of leptin.11 If California mice were leptin-resistant, their serum leptin concentrations would be expected to be higher than those of nonobese laboratory mice. Therefore, it is unlikely that California mice are leptin-resistant, and the mechanism of steatosis in this species appears to be independent of leptin metabolism. The differences in fat accumulation seen in California mice could be due, at least in part, to differences in the composition of the 12% and 26% diets. The 12% diet contained beef tallow, a saturated fat, as its sole fat source. The 26% diet contained both beef tallow and soybean oil, which is a polyunsaturated fat. There are 2 potential mechanisms for the effects of the type of dietary fat on body fat accumulation: simple accrual and energy expenditure. Laboratory mice become more obese on diets containing saturated fat, especially animal fat (such as beef tallow), than on diets containing the same amount of polyunsaturated fat of plant origin.34 The composition of the dietary fat affects the amount of thermogenesis in the brown adipose tissue of rats. Rats remain leaner and have higher rates of brown adipose tissue thermogenic activity on diets enriched with polyunsaturated Ω3 fatty acids than on diets high in

Figure 9. Mean serum leptin concentrations (error bar, 1 SD) in California mice fed the low- (12%) or moderate- (26%) fat diet. Differences were nonsignificant.

saturated fat.27 The increased mass of the subscapular fat pad in HR female mice may be due to accrual of brown fat for thermogenesis to warm pups. Although both male and female California mice cooperatively parent their young, including providing warmth for their pups,19 female mice may be better adapted for this function than are male mice. Additional experiments using semipurified diets with a single fat source are necessary to determine whether the type of fat affects body fat distribution and hepatic steatosis in California mice. The currently identified genes that are associated with partial lipoatrophy in human beings do not account for all of the documented cases of the disease. In addition, there may be human populations that have adapted to have a very limited ability to expand peripheral fat mass and who have no clinically significant metabolic syndrome when they have a spare diet but who develop insulin resistance when the diet is more luxuriant, essentially a mirror image of the pathogenesis of obesity in Pima Indians.35 Profiling the gene expression and metabolism in liver and adipose tissue may identify new genes associated with the pathogenesis of fatty liver disease and lipoatrophy and, therefore, reveal novel targets for therapy of these disorders. Putative target genes for profiling expression include those for cytokines, LDL receptor, and fatty acid binding protein and lipogenic genes including sterol regulatory element-binding protein and peroxisome proliferators-activated receptor γ.

Acknowledgments

We acknowledge Mary Tagliarino and the animal care staff of the School of Veterinary Medicine for the care of the mice, Allyson Holler for technical assistance, and Shane Huebner and Dr Mark Cook for lipid analysis. This study was supported by RO3 AG18551 to LKH.

References

1. Ahmed U, Redgrave TG, Oates PS. 2009. Effect of dietary fat to produce nonalcoholic fatty liver in the rat. J Gastroenterol Hepatol 24:1463–1471. 2. Anania FA. 2002. Leptin, liver, and obese mice—fibrosis in the fat lane. Hepatology 36:246–248. 3. Bell LN, Theodorkas JL, Vuppalanchi R, Saxena R, Bemis KG, Wang M, Chalasani N. 2010. Serum proteomics and biomarker

37

cm10000039.indd 37

3/4/2011 1:33:00 PM


4. 5. 6. 7. 8.

9.

10.

11. 12. 13.

14. 15. 16. 17. 18. 19.

discovery across the spectrum of nonalcoholic fatty liver disease. Hepatology 51:111–120. Boozer CN, Schoenbach G, Atkinson RL. 1995. Dietary fat and adiposity: a dose–response relationship in adult male rats fed isocalorically. Am J Physiol 268:E546–E550. Bowles L, Kopelman P. 2001. Leptin: of mice and men? J Clin Pathol 54:1–3. Caldwell SH, Harris DM, Patrie JT, Hespenheide EE. 2002. Is NASH underdiagnosed among African Americans? Am J Gastroenterol 97:1496–1500. Caldwell SH, Oelsner DH, Iezzoni JC, Hespenheide EE, Battle EH, Driscoll CJ. 1999. Cryptogenic cirrhosis: clinical characterization and risk factors for underlying disease. Hepatology 29:664–669. Chitturi S, Abeygunasekera S, Farrell GC, Holmes-Walker J, Hui JM, Fung C, Karim R, Lin R, Samarasinghe D, Liddle C, Weltman M, George J. 2002. NASH and insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome. Hepatology 35:373–379. Chitturi S, Farrell G, Frost L, Kriketos A, Lin R, Fung C, Liddle C, Samarasinghe D, George J. 2002. Serum leptin in NASH correlates with hepatic steatosis but not fibrosis: a manifestation of lipotoxicity? Hepatology 36:403–409. Chitturi S, Weltman M, Farrell GC, McDonald D, Kench J, Liddle C, Samarasinghe D, Lin R, Abeygunasekera S, George J. 2002. HFE mutations, hepatic iron, and fibrosis: ethnic-specific association of NASH with C282Y but not with fibrotic severity. Hepatology 36:142–149. Chong AY, Lupsa BC, Cochran EK, Gordon P. 2010. Efficacy of leptin therapy in the different forms of human lipodystrophy. Diabetologia 53:27–35. Coleman R. 2000. Personal communication. Das K, Das K, Mukherjee PS, Ghosh A, Ghosh S, Mridha AR, Dhibar T, Bhattacharya B, Bhattacharya D, Manna B, Dhali GK, Santra A. 2010. Nonobese population in a developing country has a high prevalence of nonalcoholic fatty liver and significant liver disease. Hepatology 51:1593–1602. Dewey MJ, Dawson WD. 2001. Deer mice: the Drosophila of North American mammalogy. Genesis 29:105–109. Fan J-G, Qiao L. 2009. Commonly used animal models of nonalcoholic steatohepatitis. Hepatobiliary Pancreat Dis Int 8:233–240. Folch J, Lees M, Sloane Stanley GH. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226:497–509. Fournier E, Peresson R, Guy G, Hermier D. 1997. Relationships between storage and secretion of hepatic lipids in 2 breeds of geese with different susceptibility to liver steatosis. Poult Sci 76:599–607. Garg A, Agarwal AK. 2009. Lipodystrophies: disorders of adipose tissue biology. Biochim Biophys Acta 1791:507–513. Gubernick DJ, Alberts JR. 1987. The biparental care system of the California mouse, Peromyscus californicus. J Comp Psychol 101:169–177.

20. Honda H, Ikejima K, Hirose M, Yoshikawa M, Lang T, Enomoto N, Kitamura T, Takei Y, Sato N. 2002. Leptin is required for fibrogenic responses induced by thioacetamide in the murine liver. Hepatology 36:12–21. 21. Hought D. 2003. Personal communication. 22. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, Ferrell LD, Liu YC, Torbenson MS, Unalp-Arida A, Yeh M, McCullough AJ, Sanyal AJ. 2005. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 41:1313–1321. 23. Krugner-Higby L, Shadoan M, Carlson C, Gendron A, Cofta P, Marler C, Wagner J. 2000. Type 2 diabetes mellitus, hyperlipidemia, and extremity lesions in California mice (Peromyscus californicus) fed commercial mouse diets. Comp Med 50:412–418. 24. Krugner-Higby L, Shelness GS, Holler A. 2006. Heritable, dietinduced hyperlipidemia in California mice (Peromyscus californicus) is due to increased hepatic secretion of very-low-density lipoprotein triacylglycerol. Comp Med 56:468–475. 25. Nagy TR, Clair AL. 2000. Precision and accuracy of dual-energy X-ray absorptiometry for determining in vivo body composition of mice. Obes Res 8:392–398. 26. Ogus S, Ke Y, Qiu J, Wang B, Chehab FF. 2003. Hyperleptinemia precipitates diet-induced obesity in transgenic mice overexpressing leptin. Endocrinology 144:2865–2869. 27. Oudart H, Groscolas R, Calgari C, Nibbelink M, Leray C, Le Maho Y, Malan A. 1997. Brown fat thermogenesis in rats fed high-fat diets enriched with n3 polyunsaturated fatty acids. Int J Obes Relat Metab Disord 21:955–962. 28. Prentice AM, Hennig BJ, Fulford AJ. 2008. Evolutionary origins of the obesity epidemic: natural selection of thrifty genes or genetic drift following predation release? Int J Obes (Lond) 32:1607–1610. 29. Ranadive SA, Vaisse C. 2008. Lessons from extreme human obesity: monogenic disorders. Endocrinol Metab Clin North Am 37:733–751. 30. Reitman ML, Gavrilova O. 2000. A-ZIP/F1 mice lacking white fat: a model for understanding lipoatrophic diabetes. Int J Obes Relat Metab Disord 24 Suppl 4:S11–S14. 31. Rothstein JD. 2003. Of mice and men: reconciling preclinical ALS mouse studies and human clinical trials. Ann Neurol 53:423–426. 32. Savage DB. 2009. Mouse models of inherited lipodystrophy. Dis Model Mech 2:554–562. 33. Solga SF, Clark JM, Alkhuraishi AR, Torbenson M, Tabesh A, Schweitzer M, Diehl AM, Magnuson TH. 2005. Race and comorbid factors predict nonalcoholic fatty liver disease histopathology in severely obese patients. Surg Obes Relat Dis 1:6–11. 34. Wang H, Storlien LH, Huang XF. 2002. Effects of dietary fat types on body fatness, leptin, and ARC leptin receptor, NPY, and AgRP mRNA expression. Am J Physiol Endocrinol Metab 282:E1352–E1359. 35. Wells JCK. 2009. Ethnic variability in adiposity and cardiovascular risk: the variable disease selection hypothesis. Int J Epidemiol 38:63–71.

38

cm10000039.indd 38

3/4/2011 1:33:00 PM