VJ...“ a 2.. a a 4...! .5. 5.0V wow "35 . 2.1 FF .0. l ' v) III..- .‘I1 .“u E... 2U "w... .31: an“: A.\ R: 1.. w... . “(by F‘ put ”‘ .1 91. 3.04... '1?le I. A. .3v .4. U "NA...- r A...” ‘1‘<‘1il‘1‘!‘ 11‘ “ ‘1‘ iill‘l‘ “ I‘m 4&2; “'5 ~. .~.-»'mm*' L I D it (fl K I) ‘ ‘ ‘ D 1.25313ng 8mm T . . Luvm-Sy ammé BY "GAG a SUNS' 300K BIIBEIW INL u nun-V -"""‘ Bl IIIII mm m "wt“ Lil! gm Ll H711! gin mu M 1m ll 312 M4967 ABSTRACT EFFECTS OF OVERFEEDING SUCKLING RATS UN FAT ORGAN CELLULARITY By Barbara Estelle Obst Obesity is associated with two processes, an increase in the number (hyperplasia) and size (hypertrophy) of adi- pocytes. Human obesity exists in both forms (ijrntorp and destrbm 1971, Brook §t_§l. 1972, Salans gi_gl. 1973). The methods used to successfully induce hyperplastic obesity in animals include: feeding a high fat diet (Herberg §£_§1, 1974, Lemonnier 1972, Lemonnier and Alexiu 1974) and possibly tube feeding (Marine and Hirsch 1974) and reducing litter size (Knittle and Hirsch 1968). The genetic obesity of the Zucker rat has been classified as hyperplastic (Jflvumn §$_§1, 1971, Lemonnier 1971, Lemonnier and Alexiu 1974) as has been the obese hyperglycemic mouse (Herberg §£_g1. 1970b, Johnson and Hirsch 1972) both relative to their lean controls. Little or no work has been done to determine the fat cell differences between strains of the same Species. However, weight gain, when rats are fed a high fat diet, differs among strains (Lemonnier et al. 1971, Mickelsen et al. 1955 Barbara Estelle Dbst Schemmel §£_gl. 1970a). For these studies two strains of rats, one an obesity- susceptible strain when fed a high fat diet, the Osborne Mendel, and the other an obesity-resistant strain, the S 5B/Pl (Schemmel‘§j_§l, 1970a), were used. At birth, 40 rats of each strain were suckled by dams fed a high fat diet. The other 40 were suckled by dams fed a low fat diet. Half of each of the four groups of 40 were nursed in litters of six while the other half were nursed in litters of three. Finally, half of the rats were supplemented, twice daily via tube feedings, while the other half were not. At weaning (24 days of age) 5 animals from each group of sixteen were sacrificed and the remaining 5 animals were raised until 15 weeks of age at which time they were sacrificed. Fat depots were removed, weighed and rapidly frozen for subse- quent analysis of cell size and number. Adipocyte size and number were determined on a sample of all inguinal tissue and samples of testicular and peri— renal tissues from 40 awfimals representing the extreme dietary treatments (high fat diet, supplemented, three per letter versus low fat diet, non-supplemented, six per litter) for each strain and age. SiZing of fat cells was done micro- scopically On cells isolated by a modification of Rodbell's (1964) procedure. Cells were stained with 2% cresyl violet ,and fixed with a glutaraldehyde solution prior to sizing. The diameter of cells from each depot were measured. From the resulting data cell size (weight of lipid, in ng., of Barbara Estelle Obst the average fat cell) and cell number were calculated. Osborne Mendel rats had about two times the number of fat cells in the entire left inguinal fat pad as did the similarly treated S 5B/P1 strain at 24 days of age. Both strainsfmndincreases of cell number between 24 days and lO5 days of age. For the Osborne Mendel strain this was a two to three fold increase while for the S SB/Pl it was a three to four fold increase. This resulted in all 105 day old animals having similar numbers of inguinal adipocytes, except for the Osborne Mendel strain fed the high fat diet. This group of animals had three times the number of cells as the other groups. Therefore, at 15 weeks of age, the obesity-suscep- tible Osborne Mendel strain of rat fed the high fat diet had hyperplasia. The obesity—susceptible male Osborne Mendel strain always had larger cells than the S SB/Pl strain. This difference was significant in half of the groups when similar ages were compared. The Osborne Mendel strain had larger cells than the S 5B/Pl in inguinal tissue when first evaluated at 24 days of age. When this was coupled with their greater increase in size with age (three fold versus two fold) they completed the 15 weeks of the study with cells 2% times larger than comparably raised male S SB/Pl animals. This was on an absolute scale. Similar effects of strain, age and diet on fat cell size and number as those already described in the inguinal depot were also true for testicular and perirenal fat depots. Barbara Estelle Obst In testicular and perirenal tissue the Osborne Mendel rat had fat cells twice as large as the S SB/Pl rat at 105 days of age. The largest fat cells were located in inguinal and testicular tissues. The number of testicular cells in the whole left depot was two times greater in the high fat fed 105 day old Osborne Mendel rat than in all other 105 day old groups. For peri— renal tissue this was almost a four fold increase. High values for adipocyte number were found in 15 week old Osborne Mendel rats fed a high fat diet in each tissue studied. Feeding of a high fat diet versus a low fat diet had not significantly affected the size of the adipocytes by 15 weeks of age. Neither altering litter size from six to three pups nor supplementing neonatal pups twice daily sig- nificantlly affected the eventual depot cell size or cell number in inguinal fat depots. EFFECTS OF OVERFEEDING SUCKLING RATS ON FAT ORGAN CELLULARITY By Barbara Estelle Obst A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1976 To my daughter Lindora and my husband Norman ii ACKNOWLEDGEMENTS My appreciation is extended to the following persons who contributed to this study: My academic advisor, Dr. Rachel Schemmel, and Dr. Dorice Narins, Dr. Robert Merkel and Dr. Duane Ullrey, my committee; for their time, assistance and guildance. Dr. Vance Sanger and assistants for help in finding a method to microscopically distinguish lipid droplets from adipocytes. The department of pathology for use of the Leitz Ortholux microsc0pe used in making photomicrographs. Dr. Dale Romsos for suggestions on the use of colla— genase and for clarification of calculations. Dr. Roger McFeeters for writing the program for the Wang calculator to expedite calculations. Dr. Kenneth Stevenson for use of the stage lamp and eyepiece and stage micrometers for microscopic work. Marianne Stone, Janet Grommet and Anna McKay for as— sistance with dissection of the animals. Cathy Nolan for help with calculations. Mary Andrews for statistical help and computer pro- graming. My neglected family, Norman and Lindora. iii The National Science Foundation which partially supports the Michigan State University computing facility. This project was supported in part by the Weight Watchers Foundation, Inc. iv TABLE OF DEDICATION . . . . . . ACKNOWLEDGEMENTS . . . LIST OF TABLES . . . . . LIST OF FIGURES. . . . . REVIEW OF LITERATURE . Obesity--Implication for Human Health. Risk Factor. . . . Defining Obesity and Overweight . . . The Cellular Classification of Obesity. CONTENTS Incidence of Human Obesity. Adult Obesity. . . . . Cellular Classification. Prognosis for Weight Reduction Childhood Obesity . . . Animal Studies . . . . Growth and DevelOpment . Non-Adipose Tissue. . Adipose Tissue . . . Over and Under Nutrition Species . . . . . Depot Location. . Sex . . . . . Diet . . . . . Strain . . . . Genetic Obesity . Hypothalamic Obesity INTRODUCTION. . . . . . METHODS . . . . . . . Experimental Design. . . EXperimental Conditions Procedure for Sacrificing Rats Fat Depots. . . . . Depot Lipids . . . . . Weanling Animals . . . One Hundred and Five Day Old and Removing Page ii iii vii viii 1...: \oxocncnslbbwwmmmh—nI—J 21 25 25 27 28 28 28 BO Assessment of Adipocyte Number and Size . Isolation . . . . . . . . . . Fixing and Staining . . . . . . . Identification. . . . . . . . . Tabulation . . . . . . . . . . Calculations . . . . . . . . . Photography . . . . . . . . . . Statistical Analysis of Data. . . . . RESULTS AND DISCUSSION . . . . . . . Inguinal Cell Number Age . . . . . . . . . . . . Strain . . . . . . . . . . . Ration . . . . . . . . . . . Supplementation . . . . . . . . Litter Size. . . . . . . . Summary, Inguinal Fat Cell Number Adipocyte Size in the Inguinal Fat Depot. Age . . . . . . . . . . . . Strain . . . . . . . . . . . Litter Size, Supplementation and Ration Summary, Rat Cell Size in the Inguinal Fat Relation of Results to Body Weight and Body Fat Cause of Osborne Mendel Obesity. . . . Comparison of Number and Size of Adipocytes in Three Fat Depots . . . . Adipocyte Number . . . . . Inguinal Tissue. . . . . Testicular Tissue . . . . Perirenal Tissue . . . . Comparison of the Number of Cells i Three Depots. . . . . . . Fat Cell Size . . . . . . . Inguinal Tissue. . . . . . Comparison with Other Data . Testicular Tissue . . . . . Comparison with Other Data . Perirenal Tissue . . . . . j o o o o o o 0 Comparison with Other Data Strain and Adipocyte Density . Implications for Human Obesity . CONCLUSIONS . . . . . . . . . . . SUGGESTIONS FOR FURTHER STUDY . . . . . Control of Obesity . . . . . . . . Microscopic Versus Coulter Counter Sizing Proliferation Pattern . . . . . Sex Difference .. .. .. . . . . . . LITERATURE CITED . . . . . . . . . vi Depot Page 32 32 32 33 34 35 36 36 38 38 38 39 4O 41 41 42 42 43 44 45 46 46 46 47 48 48 48 49 49 SO SO 51 51 52 54 55 56 57 82 84 84 85 86 86 87 LIST OF TABLES Diet Composition . . . . . . . . . Mean fat cell number and standard error in the entire left inguinal fat depot for each group of animals. . . . . . . . . . List of significant differences (0840.05) in left inguinal cell number between ages, strain, rations, supplementation groups and litter sizes when the other four variables are held constant . . . . . . Mean cell size and standard error, expressed in ng. in left inguinal depots for each group of five animals . . . . . . . . List of significant differences (atA0.0S) in left inguinal cell size, expressed as weight in ng. between ages, strains, rations, supplementation groups and litter sizes when the other four variables are held constant. . . . . . . . . . . The effect of age, diet and strain on mean adipocyte number in three selected depots of male rats . . . . . . . . . . . Comparison of cell number X 106 in three left depots by three variables . . . . . The effect of age, diet and strain on mean fat cell size, expressed as weight in ng.,in three fat depots of male rats . . . Comparison of cell size, expressed as weight in ng., in three left fat depots by three variables . . . . . . . . . . . . vii Page 58 59 6O 61 62 63 64 65 66 LIST OF FIGURES Figure Page 1. Experimental Design . . . . . . . . 67 2. Krebs—Ringer Phosphate Buffer . . . . . 68 3. Glutaraldehyde Solution . . . . . . . 69 4. Photomicrographs . . . . . . . . . 7O 5. Assessment of Adipocyte Diameter . . . . 71 6. Tallying of Sized Adipose Cells . . . . 72 7. Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for twenty-four day old 8 SB/Pl rats . . . . . . . . 73 8. Mean cell number X 106 versus mean cell size expressed in ng. lipid, in left inguinal depots for one hundred and five day old S SB/Pl rats. . . . . . 74 9. Mean cell number X 1O6 versus mean cell size, expressed in ng. lipid, in left inguinal depots for twenty-four day old Osborne Mendel rats . . . . . . 75 10. Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for one hundred and five day old Osborne Mendel rats . . . 76 11. Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots . . . . . . . . . 77 12. Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left testicular depots . . . . . . . . 78 13. Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left perirenal depots. . . . . . . . . 79 viii Figure Page 14. Mean inguinal cell number X 1O6 versus age in days in left depots. . . . . . . 79 15. Mean inguinal cell size, expressed in ng. lipid per cell, versus age in left depots . . . . . . . . . . . . 80 ix REVIEW OF LITERATURE Obesitv--Implications for Human Health Risk Factor Data have been obtained to show the positive relationship between obesity and many diseases and physical impairments. Micklejohn (1955) divided the complications of obesity into two categories--metabolic and mechanical. The metabolic complications included diabetis, hypertension and disorders of cholesterol metabolism. The mechanical problems were further classified by Sinclair (1955) to include postural changes, clumsiness and the direct effects from carrying extra weight. Postural changes led to respiratory and cir- culatory insufficiency; clumsiness to more accidents; and, since extra weight must be supported physically by the joints and metabolically by the heart and lungs, increased incidence of arthritis, varicose veins, hernias and pneumonia. In addition, increased incidence of high blood pressure, kidney disorders, gallstones (Eden 1975) influenza and digestive tract disorders (Wilkinson 1975) have been reported in the obese person. As a result, obesity has been associated with reduced life expectancy (Eden 1975, Micklejohn 1955, Sinclair 1955, Wilkinson 1975). 2 Defining Obesity and Overweight While obesity connotes an excess storage and depotsiton of fat, Knittle (l972b)does not believe overweight directly implies fatness. Physiologically, according to Salans §t_gl. (1973), obesity is characterized by an excess of adipose tissue mass. A person has been considered to be overweight between ten and twenty percent heavier for height than expected (Eid 1970, Shukla §:_gl. 1972) from Scott's (1961) London County Council or Tanner gt_gl. (1966) standards, respectively. Obese was twenty percent or more above weight for height (Brook §t_gl. 1972, Brook and Lloyd 1973. Eid 1970, Shukla §£_§l. 1972)both for given sex and age of subject. Asher (1966) has considered an individual obese when the weight for height was above the ninety-seventh percentile. Obesity was defined by Johnson §t_gl. (1956) to be in or above the channel A4 of Wetzel's grid (1941). The obese person, Knittle (1972b) has stipulated, has at least 40% of total body weight as fat. The Cellular Classification of Obesity There are three possible routes by which fat organ enlargement could occur; hyperplasia, hypertrophy, or a combination of the two. Hyperplastic obesity is characterized by an increase in the number of adipocytes. Hypertrophy is an enlargement of fat cells. The number of adipocytes is not increased in hypertrOphic obesity relative to a 3 non-obese subject of similar age, sex and species. Finally, obesity could result from the combined effects of both hy- pertrOphy and hyperplasia. Incidence of Human Obesity Thomas and Mayer (1973) have reported 50% of the middle aged Americans as being over-weight based on height weight tables. The Ten State Nutrition Survey (1968-1970) found 40% of adults and 39% of adolescents to be obese by weight. Johnson et_§l. (1956) reported ten percent of Boston area school children as obese in 1952. Shukla §t_§1. (1972) cited 16.7% of three hundred infants obese and an additional 27.7% as overweight. The percentages of obese persons as obtained by Eid (1970), for children aged six through eight years having shown excess weight gains at six months, closly agree with the percentages of Shukla et al. (1972). _gu1t Obesity Cellular Classification Salans §j_§l, (1973), Brook §£_gl. (1972) and ijrntorp and dester (1971) have associated the character of human obesity with the age at its onset. Hyperplastic obesity has been related to early onset, generally through the age of five years and between nine and thirteen years (Salans §£_§1. 1973). Robertson and Lowrey (1964) reported that at least 26% of the subjects (aged four through sixteen years) 4 attending a clinic for overweight children had become over— weight between the age of two and five years and 45% were overweight before they were six. Thus fat babies tended to become fat children (Eid 1970) and a high proportion of fat children became fat adults (Brook §1_§1, 1972, Knittle 1972b). Hypertrophic obesity developed, according to Salans §t_al. (1973), later in life. This is consistent with the finding of ijrntorp and Sjbstrom (1971), who found that the type of obesity (hyperplastic or hypertrophic) was asso- ciated with the degree of adult obesity. The moderately obese tended to be of the hypertrophic type whereas the most obese exhibited hyperplastic obesity. Prognosis for Weight Reduction Hirsch et_§l. (1966) was the first investigator to report that cell size, and not cell number, varied with adult weight gain and loss. This has been subsequently verified by many other investigators (ijrntorp and Sjbstrdm 1971, Bray 1970, Brook and Lloyd, Knittle 1972b, Salans §t_gl. 1971). While weight reduction for hypertrophic obesity is not easy, it is apparently even more difficult for the individual with the hyperplastic form (Knittle 1972b). CbilfibeeLubesiliL Since prognosis for hyperplastic, child-onset, obesity in the adult is poor, the cure lies in prevention. Cell number has been thought to become fixed in the late teens 5 or as young adults (Hirsch §t_§l, 1966). Prevention, there- fore, must focus through this age. Most of the obese students in the study of Johnson §£_§l. (1956) had been persistently obese throughout their school years. Some and become obese during grades seven through nine, and a few had fluctuated between the obese and non-obese states. In the non-obese subject, the ealiest age at which adult cell number was achieved was twelve (Knittle 1972b). In the obese children studied, adult cell number was achieved in many by eleven years of age and in some as early as six years. In addition, the rapid increase in cell number, which normally began from nine to twelve years of age, was observed to have begun from five to seven years in some obese children. Once obese successful weight reduction was rare (Asher 1966, Eid 1970, Robertson and Lowrey 1964). Rather than actually producing a weight loss the recommended treatment for obese children has consisted of maintaining stable weight while linear growth continues (Knittle 1972b, Wolff 1955). Little, however, is known about the cellular effects of maintaining obese children on a diet (Brook et_gl. 1972). Obesity and birthweight in humans seem to be unrelated. Wolff (1955) failed to find this relationship while Asher (1966) failed to show a significant relationship. Overweight in later years was more related to rapid infant weight gain than to the fact the child had one or two obese parents according to Eid (1970). In contrast, several other researchers (Tanner 1955, Tepperman 1958, Thomas and Mayer 1973) found that heredity was more important than 6 environment in determining the eventual weight of an in- dividual. That rapid increase in infant weights, reflected by recent trends toward shortened birthweight doubling is advantageous is now being questioned (Narins and Hirsch 1974, Wilkinson 1975). As early as 1960 Widdowson and McCance questioned the benefits from making children grow faster than they had previously. Early nutrition has been implicated in the determination of later development of obesity. Some reports have sug— gested that early weaning and introduction of solid foods was causal (Shukla §t_§l. 1972, Wilkinson 1975). Others (Beal 1957, Fomon 1969, Guthrie 1966) have disputed this due to poor infant acceptance and low calorie density of the solid foods relative to milk. Fomon (1974) speculated that early introduction of solid foods might encourage overeating and establish unsound eating habits. Shukla §t_§l. (1972) found that milk consumption of the infant did not initially decrease when solid foods were added to infant diets. The average daily intake for all infants up to three months of age was 775 Kcal. while for infants in the same age range and receiving only milk was 590 Kcal--a substantial dif- ference. Guthrie (1966) reported that infants receiving solid food had about the same caloric intake as these re- ceiving none. Many babies have been receiving solid food while very young (Beal 1957, Guthrie 1966, Shukla gt_gl. 1972). Forty percent of the three hundred infants studied 7 by Shukla et a1. (1972) had received solids by four weeks. This resulted in an increased intake of protein, carbohydrate, fat and calories compared to the intake of unweaned babies. According to the study of Beal (1957), introduction of fruit and cereal occurred at an average of one month of age. Guthrie (1966) found eight percent of the fifty infants participating in her study to have been receiving solid foods at five weeks and all by nine weeks. The relationship of bottle feeding and breast feeding to obesity is also controversial. Eid (1970) reported no difference between the two methods of feeding and the in- cidence of obesity. Bottle fed babies gained weight more rapidly than did breast fed babies, according to Fomon (1974), and he speculated the reason may have been overfeeding at the bottle. Shukla §£_gl. (1972) found a small percentage of mothers began breast feeding and that there was a high drop out rate: only 6.3% of the population continuing at twelve weeks. There was a tendency, in this study, for bottle fed infants to receive solid foods earlier than those fed at breast. Animal Studies Many factors effect the size of the fat cell and cel- lularity of the fat organ. Among them are: species, ana- tomical site, location within the adipose organ, sex, diet and strain. In addition, single gene mutation resulting in obesity and obesity produced by lesions of the hypothalamus 8 provide information to further our understanding of the whole concept of obesity. Growth and Development Non-Adipose Tissue Early work on cellular growth included information on organs of the Sprague-Dawley rat, but due to technical difficulties excluded adipose tissue (Winick and Nobel 1966, Winick and Nobel 1967). Growth was altered by over and under nourishment achieved by altering litter sizes. Parkes (1926) observed differences in body weight gains of mice reared in different sized litters; those from small litters gained weight more rapidly than those from large litters. Reducing rat litter size to three pups resulted in a persistent increase in body size throughout the whole life of the animal compared to rats from large litters of eighteen (McCance and Widdowson 1966). Winick and Nobel (1967) reduced litter sizes of Sprague-Dawley rats to three and six and found many parameters increased, including: body weight, organ protein, RNA, DNA and the number of cells per organ. The size of the cells, however, were similar in both over- nourished and control rats. They also (1966) deprived rat pups by increasing litter size to eighteen and restricting neonatal intake at various ages. They concluded that the cellular response to deprivation depended on the tissue in question, the age at onset, and the duration of deprivation. Adipose Tissue Work on the develOpmental pattern of adipose tissue, too, has utilized the Sprague-Dawley rat. Peckham §t_§l, (1962) used DNA to assess adipose tissue expansion and con- cluded that the epididymal site retained the capacity to generate adipocytes through thirty—four weeks. A similar conclusion was reached by Zingg §t_§1, (1962) using Wistar rats. Enesco and Leblond (1962) found that epididymal tissue expansion occurred, in the Sherman rat, by both hyperplasia and hypertrophy as evidenced by DNA analysis. Hirsch and Han (1968), using the technique of Hirsch and Gallian (1968) found that growth by hyperplasia ceased in the epididymal site at twelve weeks and in the retrOpe— ritoneal site at fifteen weeks. Beyond these ages depot enlargement in the chow fed Sprague-Dawley rat was by by- pertrophy alone. This has been confirmed by Greenwood and Hirsch (1975) using a pulse-dose sequence of tritiated thymidine. These reports need not be in conflict since Rodbell (1964) showed that a large proportion of the DNA in adipose tissue is present, not in adipocytes, but in their SUpporting matrix. Thus, while the adipocytes them- selves may not have been expanding in number past twelve weeks, the supporting matrix may have continued to expand. Over and Under Nutrition Many investigators have altered the size of rat (Heggness et al. 1961, Johnson et a1. 1973, Knittle and 10 Hirsch 1968) and mouse (Lemonnier and Alexiu 1974, Lemonnier §£_§l, 1973) litters in an effort to achieve either under or over nutrition with respect to adipose tissue growth. Tube feeding (Dymza gt_g1. 1964, Narins and Hirsch 1974), alteration of diets of lactating dams (Emery §t_§l, 1971, Schemmel §t_§1. 1973, Knittle 1972a) and starvation (Hirsch and Han 1969) have also been used. Heggness §t_§1. (1961) rotated two litters of six Rochester rat pups each between three dams on an eight hour schedule such that one group always went to a dam that had not been suckled the preceeding eight hours. The second group, then, always went to the dam suckled the previous eight hours. In addition, regular litters of six, twelve and eighteen pups were raised. The animals from twelve and eighteen per litter gained weight less rapidly while having equally as efficient utilization. They were presumed to have been eating less. Group one was concluded to have gained weight at a maximum rate because group two compared well to the animals raised regularly six to the litter. Therefore, there must have been enough milk left after group one had suckled to adequately nourish group two. All body consti- tuents were elevated by preweaning overnutrition but this was especially true of fat. Group one exhibited a growth spurt when weaned to the calorically more dense high fat diet. Knittle and Hirsch (1968) altered litter sizes to either four or twenty-two pups per litter at birth. Animals from the small litters were heavier and had more adipocytes in the epididymal fat organ than those raised in large ll litters. Beginning at ten weeks of age the animals from the small litters were significantly heavier and had signi— ficantly larger cells. Their epididymal fat depots were both absolutely and relatively heavier than those of their siblings. In addition, the pattern of adipose tissue growth had been altered in animals from small litters because hy- perplasia continued through the twenty weeks of the study. Female mice raised in litters of four on a control diet had more and larger adipocytes at 4.5 months of age and an increase in body fat when compared to animals raised nine to the litter (Lemonnier and Alexiu 1974). Pups tube fed by Dymza §t_gl. (1964) on basal plus five percent rats milk had significantly more carcass fat and less moisture than controls. Narins and Hirsch (1974) supplemented neonatal rats and increased adipocyte size over controls. Although there was no significant difference in cell number between the supplemented and non-supplemented groups, both had elevated cell numbers compared to the animals of Knittle and Hirsch (1968) raised simultaneously. A form of supplementation, giving lactating dams a high fat diet, was found by Emery gt_§1. (1971) to increase milk fat two fold. Schemmel §£_§l. (1973) reported that lactating Osborne Mendel rats showed this difference at eighteen days post-partum when fed grain or 60% fat (w/w) diets. Another method of altering rat milk supply was used by Knittle (1972a). He deprived pups by altering maternal protein and calorie intake. He found that pups from calorie 12 restricted dams could catch up to controls in growth while those from protein and calorie restricted dams could not. Hirsch and Han (1969) reported that starving six week old Sprague-Dawley rats for one week resulted in a reduction of both cell size and number. Both parameters returned to control levels with refeeding. Prolonged semi-starvation from fifteen through twenty-six weeks of age resulted in a reduction in cell size only. Hollenberg §£_§1. (1970) have confirmed this in the lumbar and epididymal sites using animals previously in- jected with a dose of tritiated thymidine. More recently, Greenwood and Hirsch (1974) have used tritiated thymidine to determine the developmental pattern of Sprague-Dawley rat epididymal tissue. Mature adipocytes have gone through three characteristic developmental stages, those being: proliferation, differentiation and maturation. During proliferation the cells are thought to be produced at varying rates and times and be influenced by environmental conditions such as: hormones, temperature, age, and dietary intake. Cells leave the proliferative stage and enter the differen- tiating stage where they stay again for varying periods depending upon external stimuli. Finally, cells enter the mature population and enlarge or are lost from it. Several feedback 100ps are thought to influence the whole process. In epididymal adipose tissue of the nine day old Sprague-Dawley rat the entire process was uninhibited, and proliferation occurred rapidly (Greenwood and Hirsch 1974). 13 Between two and four weeks of age proliferation was rapid but differentiation was slow. From five to seven weeks proliferation slowed, but time spent differentiating again decreased. Beyond this age, the adipocytes were filling with the lipids. Thus the pattern was hyperplasia and hypertrophy early, maturing to hypertrophy alone. Species Expansion of adipose tissue mass varies from species to species. In male Wistar rats and male Golden hamsters postnatal depot enlargement was primarily by hypertrophy while in male albino guinea pigs hyperplasia was causal (Di Girolamo et al. 1969, Di Girolamo and Mendlinger 1971). Depot Location The location of the fat organ within the animal affects both the number and size of cells that are established. In adult male rats (Di Girolamo §t_gl. 1969, Di Girolamo §t_§1. 1974, Hellman §t_gl, 1962, Johnson et_gl. 1971) and mice (Lemonnier 1972) the perirenal and retroperitoneal depots have the largest cells, the epididymal depot the next largest and the subcutaneous depot the smallest. The order was dif- ferent in genetically obese mice, however, being subcutaneous 7epididymal7perirenal (Johnson and Hirsch 1972). A species difference existed also. In New Zealand rabbits subcutaneous adipocytes were the largest while those from the epididymal site were smallest (Di Girolamo et al. 1974). 14 Most fat cells are generally found in the male subcu- taneous depot, the next largest number in the epididymal site with fewest in the perirenal or retrOperitoneal for rats 26 weeks of age (Johnson E£_Elv 1971) and mice 32 weeks of age (Lemonnier 1972) both raised on control diets. In addition, different depots are affected differently by dietary alterations. Lemonnier and Alexiu (1974) and Lemonnier (1972) reported hyperplasia in the perirenal and perigenital sites while hypertrophy was evident in the ab- dominal subcutaneous location at thirty-two weeks of age as a result of feeding a high fat diet to mice since birth. When the high fat diet was given to male Swiss mice hyper- plasia appeared in the perirenal depot too. Work on adult human subjects has also revealed variations in cell size. Salans §t_§l. (1971) found site to site variations which were defined further as glutealZpreperi- tonealZabdomial7mesentarthricepts7omentum (Salans 23121- 1973). In general they found most cells in the omental site and fewest in the gluteal location. Brook (1971) found cells from deep human sites in child- ren aged 0.8—ll.4 years to be smaller than those from sub- cutaneous sites. It has also been shown that cells along lobule borders are smaller than those centrally located (Smith 1971). Goldrick (1967) has observed variation of cell size from the periphery to the center of the lobule, which he described as a sectioning artifact. 15 Sex Lemonnier (1971) studying ten week old genetically obese Zucker rats fed standard diet, observed that females had a 2.3 fold increase in cell number while males had sig- nificantly fewer cells than their loan littermates in the perigenital location at 2.5 months of age. Depot weight increased 9.7 times for females and 3.2 times for males. Cell size increased about four times in both sexes. These differences in sex and obesity disappeared by twelve months of age but were replaced by hyperplasia in the perirenal and sub-cutaneous depots of obese animals. Further evidence of sex difference was reported by Lemonnier and Alexiu (1974). Female control rats exhibited a two fold increase in cell number between the ages of 2.5 and 12 months while males did not. Sex differences were also apparent in adult mice with respect to cell size and number in perigenital, peri- renal and abdominal subcutaneous locations (Lemonnier 1972). Johnson and Hirsch (1972) have reported the most st- riking hyperplasia in six strains of genetically obese mice as occurring in the female obese hyperglycemic mouse. Similar results to those found in female versus male rats (Lemonnier and Alexiu 1974) and mice (Lemonnier 1972, Johnson and Hirsch 1972) were also found in women by ijrntorp and destrbm (1971). Women tended to have more fat and more adipocytes than men. 16 Diet Diets high in various types of lipids have been related to increased rat weight gain. Barboriak et_al. (1958) reported solid fats such as Crisco were more effective in promoting growth than oils. Osborne Mendel rats, fed a high fat (60% w/w) diet from weaning, had fat depots two to three times heavier than grain fed controls (Schemmel gt_§1. 1970a). In rats such as the Osborne Mendel strain, which readily became obese when offered a ration high in fat; the diet after weaning had an equally important effect on the eventual weight of the animal as did the diet during the preweaning period (Schemmel §t_§l. 1973, Schemmel and Mickelsen 1974). Peckham §t_§l, (1962) also fed a ration high in fat. Using DNA as the indicator of cell proliferation, they determined that weight gain in the epididymal site of Sprague-Dawley rats was accompanied by increase in cell number. Switching animals from the high fat (57% w/w) to a calorically less dense pellet diet at 34 weeks of age resulted in a lower DNA value at 65 weeks than animals fed a high fat diet since weaning. When NMRI mice were fed a high fat diet, cell size and cell number increased in epididymal and scapular sub- cutaneous locations (Herberg §t_§l. 1974). Hypertrophy was the result in genetically obese obob mice (Lemonnier §t_§l. 1971). Thirty-two week old Swiss mice, fed a high fat diet since birth, showed hypertrophy in abdominal subcutaneous l7 and perigenital depots with hyperplasia in the perirenal site. Lemonnier and Alexiu (1974) reported hypertrophy in the epididymal site when either corn oil or hydrogenated palm oil was given as the energy source to weanling rats. Corn oil was associated with hypertrophy and hydrogenated palm oil with hyperplasia in the perirenal site. Genetically obese adult male Zucker rats fed a high fat diet from five through twelve months showed hyperplasia in perirenal and subcutaneous sites. In spite of the fact that the cell number of the lean animals had stabilized, when a high fat diet was fed to adult Zucker rats hyperplasia was observed in the perirenal site. Strain Different strains of the same species show marked dif- ferences in ability to gain weight on the same diet (Hammond 1955, Schemmel 1970a). Lemonnier §t_gl. (1971) noted that Swiss albino mice were more prone to become obese when a high fat diet was consumed than were Bar—Harbor mice. Osborne Mendel rats gained the most weight and S 5B/P1 rats least when seven strains were fed a high fat diet (Schemmel g£_§l. 1972, Thiel §t_§l. 1972). Osborne Mendel rats were more efficient than the SSB/Pl strain in utilizing food calories for deposition within the body regardless of whether eating a high fat or grain ration. This was especially true of the high fat diet (Schemmel et al. 1972). Age and strain 18 did affect weight, but had no effect on the percentage of triglyceride or on fatty acid composition within the depot (Shier and Schemmel 1975). Genetic Obesity Studies have been conducted on genetically obese animals in order to define the character of obesities caused by a single gene mutation. Johnson and Hirsch (1972) found that the obesity of most varieties of mice exhibiting this charac- teristic was of hypertrophic origin. Exceptions included the obese-hyperglycemic mouse (Herberg §t_gl, 1970b, Johnson and Hirsch 1972) and the New Zealand Obese mouse which exhibited hyperplasia as well. The latter strain, however, is not caused by a single gene mutation and thus has no lean control. When compared to NCS/R it exhibited hyperplasia (Johnson and Hirsch 1972) but Herberg ej;§l, (1970b) classified it as hypertrophic. The only rat which carries the trait for genetic obesity the Zucker strain, has been investigated. Bray (1969) noted that hypertrophy caused enlargement of the epididymal and parametrial depots of the genetically obese animal. Johnson §£_gl. (1971) concluded from their developmental study that post-weaning growth of the lean Zucker rat was similar to the Sprague-Dawley. Tissue expansion was the result of com— bined hypertrophy and hyperplasia through fourteen weeks beyond which enlargement took place by hypertrophy alone. The exception was the scapular subcutaneous site which showed 19 continued hyperplasia through 26 weeks. The developmental pattern in the obob varied from his lean littermate. The fatty tended to be born with fewer cells but then caught up and, in fact, surpassed the control. Rate of proliferation exceeded that of the control, continuing through 26 weeks. Cell size was equivalent in all locations unlike the normal rat where subcutaneous7epididyma17retreperitoneal. Hyper— plasia was observed in all adipose depots of the obese Zucker and, in addition to his reduced skeletal size, the animal had a greater percentage of total body fat. Johnson gt_gl. (1973) reported the effect of raising Zucker rats in litters of three or four and twelve to nine- teen pups per litter in order to determine if genotype or nutritional effects predominated. During the first thirty days the nutritional effect was the predominant influence on weight gain while beyond twelve weeks genotypic difference became more important. Lemonnier (1971) reported hyperplastic obesities in female genetically obese Zucker rats. Hypothalamic Obesity Lesions in the ventromedial nucleus of the hypothalamus (VMH) are a method of inducing obesity in an animal. Johnson 23:21, (1971) studied VMH lesioned lean Zucker rats. They found that the cellular obesity was similar to that of the obese Zucker in the epididymal site, cell sizes being much larger than the controls. The retroperitoneal site of the VMH animal exhibited hypertrOphic obesity with respect 20 to the control. In the scapular subcutaneous site the VMH lesioned Zucker had fewer cells than the obese Zucker and larger cells than the controls. Reduced skeletal size was observed in these rats. Han EI_E£° (1965) lesioned Sprague—Dawley rats and noted that thirty percent became obese when pair fed with controls. Lesioning caused an increased percentage of body fat but nose to occipital length, femur length and stomach weight decreased. The VMH animals gained equivalent amounts of weight while growing less rapidly. Han and Lui (1966) produced lesions in weanling Sprague-Dawley rats and then force fed them to control both quantity and pattern of intake. They concluded that the induced obesity was not caused by hyperphagia. Thomas and Mayer (1973) reported VMH obesity was caused by hyperphagia. A two to three fold increase in initial intake was reported. Kennedy (1955) reported VMH rats as having an increased drive to work for food. It seems, then, that they also have altered metabolic pathways allowing them to become obese when pair fed. Hirsch and Han (1969) lesioned Sprague-Dawley rats at 7.5 and 13 weeks. They concluded that the induced obesity was hypertrophic. Han and Lui (1966) lesioned rats at 21 days. This resulted in increased body weight and reduced bone growth. They speculated that this might be due to close proximity between eating centers and growth hormone releasing factor centers in the hypothalamus. INTRODUCTION Human obesity exists in both the hypertrophic and hy- perplastic forms (ijrntorp and Sjbstrbm 1971, Brook §£_§l. 1972, Salans §t_gl. 1973). The former has been associated with adult onset obesity. The hyperplastic form, which according to Knittle (1972b) is more difficult to correct because of the increased number of adipocytes, has been associated with onset in childhood (ijrntorp and destrbm 1971, Brook gt_gl. 1972, Salans gt_§l. 1973). In contrast to working with human subjects, laboratory animals, such as rats and mice, permit rigid dietary and environmental control. In addition, genetic differences, which are important in influencing the build of an individual (Tanner 1955, Tepperman 1958, Thomas and Mayer 1973), can be controlled when animals are used. Most obese laboratory rats and mice have been reported to have hypertrophic obesity (Lemonnier 1971). Few animal models exist for the hyperplastic form. Genetic obesities (Johnson and Hirsch 1972, Johnson §t_§l. 1971, Lemonnier 1971, Herberg 1970b), feeding a high fat diet (Lemonnier 1972, Lemonnier and Alexiu 1974, Herberg §t_§l, 1974) and possibly tube feeding (Narins and Hirsch 1974) or reducing litter size (Knittle and Hirsch 1968) have been associated 21 22 in some situations with hyperplasia. When observed it is most often in perirenal tissue (Lemonnier and Alexiu 1974). Females seem to have a greater susceptibility to hyperplasia than do males (ijrntorp and Sjbstrdm 1971, Johnson and Hirsch 1972, Lemonnier 1971). Little work has been done on differences in strain, although differences in ability to gain weight on a given nutritional plane have been docu- mented (Hammond 1955, Lemonnier §t_§l. 1971, Schemmel §t_gl. 1970a). This study was undertaken to determine the cellular effect of a combination of obesity inducing agents in two rat strains. Male Osborne Mendel rats served as obesity susceptible and male S 5B/Pl rats as obesity-resistant models (Schemmel et_gl. 1970a). Induction of obesity was attempted through the following manipulations: reducing litter size from the control level of six (normal for the S 5B/Pl strain) to three; feeding a high fat (44% w/w) diet to post-partum dams, their nursing litters, and their weanlings; and sup- plementing the intake of neonatal pups with tube feedings. The primary purpose of this study was to evaluate how obesity-resistant and susceptible rat strains would respond from the viewpoint of adipose tissue cellularity, to dif- ferent methods of overfeeding. Procedures known to have been associated with the production of obesity between birth and weaning as well as after weaning were used. Which dietary manipulations were associated with the pro- duction of hyperplastic obesity and which with hypertrophic 23 obesity were also of interest. In addition, the relation- ship, if any, of the number or size of adipocytes to body weightem sacrifice or percent carcass lipid was of interest. The correlation between body weight and body fat was also assessed. Two primary methods exist for the assessment of adi- pocyte cell size and number. The technique of Hirsch and Gallian (1968) used osmium tetroxide to fix adipocytes which were then sized in a Coulter counter. This method has been widely used in recent years (Brook 1971, Brook and Lloyd 1973, Brook gt_§l, 1973, Hirsch §£_§l. 1966, Hirsch and Han 1969, Johnson §t_§l. 1971, Johnson §t_31. 1973, Johnson and Hirsch 1972, Knittle 1972a, Knittle and Hirsch 1968, Narins and Hirsch 1974, Salans gt_gl. 1968, Salans §t_gl. 1971, Salans gt_§l. 1972) but has two main disadvantages. One is cost, since both the osmium tetroxide and Coulter counter are expensive. The other is size limitation since the method does not assess adipocytes smaller than 25/ym. in diameter. This makes the technique of Hirsch and Gallian (1968) especially poor for young animals which have a higher proportion of small cells. The second technique involves microscopic assessment of adipocyte diameters. This has been done on fixed sec- tioned whole tissue (Bjorntorp and Sjbstrom 1971, Hellman 1962, Lemonnier 1971, Lemonnier 1972, Lemonnier et_§l. 1971, Sjbstrbm et al. 1971, Zingg et al. 1962) and on cells isolated from their supporting matrix by the method of Rodbell (1964). 24 Unstained isolated celk3(Bray 1969, Bray 1970, Kovanen §t_al. 1975, Zinder §£_§1. 1967) have been manually measured using an eyepiece micrometer in a microscope. Unstained cells have been mechanically sized by a Zeiss particle size analyzer using photomicrographs (Hirsch and Gallian 1968, Smith 1971) and on non-isolated cells (Goldrick 1968). Isolated cells have also been stained with methylene blue prior to manual sizing (Booth §t_§l. 1974, Herberg et_gl, 1970b, Di Girolamo §£_gl. 1971, Di Girolamo and Mendlinger 1971). The disad- vantage of the sectioning procedure is that the cells are not spherical, as they are when isolated, making calculations of size (such as volume, surface area or weight) less ac- curate. Investigators using no stain or methylene blue may have been sizing lipid droplets in addition to adipocytes since the two are indistinguishable. The latter argument is true for sizing done by the Zeiss particle size analyzer as well. For these reasons a new method of microsc0pically identifying isolated adipocytes has been developed and de- scribed in this thesis. METHODS Experimental Design Ninety-six Osborne Mendel (OM) and ninety-six S 5B/Pll (S) male rat pups were assigned in groups of three or six, on the day following birth, to dams of their respective strains. For a complete representation of the experimental design see figure 1. Half of the dams assigned each litter size were offered a high fat and half a low fat diet. These diets are detailed in table 1. In addition, half of the animals from each strain, diet and litter size were given two supplemental tube feedings each day until weaned. The composition of the supplemental feedings were 24% and 18% corn oil for OM and similarly 20% and 15% for S. Milk was prepared with 15 gm of dry fat free milk solids, the above listed percentages of corn oil and distilled water to a volume of 100cc. This was mixed in a blender. Later groups were randomly reduced in number so that there were ten animals for each group. Five of each group were sacrificed at 24 days and the remainder at 105 days. lThe pedigreed S SB/Pl Cr breeding stock was secured from Samuel M. Poiley, Head, Mammalian Genetics and Animal Production Section DR and D, Chemotherapy, National Cancer Institute, DHEW. 25 26 The diets were designed so that for each Kcal. consumed, the animals would receive equivalent amounts of protein, vitamins, and minerals. The Kcal.:protein ration as cal- culated from Kcal. consumed/gm of dietary protein eaten was 17.3 for both diets. The diets differed in the source of the primary energy component. The low fat diet provided 66% of the calories from the carbohydrate cerelose. This diet contained 3.8 Kcal. per gram based on 4, 4 and 9 Kcal. per gram of protein, carbohydrate and fat respectively. Using the same basis the high fat ration contained 5.7 Kca1.per gram. It contained 44% fat in the form of hydro- genated vegetable shortening.. At weaning (24 days) animals were assigned to one of two groups. One group was sacrificed immediately and the second group was maintained on the same diet their mothers had been given until 105 days of age. At this age the re- maining animals were sacrificed. Adipose tissue from the inguinal region for all rats and from the testicular and perirenal regions of animals representing the extreme ex- perimental conditions for both ages and strains (supplemented, high fat diet, three per litter; and non-supplemented, low fat diet, six per litter) were assessed for adipocyte cell size and number. The percentage of lipid in each depot was also determined. 27 Experimental Conditions All animals were housed in a ventilated room where the temperature remained at 23: 10C. Twelve hours of light were permitted each day. Activity was neither promoted nor restrained. Dams, having previously cast one litter, were mated four to a male. When pregnant, as determined by weight gain, they were transferred to individual plastic cages of the dimensions 33 x 28 x 13 cm. The tOp of the enclosure was wire to permit air circulation. Wood shavings were placed at the base of each cage and were changed at three day intervals. Water and pellets were offered ad libitum. On day one, litters were reduced numerically to rearrange appropriate numbers as previously described. The dams were changed from the pelleted diet to either the high or low fat diet which was available ad libitum in porcelain cups. The pups, so designated, began receiving supplemental feedings. The pups were supplemented daily on days one through twenty-four. Supplementation occurred between 8 and 9 AM and between 4 and 5 PM i 1 hour. Pups were fed up to one milliliter per feeding depending on their size and the full- ness of their stomachs. They were fed with a bulbous syringe inserted into the stomach via the seephagus. As the pups grew older they were allowed to drink and eat the food given their dams. At twenty—four days half of the pups were sacrificed, while the others were individually housed in wire screened 28 bottom metal cages. The dimensions were 18 x 18 x 25 cm. Each rat, in a porcelain cup, had a diet of the same com- positon of his mother. Food and water were available ad libitum. Procedure for Sacrificing Rats and Removing Fat Depots 0n the day of sacrifice each experimental animal was lightly anesthetized with ether and decapitated. Inguinal, testicular and perirenal adipose depots were excised. In the weanling animals this involved both depots, while in the older animals only the left depot was removed. An incision was made down the midline of the ventral side of the rat. The inguinal depot(s) was removed using the ventral midline, lowest rib and urogenital organs as boundaries. The tissue was cooled rapidly than weighed to the nearest milligram. The testicular and perirenal fat organs were subsequently removed and treated in the same manner. Immediatly after weighing, all depots were individually wrapped in aluminum foil, labeled, placed in self-sealing plastic bags, frozen on dry ice and subsequently transferred to a chest-type freezer (—20i2°c.). Depot Lipids Weanling Animals Using forceps, approximately 0.3 gm. of frozen adipose tissue was transferred into a tarred Roese-Goettlieb 29 flaskz. The flask plus tissue were weighed and the difference yielded the wet weight of tissue to be analyzed. A correction was made for any moisture lost during storage according to the following formulae: Moisture lost per tissue sacrifice tissue weight gm sacrifice = weight (gm) _ at analysis (gm) weight (gm) tissue sacrifice weight (gm) Corrected weight _ gm moisture lost weight of of tissue sample T per gm sacrifice X tissue used (gm) weight (gm) + weight of tissue used (gm) Duplicate samples were analyzed wheneverquantity permitted. Ten m1.of diluted 37% HCl (4 parts acid to one part deionized water) were added to each flask. The contents were boiled for thirty minutes in a water bath in the hood. All samples were cooled for at least one hour but never longer than twenty-hour hours. When left in excess of two hours, flasks were stoppered. Deionized water was added until the level of liquid in the flask reached the base of the flask neck. Blanks of diluted hydrochloric acid and water only were treated similarly and handled the same as samples throughout the rest of the analysis. Twenty-five m1.of anhydrous ethyl ether were added and each flask was covered with a rubber stepper. (The same cork and flask remained together throughout the experiment and the part of the cork in contact with the flask contents was not touched). Flasks were shaken sixty times, the stoppers 2Purchased from Mejonnier Brothers Company, Chicago, Illinois. 30 removed and 25 m1. of petroleum ether were added. Each flask was subsequently shaken ninety times and the corks were again removed. Each flask stood thirty minutes while the layers separated. The upper ether layer was decanted into tarred fat f1asks3. Ethyl and petroleum ethers were added then decanted twice more as previously described, however 15 m1. of each ether rather than 25 ml. were used. Prior to the third extraction, deionized water was added to raise the level of the water layer to the tOp of the neck of the Roise-Goettlieb flask. Ether was evaporated until no fumes were discernable. The flasks containing fat were placed in a 700 C. oven for 15 minutes, cooled in a dessicator for 30 minutes and weighed on an analytical balance to an accuracy of four decimal places. Weight of lipid in the flask was calculated by dif- ference. Percent depot lipid was calculated as follows: Percent of weight of lipid in sample (gm) lipid in = corrected wet weight of tissue X 100 wet tissue used (gm) Weight of lipid percent of lipid depot weight . = . . X . . 1n depot in wet t1ssue at sacr1f1ce One Hundred and Five Day Animals Approximately one gram samples of fat depot were minced while frozen. They were then transferred to tarred, labeled 3The fat fla ks had been prepared previously by heating overnight in a 70 C. oven, cooling for thirty minutes in a dessicator then weighing to the fourth place. 31 aluminum pans. Duplicate samples were taken whenever pos— sible. The pans plus minced fat depot were immediatly weighed to the fourth place and the weight of depot in each sample was calculated by difference. All samples were placed in a 70°C. oven for 48 hours. They were then moved to a vaccume oven also at 700C. where they remained an additional 48 hours. The samples were weighed daily thereafter until constant weight (to the third place) was reached. Before each weighing samples were cooled for 30 minutes in the dessicator. Once dry, the samples were stored in dessicators until the lipids were extracted on the Goldfisch fat extraction apparatus. Extraction with ethyl ether was allowed to proceed for 6% hours. This length of time was found sufficient to remove all lipids as evidenced by no additional lipid following re-extraction. Extraction flasks had been previously weighed to the fourth place after drying overnight in a 70°C. oven 15 minutes and cooled in a dessicator for 30 minutes. Following extraction of the lipid, flasks were dried in a 70°C. oven 15 minutes, cooled in a dessicator for 30 minutes and weighed to the fourth place. The weight of lipid was calculated by difference. Calculations to determine depot lipid after correction for moisture losses were made as previously described for 24 day old animals. 32 Assessment of Adipocyte Number and Size Isolation Adipocytes were isolated using a modification of Rodbell's (1964) procedure. Frozen adipose depots were thinly sliced, using a microtome blade. Forty to 100 mg. of tissue were used per sample. Frozen minced tissue was quickly added to clear plastic test tubes. The tubes were tapped to facilitate transfer of the tissue to the bottom of the tube. One ml. of Krebs-Ringer phosphate buffer with bovine serum albumin and collagenase (figure 2) at pH 7.4 was added to each tube. This amount was found to cover the tissue and to contain a sufficient amount of enzyme to release adipocytes. Samples were incubated at 370C. in a shaking water bath for thirty minutes. Fixing and Staining Tubes were removed from the water bath and two drops of 2% cresyl violet were added using a Pasteur pipette. Tubes were gently agitated to facilitate stain diffusion. One drop of the glutaraldehyde fixing solution (figure 3)4 was added with a Pasteur pipette. The tubes were again gently agitated. 4The fixing solution stepped enzyme activity making washing to remove the enzyme unnecessary. This was helpful in maintaining the integrity of the fragile cells. 33 Identification All glass coming in contact with the isolated cells was first coated with silicons. This included microsc0pe slides and pasteur pipettes. Suspensions were prepared the day of measurement to avoid bias from rupture of fragile cells. Tubes were lightly agitated prior to sampling. Small drops, of the stained fixed cell su5pension were transferred to the treated slides using a Pasteur pipette. No cover slips were used. The diameters of 400i 12 adipocytes per depot were optically measured. This was accomplished using an eyepices micro- meter which had been previously calibrated. Calibration checks were made at intervals throughout the study. The magnification of the binocular microscope was 150 power. The eyepiece was 15 power and the objective 10 power. Illuminatunwwas directly from a stage lamp. The microsc0pe had slide stage adjustment controls so that movement of the slide proceeded smoothly along the stage. Isolated adipocytes were round and identified by their rough appearance (basketball-like) and a dark ring around their circumference. In addition most cells had picked up the violet hue of the stain. Cresyl violet stain is specific for protein and does not complex with lipid. Lipid droplets were smooth, the rims were not dark, and never picked up 5Dri-coat, purchased from Fischer Scientific Company, Livonia, Michigan. 34 the stain (figure 4). Membranes and cellular debris will, however, pick up the stain--but will not be round. When— ever too much debris was present, as evidenced by purple colored irregular shapes, the sample was discarded and re- isolated from the frozen tissue. Measurement began at the tsp of the microscopic field for each drop of cell suspension. Assessment then proceeded from left to right, then back to left, etc. until measurement was completed at the bottom of the droplet. Each cell was aligned with a micrometer grid using the stage slide ad— justors. The grids between the grid already aligned and the opposite side of the cell were counted (see figure 5). Tabulation A tape recorder was placed adjacent to the microscOpe. Conveniently, it had a pause arrangement, allowing the person measuring cells to actually record, by the flick of a switch, only after three to four cells had been measured. This made tabulation from the tape to paper more expedient as it cut out all pauses while actual measurement was being made. Cells were measured and their diameters recorded without looking away from the microsc0pe field. Four hundred plus or minus twelve cell diameters were thus measured for each depot. When assessment was approximately complete as judged by the counter on the tape recorder, the tape was played back and the number of cells from each diameter division were tallied. The sum of cells of each diameter was then 35 recorded for future calculation of cell size and cell number (see figure 6). Calculations The volume of cells from each depot was computed using the following formula: volume ( m3) = ("/6) (302+ :2 ) I where the terms Y (mean) and ¢(standard error) are computed from the diameters of the 400 adipocytes measured from each tissue. Computation was done in this manner to overcome the skewedness that would result if volume (x3) were computed in the usual manner. These calculations were done on the Wang calculator, grouped mean program. The lipid content of the "average" cell in each depot was calculated in the following manner: Mean adipocyte lipid mean vglume triglyceride = X . content (ag.) 93m ) den81ty The units of triglyceride density are g./(106)(ym3) as determined by conversion of density in gm/cc toyag/cc and by calculation of the number of/bm/cc. (Measurement was in‘flm.). The density of 0.918 was used throughout the study. Cell number was calculated by dividing the weight of the "average" cell into the tissue lipid content. Cell number = tissue lipid content mean adipocyte lipid content 36 Photography Photomicrographs were made representing many of the tissues and treatments. Both a Leitz 4"x5" and a 35 mm camera were used on a Leitz ortholux microscOpe6. The former used black and white Ektapan film7 and the latter used 35 mm colored films. Both types of pictures were taken at the same magnification, which was determined by the following Leitz magnification formula: bellows extention 25 or: 10 x 10 x 1.25 x (28/25) = X140. Prints of the 4"x5" occular x objective x 1.25 x were made directly from the negative without enlargement. Prints of 35mm slides, however, were enlarged (see figure 4). Statistical Analysis of Data A five way multivariate analysis of variance and uni- variate analysis of variance were used to identify the dif‘ ferences in the data. This was accomplished on the Michigan State University computer and included inguinal cell size in‘flg lipid; cell number in left depot and percent depot fat. While two variables remained fixed, differences between 64"x5" Leitz automatic camera and Leitz Wetzlar ortholux microsc0pe: Distributed by Donald Main and Co., Scientific Instruments, 510 North DBarborn Street, Chicago, Il. 60610. 7Ektapan film: 4"x5", 4162 thick, Eastman Kodac Co., Rochester, NY. 14650. 835 mm film: Eastman Kodac Co., Rochester, NY. 14650. 37 the third variable were analyzed by the Students T test. The three groups included age, strain and treatment. The two treatments were control fed (low fat diet, non-supple- mented, 6/litter) and overfed (high fat diet, supplemented, 3/litter). Data on the forty animals for which inguinal, testicular and perirenal tissue had been analyzed were used. The Michigan State University computer performed these com- putations as well. Students T tests (Chou 1963) for individual group dif- ferences were mechanically performed to decipher differences in cell size and number between the three tissues. Pearsons correlation coefficients were computed on the Michigan State University computer. Cell size and cell number for inguinal fat pads were compared with percent body fat and sacrifice weight in 105 day old animals. RESULTS AND DISCUSSION Inguinal Cell Number Data are summarized in Tables 2 and 3 and Figures 7, 8, 9 and 10. Table 2 gives mean cell number values and standard errors about the mean for each group of five animals. Table 3 lists situations where each variable resulted in a significant (6(L0.05) alteration in cell number. The mean cell number as affected by the various treatments is graphi- cally shown in Figures 7, 8, 9 and 10. Each treatment variable, as litter size, supplementation, ration, strain and age and its influence on cell number is treated separa- tely. Age In general fat cell number in whole left inguinal fat depots significantly increased between 24 and 105 days of age. However,there were three exceptions. These exceptions were: OM, high fat fed, supplemented, 3/litter; OM, low fat fed, non-supplemented, 6/litter; and OM low fat fed, non- supplemented, 3/litter. The number of cell in whole left depots generally doubled with age in similarly treated groups. 38 39 Strain Even though S rats are very resistant to obesity (Schemmel ejLal, 1970a) it was very suprising to note that they had approximately the same number of adipocytes in the left inguinal fat depot as did the OM rats when they were fed the low fat ration. Twenty-four day old, low fat fed, OM rats which had been supplemented and raised three to the litter (cell number in left depots; OM 5.55 i 1.14, S 3.30 i 0.52, all values for cell number are X 106) as well as those non-sup— plemented and raised three to the litter (cell number in left depots; OM 7.42 i 1.88 X 106, S 4.85 i 1.14 X 106) had significantly more cells than similarly treated S rats. When fed the high fat diet there were at least three fold differences in cell number between the two strains whether assessed at 24 or 105 days of age. More specifically, rats fed the high fat ration and which had been supplemented and raised in litters of three had 12.07 i_3.92 X 106 versus 6.23 i 1.67 X 106 cells for 5 rats; if supplemented and raised in litters of six OM rats had 15.07 i 4.24 X 106 and S rats 5.00 i 1.12 X 106 cells; and if non-supplemented and raised six to the litter the 0M had 12.38 i 2.04 X 106 cells whereas 5 rats had 5.96 i 1'47 X 106 cells, all in whole left depots. The latter two sets of 0M rats compared, also had more cells in whole left depots than the S strain at 105 days of age, OM 60.08 .t 25.88 x 106, 5 24.48 i 7.35 x 106; OM 54.79 1: 18.71 X 106, S 11.41 i 3.23 X 106, respectively. Finally 40 high fat fed 105 day OM rats, non-supplemented and raised in litters of three had significnntly more cells (in whole left depots) 39.55 i 12.38 X 106 than similarly treated 8 rats, 13.26 1 5.46 x 106. Ration The most meaningful increase in adipocyte number was caused by feeding 0M rats a high fat diet. Other less meaningful differences related to ration were observed. Significant diet differences were shown in both strains of rats at 24 days of age when supplemented and raised in litters of six. For 0M rats this was a three fold increase. Cell number for OM rats fed the high fat diet was 15.05 i 4.24 X 106 and if fed the low fat diet, 5.55 1 1.14 X 106, in whole left depots. These differences were even more pro- nounced in OM rats at 105 days of age because rats fed the high fat diet had 60.08 1 25.88 X 106 cells and those fed the low fat diet had 15.60 i 2.64 X 106 cells. Although differences in cell number were significant for S rats in the one case, the fact that it failed to occur in the other three situations makes this observation less meaningful. For the 24 day old S strain cell number for the whole left inguinal fat depots were 5.00 i 1.12 X 106 for rats fed the high fat diet and 3.30 i 0.52 X 106 for rats fed the low fat diet. Cell numbers in whole left depots at 105 days of age were 24.48 i 7.35 X 106 and 12.97 i.5-30 X 106 for rats fed high and low fat diets, respectively. 41 Supplementation Dietary supplementation by directing a bulbous syringe into the stomach had little effect on the number of adipose cytes. Although there were significant differences in two situations, both of which were observed in the S 5B/Pl strain, the fact that one caused an increase in cell number and the other a decrease tends to minimize the effect of supplementation, in general. In the 24 day old animals the only difference between supplementation groups was in low fat fed S rats raised in litters of six. The non-supplemented rats had more cells (5.46 i 1.59-X 106 versus 3.30 1 0-52 X 106) than did the supplemented animals in whole left depots. In 105 day old animals, however, supplemented S rats had more cells when raised in litters of six and fed a high fat diet. The supplemented animals had 24.48 i 7.35 X 106 cells and the non-supplemented rats 11.41 i 3.53 cells for the total left inguinal fat depot. Litter Size Generally, adipocyte number was not affected by raising rats in litters of 3 or 6. However, there was one exception. This was in 24 day old rats of the S strain, fed the low fat ration and supplemented. Those raised in litters of three had 4.58 i 0.83 cells while those raised in litters of six had 3.30 i 0.52 cells in whole left depots. Since a significant difference did not persist to 105 days of age, 42 altering litter size did not have a permanent effect on the number of inguinal adipocytes. Summary, Inguinal Fat Cell Number Inguinal fat cell number was significantly increased by feeding OM rats a high fat diet but but this was not true for S rats. Furthermore, in general, there was an increase in fat cell number between 24 and 105 days of age for both strains of rat regardless of diet, supplementation or litter size. At 24 days, S rats bad 1/3 to 1/2 as many adipocytes in the whole left inguinal fat depot as 0M rats. This dif- ference in strains was also true for high fat fed animals at 105 days of age but not for low fat fed rats at 105 days of age. Suprisingly, at this age both OM and S rats had a similar number of fat cells when fed a low fat diet. Adipocyte Size in the Inguinal Fat Depot Mean cell size has been expressed as weight in ng. of lipid per cell in Table 6 and the significance of the treat- ments, that is, the effect of age, strain, ration, supple- mentation and litter size are presented in Table 7. Graphic presentations of these effects are represented in Figures 7, 8, 9 and 10. Meaningful effects of each treatment will be discussed separately. 43 Age In eleven out of sixteen possible treatments, 105 day old rats had significantly larger fat cells in the inguinal fat depot than 24 day old rats when strain, ration, sup- plementation and litter size were held constant. For the OM strain, age caused a significant increase in the quantity of lipid per cell in every case except one. The exception was the group of rats which was fed a low fat diet, non— supplemented and raised 3/litter. In genreal, S rats approximately doubled the weight of lipid in the fat cells of the inguinal fat depot between 24 and 105 days of age. For % of the gruups the increase was significant. Three of these were fed the high fat diet. Further breakdown included: supplemented 3/1itter (162 1.42 for 105 day old rats, 81 i.5 for 24 day old rats--all cell size values are mean cell lipid weight in ng.); supplemented 3/litter (160 i 39 for 105 day rats, 83 i 16 for 24 day old rats); and non-supplemented 6/litter (118 i 34 for 105 day old rats, 70 i 17 for 24 day old rats). Only one low fat fed 5 group of rats showed a significant increase; 90 i 22 for 105 day old rats, 46 i 12 for 24 day old rats, in cell size with age. This was in the group that had been non-supplemented and raised 6/litter. 44 Strain The quantity of lipid per cell was related to the strain of the rat, with OM rats having from two to four times as much lipid in their cells as S rats when the other four variables were held constant (Table 6). From among the sixteen treatment groups, eight were significantly different and these differences were more likely to occur among 105 day rather than 24 day old rats. In general, large standard errors tended to minimize the significance of the other groups despite a two to four fold difference in mean value. More specifically, 24 day old 0M rats fed a low fat ration, supplemented and raised 6/litter had significantly more fat in their fat cells (137 i 35)than did 5 rats (84 i 10). There was nearly a three fold difference by 105 days of age (405 i.l45 in the OM versus 143 i 61 in the 5 rats) which also was significant. Twenty-four day old low fat fed, non-supplemented 0M rats raised 3/litter had significantly more fat in their inguinal depot adipocytes than did S rats (114 i 34 versus 65 i 19). By 105 days of age the mean weight of fat in the inguinal fat depot adipocytes of 0M rats held constant for diet, supplementation and litter size was appro- ximately 2% times as much as S rats. Groups 105 days of age where strain significantly affected cell size included: high fat fed animals, supplemented and raised in litters of six (OM 373 i 70, S 162 i 42); non-supplemented, and raised in either size litter (6/1itter, 0M 285 i_55,S 118 i 34; 3/litter, OM 346 i 43, S 138 i 36); low fat fed animals, 45 non-supplemented and raised in litters six (ON 350 i 110, S 91 i 22). Thus, in animals fed the high fat diet and raised to 105 days of age, the 0M strain always had larger inguinal cells than the S strain when similar groups were compared. In the low fat fed 105 day old animals, those 0M rats raised in litters of six had larger cells than similarly treated S rats regardless of whether or not they were supplemented. Litter Size, Supplementation and Ration The size of the litter in which the rat was reared made a significant difference ($40.05) only in OM, high fat fed, supplemented rats 24 days old. For this group, rats raised in litters of three had larger adipocytes in the inguinal depot than rats raised in litters of six (148 i 27 versus 107 i 37). Feeding supplementary milk by tube resulted in a significant increase in lipid weight per cell only for 24 day old, S low fat fed rats raised 6/litter. The supplemented group had a mean cell weight of lipid of 84 i 10 and the non- supplemented group had a mean cell lipid weight of 46 i.12 ng. per cell. Neither of the above differences persisted into later (105 day) life. Variation in litter size from three to six rats and supplementary feeding of the pups from 1-24 days of age therefore contributed very little to the eventual enlargement of inguinal adipocytes of either strain of rat when the conditions were controlled as in this study. Diet, high versus low fat, was not important either since it never resulted in a significant difference in cell size. 46 Summary, Fat Cell Size in the Inguinal Fat Depot Age and strain were the two most effective variables which significantly increased the size of inguinal adipocytss. When strains were compared, significant differences were most often observed in older animals; when age was compared dif- ferences were most often in OM animals. This was confirmed by the two way interaction observed when the five was analysis of variance was run. This graphically depicted in Figure 15. Relation of Results to Body Weight and Body Fat At 105 days of age inguinal cell number was highly related (r=0.85) to percent body fat. The number of inguinal cans was related to body weight (r=0.70) and percent inguinal fat (r=0.59). Inguinal cell size was related (r=0.74) to body weight at sacrifice and to a lesser degree (r=0.63) to percent body fat. Body fat was also highly correlated (r=0.88) to body weight at sacrifice. Cause of Osborne Mendel Inguinal Obesity The OM strain of rat had larger cells at 24 and 105 days of age regardless of diet, supplementation, or litter size than did the 5 strain. The S strain resists obesity (Schemmel et al. 1970a) and this study showed that the size of their adipocytes had not increased to the size of the OM 47 strain. Thus the OM strain, in gereral, had larger fat cells in the inguinal depot than the 5 strain, when similar ages were compared, through fifteen weeks of age. The OM strain also had more adipocytes in inguinal tissue at 24 days of age, when similar diets were compared, than did the S strain. Both strains showed an increase in the number of inguinal adipocytes with age, but the high fat fed OM strain had three times the number of adipocytes as did all other 105 day old animals. Therefore, when the OM, obesity-susceptible, strain of rat was fed a high fat diet from birth, at 15 weeks of age they gave evidence of hyperplastic obesity. In 105 day animals, when the number of fat cells was expressed on a per gram basis, the 5 strain had at least twice as many cells as the OM strain regardless of diet. For all low fat fed animals the S rats had 7.02 X 106 cells per gram of inguinal tissue while the OM had 2.49 X 106 cells. 6 In animals fed the high fat diet values were 5.78 X 10 and 2.72 X 106 for S and OM strains, respectively. Comparison of Number and Size of Adipocytes in Three Fat Depots In order to compare adipocyte number and size among fat depots, the testicular and perirenal fat depots were also analyzed for cell size and number in four groups of rats at both 24 and 105 days of age. The four groups selected were: both the Osborne Mendel and S 5B/P1 strains fed the low fat ration, non-supplemented, and raised in litters of six; and 48 both the Osborne Mendel and S SB/Pl strains when fed the high fat ration, supplemented, and raised in litters of three. This allowed for comparisons to be made in relation to strain and age for adipocyte number and size as well as the extremes insofar as dietary treatments were concerned. Adipocyte Number Inguinal Tissue Comparisons of age and strain have already been made for inguinal tissue. Age, 105 days versus 24 days, and strain, 0M versus S SB/Pl, made significant differences in the number of inguinal adipocytes. Direct comparisons between the two treatments, that is high fat diet, supple- mented, 3/litter and low fat diet, non-supplemented, 6/litter were not previously made. They were not found to signifi- cantly different (Table 7 and 8 and figure 11). Testicular Tissue Between 24 and 105 days of age fat cell number in the testicular fat pad showed six to thirteen fold increases (Tables 6 and 7 and Figure 12). If these were considered altogether, cell number increased from 1.68 X 106 at 24 days of age to 15.83 X 106 cells at 105 days of age in the whole left depot. At 105 days of age, the Osborne Mendel, low fat fed and the two groups of S SB/Pl rats had comparable numbers of fat cells in the testicular fat depot whereas Osborne Mendel rats fed the high fat diet, raised 3/litter 49 and supplemented had at least twice as many cells as the others. This is an extention of earlier data from our laboratory which indicated that weight of this tissue was substantially increased between 24 and 100 days of age when these rats were fed a high fat ration (Schemmel et al. 1970b). Perirenal Tissue At 24 days of age, the Osborne Mendel rats fed the high fat diet, supplemented and nursed 3/litter had twice as many cells in the entire depot as did the three other groups of rats. Between 24 and 105 days of age there was a seven to seventeen fold increase in the number of cells in this whole left depot. The greatest increase was shown by the Osborne Mendel rats fed the high fat diet, supplemented and raised in litters of three. This was from 5.29 X 106 at 24 days of age to 85.77 X 106 at 105 days of age, both for whole left tissues. Even at 105 days of age, the other three groups of rats had no more than a two fold difference in the number of fat cells. The Osborne Mendel rats fed the low fat diet had even more cells than the S SB/Pl rats fed the high fat diet in whole left depots (Table 6 and 7 and Figure 13). Comparison of the Number of Cells in the Three Depots At 105 days of age, Osborne Mendel rats, regardless of diet, had 1% to 2 times as many fat cells in perirenal tissue as they did in inguinal tissue. For 0M rats, the number of cells in the testicular tissues were comparable to those in 50 the inguinal tissue when evaluated on the entire weight of the tissue. For S SB/Pl rats at 105 days of age, the three fat depots tended to contain the same numbers of cells (Table 6). The trend in adipocyte proliferation seemed to be that the inguinal depot proliferated early in life, followed by proliferation in fat cell number in the testicular and perirenal depots. Fat Cell Size Adipocytes always significantly increased in size with age regardless of anatomical site of the fat depot (Table 8 and 9 and Figures 11, 12 and 13). In general, the increase in cell size was two fold for inguinal tissues but three to eight fold for testicular and three to four fold for peri- renal tissues. Inguinal Tissue Results for age and strain differences with respect to inguinal fat cell size have already been discussed. Treatment extremes, that is high fat diet, supplemented, 3/1itter versus low fat diet, non-supplemented 6/1itter have not been previously compared. They were found significantly different when subgroups, except 105 day old Osborne Mendel rats were compared. All other inguinal groups, that is 24 day old S and OM as well as 105 day old S showed treatment differences at the 0.02 level of confidence. 51 Comparison with Other Data The inguinal tissues, mean of all 105 day OM rats, had larger cells (349 ng./cell) than 10 week (57 ng./cell) or 20 week (136 ng./cell) old Charles River rats in the scapular subcutaneous site (Salans §t_gl. 1972). These Charles River rats had been raised 6/litter on chow-type diet + 5% glucose water. The S strain (mean of all 105 day rats) had inguinal cells (138 ng./cell) similar in size to 20 week old Charles River rats in the scapular subcutaneous site. The size of the S inguinal adipocytes at 15 weeks was comparable to the size of the non-obese male Zucker scapular cells at 26 weeks of age (162 ng./cell) while the OM had cells about twice as large (349 ng./cell)(Johnson §t_al. 1971). The Zucker rats, both non-obese and genetically obese, were raised 8/1itter on a chow-type ration. The size of the obese Zucker rat scapular cells were much larger than either the non-obese Zucker, S or 0M animals. The cells of the obese were 1263 ng./cell as compared to 162 ng./cell, 138 ng./cell and 349 ng./cell, respectively. Twenty-six week old male Zucker rats overfed by raising in litters of four and offering a high fat paste in addition to the chow had cells 212 ng./ cell and 1160 ng./cell for non-obese and obese rats, respec- tively (Johnson et al. 1973). Testicular Tissue In this study testicular fat cell size was similar to inguinal fat cell size when age, strain and dietary treatments were held constant. Levels of significance changed only 52 slightly between the two tissues. Older animals had larger cells (285 ng./cell) than younger animals (63 ng./cell) when compared over all treatment and strains. This was significant at.dE0.000 and for subgroups at‘440.024. The 0M strain had two to three times larger cells (242 ng./cell) than the 5 strain (106 ng./cell), significant ata(=0.013. The above means are representative of all four groups of rats of each strain (Tables 8 and 9). For individual groups the OM strain had significantly larger cells at the 0.024 level of confidence, except 24 day old, low fat fed, non- supplemented rats raised in litters of six where no sig- nificant difference was noted (Figure 12). In general, the testicular fat depot had smaller fat cells if rats were fed the low fat diet, non-supplemented and raised 6/litter than if rats were fed the high fat diet, supplemented and raised 3/litter (£40.01). The exception was the group of Osborne Mendel rats fed the high fat diet, supplemented and raised 3/litter at 105 days of age where cell size was comparable to the group of rats matched for strain and age. Comparison with Other Data Testicular tissue of the male OM rat had cells slightly larger (438 ng./cell in the high fat diet group and 394 ng. /cell in the low fat group) at 15 weeks than the Sprague- Dawley at 12 weeks of age (330 ng./cell) by comparison with Knittle's (1972a) data (for male rats raised 12/litter on a semipurified ration). This was less clear at 15 weeks 53 according to the data of Hirsch and Han (1969) who reported values of (330 ng./cell and of Narins and Hirsch (1974) whether supplemented (348 ng./cell) or not supplemented (267 ng./cell).1 Knittle and Hirsch (1969) reported 15 week old male Sprague-Dawley rats, fed chow-type ration, as having 537 and 281 ng.lipid/cell for rats raised in litters of four or twenty-two, respectively. By 20 weeks of age the cells had expanded to 639 and 375 ng./ce11 in the Knittle and Hirsch study (1969). In any case, the S male rats have smaller fat cells in testicular tissue, 118 and 190 ng./ cell than any of the previously mentioned sizes listed for Sprague-Dawley rats. In fact, the size of fat cells in S rats is comparable 157 ng./cell) to the 40 day old male cell size of supplemented rats raised by Narins and Hirsch (1974). Their Sprague-Dawley rats had 169 ng. lipid per cell. The Charles River strain at 10 weeks of age bad calls about the same size as the S strain (110 ng./cell for Charles River and 157 ng./cell for S) at 15 weeks (Salans §t_gl. 1971). Male 0M rats had larger testicular adipose cells at 15 weeks of age (438 ng./cell in the high fat fed and 395 ng. /cell in the low fat fed) than the Charles River strain at 20 weeks (302 ng./cell, on a chow type diet + 5% glucose water)(Sa1ans et a1. 1971). 1Hirsch and Han (1969) raised their male rats on chow- type diets; Narins and Hirsch (1974) reared their male pups, beginning with lZ/litter but systematically reducing by 2 weeks of age to 6/litter, on a semipurified ration. 54 Johnson §£_al. (1971 and 1973) have reported non-obese male Zucker rats fed chow as having cells containing 344 ng./ cell at 26 weeks of age. This is less than the OM and more than the S both at 15 weeks of age. The obese Zucker rat, however, had larger cells than either the OM or S strain. Male Zucker rats overfed by offering a high fat paste in addition to chew and rearing in litters of four had 1250 ng./ cell and 440 ng./cell in genetically obese and non-obese rats, respectively (Johnson §t_gl. 1973). The latter is similar to the size of 15 week old OM rats fed the high fat diet. Male genetically obese Zucker rats were reported by Bray (1969) as having a mean cell size of 1350 ng./cell, by Johnson §t_§l. (1971 and 1973) of 1225 ng./cell at 26 weeks, and by Lemonnier (1971) of 1275 ng./cell at 10 weeks of age (all reared on chow type diets). Perirenal Tissue Between 24 and 105 days of age there was a four fold increase in the size of perirenal fat cells for male Osborne Mendel rats fed the low fat diet. It was less (2% fold) for UN rats fed the high fat diet (Tables 8 and 9 and Figure 13). However, themehad been a large increase in the number of fat cells between these two ages for this latter group of rats and it would suggest that the perirenal tissue had not yet reached its full capacity for growth in rats 105 days of of age. Previous data from our laboratory indicated that this tissue, in OM rats fed a high fat diet, extensively increases in weight after 105 days of age (Schemmel et a1. 55 1970b) which further suggests incompleteness of develOpment. In general, the fat cells in the perirenal depot were not as large as fat cells of the inguinal or testicular depot for male 0M rats regardless of dietary treatment. Likewise, between 24 and 105 days of age, fat cell size in the perirenal depot for S rats increased three fold. However, unlike the OM rats, perirenal fat cell size was similar to the cell size of inguinal and testicular fat depots. Comparison with Other Data Hirsch and Han (1969) have reported Sprague-Dawley perirenal adipocytes as being larger than the S or OM cells in perirenal tissue. At 15 weeks of age the chow—fed Sprague -Dawley rats had cells which were 270 ng./cell while the S and 0M strains had 111 and 227 ng./cell, respectively, at the same age. By 26 weeks of age the cells in Sprague-Dawley rat retroperitoneal tissue had expanded further to 550 ng./ cell. Johnson §t_gl. (1971 and 1973) have found both obese (15 weeks 1500 ng./cell, 26 weeks 1180 ng./cell) and non- obese (15 weeks 400 ng./cell, 26 weeks 490 ng./cell) male Zucker rats raised 8/litter on chow type rations to have larger perirenal fat cells than either the ON (227 ng./cell) or the S strain(1ll ng./cell) at 15 weeks of age. This was even more pronounced when genetically obese or non-obese Zucker rats were overfed by rearing in litters of four and 56 offering a high fat paste in addition to the chow diet (894 ng./cell and 553 ng./cell, respectively)(Johnson et al. 1973). Strain and Adipocyte Density At 15 weeks of age the S strain consistently had more cells per gram of adipose tissue than did the 0M strain. This was true regardless of tissue or diet. Collectively utilizing all data on 105 day old rats fed the high fat diet the 5 male rats had 5.78 X 106 while the OM had 2.73 X 106 cells per gram of wet tissue in the inguinal fat depot (depot weights 3.01 gm. and 18.20 gm., respecitvely). When fed the low fat diet values were 7.02 X 106 and 2.49 X 106 for S and OM strains, respectively (depot weights 2.24 gm. and 6.31 gm.). Results in testicular tissue in similar order were: high fat 5.81 X 106 and 1.81 X 106 (depot weights 1.09 gm. and 11.21 gm.); low fat 7.49 X 106 and 2.67 X 106 (depot weights 1.49 gm. and 5.77 gm.). In perirenal tissue values in similar order were 7.50 X 106 and 5.46 X 106 for high fat fed rats (depot weights 2.29 gm. and 14.82 gm.) and 10.17 X 106 and 3.68 X 106 for low fat fed rats (depot weights 2.06 gm. and 6.75 gm.). All calculations of cellular density were carried out on 15 week old animals. It is interestkg that the S rat, which is resistant to obesity even when fed a high fat diet (Schemmel et al. 1970a) has a greater complement of fat cells than the obesity- susceptible 0M strain when expressed per gram wet tissue. Some control mechanism must prevent the 5 strain adipocytes from expanding, 57 thus maintaining the more cellularly dense, yet smaller cells and fat depots. Furthermore, a high fat diet did not cause hyperplasia in the S strain of rat. Implications for Human Obesity The difference between the OM and S strains of rat provide a model for human obesity. Some humans are resistant to obesity as are the S strain of rat. Other less fortunate humans are susceptible to obesity of hypertrophic, hyper- plastic, or their combined origin. The OM strain of rat provides a model for the investigation of obesity for such people. The obesity of the OM strain at 15 weeks of age has been associated with hypertrophy when fed a low fat diet from birth and with hyperplasia and possibly also hypertrophy when fed a high fat diet over the same period relative to the S strain. This makes the OM strain excellent as a model for human obesity. The cellulaiity of the S strain, too, invites further investigation as to the control of the non-obese state. 58 TABLE l.--Diet Composition. Ingredients gm./67 gm. diet gm./100 gm. diet high fat low fat proteina 22.0 22.0 salt mix 4.0 4.0 fiberc 2.0 2.0 vitamin mix e 1.0 1.0 DL-methiogine 0.1 0.1 liver mix 1.5 1.5 fat: crisco 29.51 --- corn oil --- 3.0 carbohydrateg 7.0 66.39 Total Kca%.h 381 381 Kcal./gm. 5.7 3.8 L_‘ aCasein: General Biochemicals, Chargin Falls, Ohio. Rogers and Harper salt mix: General Biochemicals, Chargin Falls, Ohio. CCellulose type: General Biochemicals, Chargin Falls, Ohio. dA.0.A.C. Vitamin mix: General Biochemicals, Chargin Falls, Ohio. Supplied the following (gm./kgm. diet): p-aminobenzoic acid, 0.10; B1 , (0.1% in mannitol), 0.03; biotin, 0.0004; calcium panto%henate, 0.04; choline, free base, 2.0; folic acid, 0.002; 1-inositol, 0.10; menadione, 0.005; niacin, 0.04; pyrodoxine HCl, 0.04, riboflavin, 0.008, thiamin HCl, 0.005; dextrose, anhydrous, q.s.; (units/kgm) Vitamin A, 20,000.00; Vitamin D2, 2,000.00; Vitamin E acetate, 100.00. 8Purchased from General Biochemicals, Chargin Falls, Ohio. fPurchased from General Biochemicals, Chargin Falls, Ohio. gCerelose, purchased from Michigan State University Stores, East Lansing, Michigan. hValues used for calculating Kcalories were 4, 4 and 9 for one gram of protein, carbohydrate and fat, respectively. 59 TABLE 2.--Mean fat cell number and standard error in the entire left inguinal fat depot for each group of animals - STRAIN DSBDRNE MENDEL S 5B/P1 DIET SUPPL.a LITTER CELL STANDARD STANDARD FAT SIZE NUMBEg ERRDR ERRDR 105 DAY DLD L0w YES 6 15.60 2.64 12.97 5.30 LOW YES 3 14.39 3.83 16.55 8.69 Low N0 6 16.53 6.16 16.66 5.23 LOW N0 3 14.79 5.99 14.13 4.07 HIGH YES 6 60.08 25.88 24.48 7.35 HIGH Yes 3 40.19 24.05 19.41 3.20 HIGH N0 6 54.79 18.71 11.41 3.23 HIGH N0 3 39.55 12.38 13.26 5.46 24 DAY DLD Low YES 6 5.55 1.14 3.30 0.52 Low YES 3 7.84 2.65 4.58 0.83 Low N0 6 8.00 4.61 5.46 1.59 L0w N0 3 7.42 1.88 4.85 1.14 HIGH YES 6 15.07 4.24 5.00 1.12 HIGH YES 3 12.07 3.92 6.23 1.67 HIGH N0 6 12.38 2.04 5.96 1.47 HIGH N0 3 12.69 8.51 6.04“ 1.67 asupplementation; yes means supplemented and no means not supplemented. 60 TABLE 3.-—List of significant differences (440.05) in left inguinal cell number between ages, strains, rations, supp- 1ementation groups and litter sizes when the other four variables are held constant. VARIATION AGEa STRAINb DIETc SUPPLEa BY (days) (fat MENTED content) AGE8 5 16w no three 5 low no six 5 low yes three 5 low yes six 5 high no three 5 high no six 5 high yes three 0M low yes three 0M low yes six 0M high no three 0M high no six OM high yes six STRAINb 24 low no three 24 low yes six 24 high no six 24 high yes three 24 high yes six 105 high no three 105 high no six 105 high yes six RATIDNC 24 5 yes six 24 0M yes six 105 5 yes six 105 UN no three 105 GM no six I 105 UM yes six SUPPLEMEN- TATION 105 5 high six LITTEQ SIZE 24 5 low yes L a105 days versus 24 days. bOsborne Mendel (0M) versus 5 SB/Pl (S). CHigh fat diet versus low fat diet. dSupplemented versus non-supplemented. eThree per litter versus six per litter. 61 TABLE 4.--Mean cell size and standard error, weight in ng., in left inguinal depots for each group of five animals. expressed as STRAIN DSDDRNE MENDEL 5 5B/Pl l DIET SUPPL. LITTER MEAN wEIGHT STANDARD MEAN WEIGHT STANDARBI FAT SIZE LIPID/CELL ERRDR LIPID/CELL ERRDR ng. ng. 105 DAY DLD L0w YES 6 405 145 143 61 L0w YES 3 361 141 177 92 LOW N0 6 350 109 90 22 LOW N0 3 279 143 116 28 HIGH YES 6 373 72 162 42 HIGH YES 3 391 153 160 39 HIGH N0 6 285 55 118 34 HIGH ND 3 346 43 138 36 24 DAY DLD L0w YES 6 137 35 84 10 Law YES 3 112 36 81 15 Low N0 6 83 30 46 12 L0w N0 3 114 34 65 19 HIGH YES 6 107 36 81 8 HIGH YES 3 148 27 83 16 HIGH N0 6 101 23 7D 17 HIGH ND 3 138 52 88 25 not supplemented. aSupplementation; yes means supplemented and no means 62 TABLE 5.—-List of significant differences «1(0.05) in left inguinal cell size, expressed in ng., between ages, strains, rations, supplementation groups, and litter sizes when the other four variables are held constant. VARIATION AGEa STRAINb DIETc SUPPLEa LITTEg BY (days) (fat MENTED SIZE content) A558 5 low no six S high no six 5 high yes three S high yes six 0M low no six OM low yes three 0M low yes six OM high no three 0M high no six OM high yes three 0M high yes six STRAINb 24 low no three 24 high yes three 105 low no six 105 low yes six 105 high no three 105 high no six 105 high yes three 105 high yes six RATIDNC SUPPLEMEN- 24 5 low six TATION LITTE 24 OM high yes SIZEa _J 8105 days versus 24 days. stborne Mendel (OM) versus 5 SB/Pl (S). CHigh fat diet versus low fat diet. dSupplemented versus non-supplemented. eThree per litter versus six per litter. 63 TABLE 6.--The effect of age, diet and strain on mean adi— pocyte number in three selected depots of male rats. (n=5) AGE 24 DAY DLD 105 DAY OLD DEPOT STRAIN TREAT- EELL STANDARD EELL »STANDARD MENT NUMBER ERROR NUMBER ERROR 1 (x 10 ) ' (x 10 ) I 0Ma Cb 8.00 4.61 16.65 6.16 N , G 0c 12.07 3.92 40.18 24.05 U I N 5B/Pl C 5.46 1.59 16.52 5.23 A L O 6.22 1.67 19.41 3.20 T E 0M c 1.94 0.43 13.26 3.21 S T 0 2.11 0.84 26.94 8.98 I E 0 SB/Pl c 1.32 0.43 11.19 1.93 L . A 0 1.35 0.25 11.95 7.43 R P OM C 1.65 0.38 24.98 9.39 E R 0 5.29 1.00 85.77 47.91 I R E 5B/Pl c 1.09 0.42 11.17 4.09 N A 0 2.39 1.52 16.65 7.37 L a OM: Osborne Mendel strain. bc; control (low fat diet, non-supplemented, 6/litter). C0; overfed (high fat diet, supplemented, 3/litter). 64 .pcmnmmmwn zdvcmowmflcmflm voc "mzm .camswm Hmocmz mcnonm0 ”20U .Hmpvfla\m .UmpcmEmadaom .pmwv vow Low; “ummnm>o "00 .Hmppfla\0 .UmpcmEmaaaomlco: .umfin vow 30a “Houpcoo nun .A0v ummum>o Ho Auv Homecoo "PcmEpmmHem mz Hm.~m m~.m mz 00.0a No.0 mz pv.ma wm.aa 0000(4400 emeo.o Ne.me em.em_ memo.o em.e~ em.ma m2 ee.oe No.84 20 em 6% m2 00.0a NH.HH mz mm.ad ma.aa mz av.ma 00.0H Hm\mm 0 00a 0 N000.0 mm.m mm.a mz Ha.m am.a mz ~0.NH 00.0 20 cm a 02 mm.m m0.H m2 [mmnH 00.H mz NN.0 mmnm HMNmm m am e .szm 00 20Hm 00 .zuHm 00 T 3u>uu o u Su>uu o o 3u>us 1 o u zHemem owe mz av.mm mm.» 02 ~0.HH mu.m mz ma.ma vm.aa numm<4400 0NH0.0 -.mm 00.0a Nav0.0 vm.0m mm.aa mz 0H.0v av.ma 0 00a 2 amm0.0 mm.vm NH.HH m2 0m.ma ma.aa mz mm.wa 00.0a 0 00H H moa0.0 mm.m mm.m mz HH.N mm.a 0000.0 ~0.NH NN.0 0 am < .Illle mo.H 00.H NW mz mm.H DNMH mz 00.0 wwwm m am 0 .szm 00 . . zmHm 00 .szm m0 #202 e 00>03 so He\mm m au>uu zo 46\mm 30>00 zo .rwmmmwue<0 m><0 m><0 .zuH ><0 mz 2H00 mos em ao>ua was em So>os mos em fipewee sezoeHeoe _ eelsuflemue Jeszqu _T eoeoq 9 .mmaanHm> mouse >0 mpoqmo pom pmma mouse cw m 0H x Romano aamo mo comflumdeo0|l.n 040o "00 .Hmp»aa\m .Umvcmemaaoomlcoc .pmfiu pm; Boa “Hoapcoo nun .A0v nmmnm>o no Auv HHopcoo "9:02pmmuem m2. 444 eoe m2 ean Hes m2, Noe Noe, oemseJJoo m2 eea eem m2 eme eme ms eem «em so eoe me m2 eeH ee ooeo.o ome eds emao.o eeH ee Ho\me m eoa s m2 ee ee eooo.o No es emoo.o eea es so em e m2 ee em flmeooo . me we Neao.o e» we Hexme e em s wwnuo on sow \ .onm Lo DAT e oo>oo o o 30>00, o o om>mo o o 2Heo so 46\me m 3o>o0 so HE\me m 3e>eo so HE\me m nesese mos m ooo.o med oe oooo.o eem me Nooo.o new ee oemesauoo eoo.o eee ee mmoo.o eon , No eeeo.o eem eea o so eooo.o eem ee eeoo.o eee ee eaoo.o , eem es o Uso o emeo.o eee ee Nooo.o ooe we omoo.o eea ee Uo H6\me m o omoo.o es em mooo.o e44 me eooo.o ee me am» eo\me m <1 mew so (mseo mweo Hm,eo,msoq m»e0 eoe so Josue eoa em 3o>e3 eoe em neseee fl. oszososee _ ssoooeemee 3<2HoozH eoeeo so amanmeum> mmnre ho mwoomo pm» pmma mmncp ce ..0c ca armemz mm ommmmnoxm .mNem Hamo mo cowflHmQEoUIl.m mum<0 00H m><0 vm 68 Solution X: 27.8 gm. monobasic sodium phosphate1 in 1000 m1. deionized water. Solution Y: 53.65 gm. NaZHPO4 . 7H 01 OR 71.7 gm. of Na2PO . 12H20; ID 1006 ml. deionized water. (0.2M? To achieve a pH of 7.4, 19 ml. of solution X and 81 m1. of solution Y are combined and diluted to 200 ml. with deionized water. The pH is again adjusted to 7.4. For adipocyte isolation work this solution should contain 4% bovine serum album'n. Immediatly prior to use 3mg./ml. bacterial collagenase is added. The purification of this enzyme varies so that the concentration is not exact and incubation time may require adjustment. The enzyme is not stable at room temperature. It should be stored frozen and used within three months of purchase. 1Purchased from Fischer Scientific Company, Chemical Manufacturing Division, Fairlawn, N.J. 07410. Purchased from Worthington Biochemical, Freehold, N.J. FIGURE 2.--Krebs-Ringer Phosphate Buffer. 69 S-Collidine buffer: 4.65 gm. of 2,4,6 tri—methyl pyridine were weighed into a 100 m1. beaker. Approximately 70 m1. of deionized water were added. The pH was adjusted to 7.4 using a pH meter and 50% HCl. The solu- tion, at pH 7.4, was transferred to a 100 ml. volumetric flask which was then filled. Glutaraldehyde Solution: 30 m1. of 25% glutaraldehyde and 31.25 ml. of S-Collidine buffer (above) were diluted to 250 ml. in a volumetric flask. The pH was again adjusted to 7.4. FIGURE 3.--Glutara1dehyde Solution. Figure 4.--Photomicrographs. 70 Adapocytes as they appear, stained, microscopically. Adipocytes, dark circular shaped objects, and lipid droplets, clear round shapes. Magnification 140 power. 71 TITI"V'V IIIVIj‘I ‘VIj—I I tr 7 1‘0 ‘20 "T ‘ ""‘40 FIGURE 5.--Assessment of Adipocyte Diameter. The adipocyte is aligned, using stage adjusters, with the grid marked "0". The number of grids the adipocyte spans, in this case 22, is recorded. 72 1 2 3 4 5 6 7 8‘ 9 10 11 12 13 14 15 16 l7 18 19 I NIINNMNMMM‘WNMMMI H m pun pawl-INHMINI I In ' Each cell was tallied in a column representing its specific width as determined in Figure 5. The total number of cells in each column was then recorded for further calculations of surface area, volume, weight, and cell number. For the above situation, this would be: 2 3 4 5 6 7 8 9 0 11 12 13 14 15 16 17 18 19 10 7 ll 13 9 10 11 8 6 3 1 0 0 FIGURE 6.--Tallying of Sized Adipocytes. 73 3/litter - 6/litter 20. Q 10. , LOW FAT M E A N 0 . - as . , 2_1 C E 40. 1 L L N u M 301 d B E R 6 x 10 20‘ ‘ 101 4 HIGH FAT O. Y V _V v ' t 0 100 200 300 400 0 100 200 300 400 WEIGHT, ng. FIGURE 7.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for twenty-four day old S SB/Pl rats. High fat and low fat diets are represented in rows; litter sizes in columns; and supplementedU or not supplemented. within each graph. 74 3/litter 6/litter 20] 10 LOW FAT- M E A N O ' V fi‘ fi F i C E 40. L L N u M 30. B E R x 106 20, 10 HIGH FAT U 10 200 300 400 1 0 200 300 400 WEIGHT, ng. FIGURE 8.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for one hundred and five day old 5 SB/Pl rats. High fat and low fat diets are represented in rows; litter sizes in columns; and supplemented E] or not supplemented Iwithin each graph. 75 3/1itter 6/litter 20‘ 1 10' ‘ LOW FAT M E A N 0 . 1 1 . . . fl C E 404 9 L L N U M 30" d B E R X 106 20< . 10 HIGH FAT D r v 1 ~ f v —r 100 200 300 400 0 100 200 300 400 WEIGHT, ng. FIGURE 9.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for twenty- four day old Osborne Mendel rats. High fat and low fat diets are represented in rows; litter sizes in columns; and supplementedD, or not supplemented I within each graph. 3/litter 6/1itter 20 10 LOW FAT M E A 0 0., N C 40 E L L N U 30 M B E R 20 x 106 10 HIGH FAT 0 , _, U 100 200 300 400 I 100 200 300 400 WEIGHT, ng. FIGURE 10.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left inguinal depots for one hundred and five day old Osborne Mendel rats. High fat and low fat diets are represented in rows; litter sizes in columns; and supplementedU, or not supplemented I within each graph. 77 5 5B/Pl OSBORNE MENDEL 20‘ 7 24 days 1 . 10. u M E A N U V V 1 1 I C E L 40‘ L N u M B 30‘ - E R x 106 20. - 10 105 days 0 0 100 200 300 400 WEIGHT, ng. FIGURE ll.--Mean cell number X 106 versus mean cell size, Expressed in ng. lipid, in left inguinal depots. Strains are represented in columns; ages are represented in rows; arm: treatments, control (low fat diet, non-supplemented, 6/J.itter) I , or overfed (high fat diet, supplemented, 3/litter) [:1 within each graph. 78 S 58/81 OSBORNE MENDEL 20 5 24 days 7 j 10* 4 M . , E A I N oil. . t . ll, 0 E 40. . L I L N U M 30. . B E R x 106 201 .7 10 105 days 0 100 200 300 400 0 100 200 300 400 WEIGHT, ng. FIGURE l2.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left testicular depots. Strains are represented in columns; ages are represented in rows; and treatments, control (low fat diet, non-supplemented, 6/litte’r) I , or overfed (high fat diet, supplemented, 3/litter) Cl within each graph. 79 S 58/81 OSBORNE MENDEL 201 . 24 days. 4 10. 1 z:>n13 C) d 7 d J 1 d 1 C E 40. 85 L L N U 1 M 30 4 B E R 6 X 10 20‘ 10 105 days 0 T 1 fl 0 100 200 300 400 0 100 200 300 400 WEIGHT, ng. FIGURE l3.--Mean cell number X 106 versus mean cell size, expressed in ng. lipid, in left perirenal depots. Strains are represented in columns; ages are represented in rows; and treatments, control (low fat diet, non-supplemented, 6/litter) I, or overfed (high fat diet, supplemented, 3/1itter) D , within each graph. 80 M 501 E A N I N 401. G U I N A L 30‘ C E L L 201 N U M B E R 10 D X 106 l I 24 days 105 days AGE FIGURE l4.-—Mean inguinal cell number X 106 versus age in left depots. The Osborne Mendel strain , S SB/Pl strain ---—, high fat dietA and low fat dietA are re- presented. The three way interaction between age, strain and diet on left inguinal cell number is significant (p(0.001). 81 M 400' E A N I N 300. G U I N A L 2001 C E L L ’,---'. 1001 ”—v"’ W "”._- E .y"' I G H T 0 fi_ 0 241days lOS‘days ng. AGE FIGURE 15.--Mean inguinal cell size, expressed in ng. lipid per cell, versus age in left depots. The Osborne Mendel , and S SB/Pl strains --- are represented. Five way multi- variate analysis of variance found the two way interaction between age and strain for cell size significant (p(0.0001). CONCLUSIONS Male Osborne Mendel rats have larger fat cells than do the S SB/Pl rats. In 24 day old rate values for 0M and 5 strains in inguinal tissue were 118 ng. lipid/cell and 75 ng. lipid/cell, respectively; in testicular tissue 74 ng. lipid/cell and 61 ng. lipid/cell; and in perirenal tissue 68 ng. lipid/cell and 40 ng. lipid/cell. One hundred and five day old animals had values of 349 ng. lipid/cell and 138 ng. lipid/cell in inguinal tissue; 419 ng. lipid/cell and 157 ng. lipid/cell in testicular tissue; and 232 ng. lipid/cell and 116 ng. lipid/cell in perirenal tissue, respectively. However, when the OM strain was given a high fat (44% w/w) diet from birth, at 15 weeks of age they had at least twice as many adipocytes in inguinal, testicular and perirenal tissues as did those fed a low fat diet. In the inguinal tissue the high fat fed OM rat had 48.65 X 106 cells while 15.33 X 106, 17.14 X 106 and 15.08 X 106 were values for the number of cells in low fat fed 0M rats and for S rats in whole left depots, fed high and low fat diets, respectively. Testicular tissue had 26.95 X 106 as compared to 13.27 X 106, 11.95 X 106 and 11.20 X 106 cells while perirenal tissue had 85.78 X 106 versus 24.98 X 106, 16.66 X 106 and 11.18 X 106 for UN rats raised on 82 83 high fat diets versus 0M rats raised on low fat diets, S rats raised on high and low fat diets, respectively (all cell number figures are in whole left depots). Thus offering a high fat diet from birth, used in conjunction with an obesity-susceptible strain of rat (Osborne Mendel), can induce hyperplastic obesity. This is not true of the obesity-resistant S 5B/Pl strain, however. Although there were trends which may have been sug- gestive of some effects on fat cell number and size from raising rats in litters of three rather than six or tube feedings, these were not statistically significant. Aging, between 24 and 105 days, increased the size and number of adipocytes as did being of the OM rather than the S strain. In addition, the feeding of a high fat diet from birth was important in increasing the number of fat cells in the Osborne Mendel strain. It had not affected the number of cells in the S SB/Pl strain by 15 weeks of age and had not affected the size of cells in either strain. SUGGESTIONS FOR FURTHER STUDY Control of Obesity Why are animals of the S strain smaller in body weight, adipose cell size and adipocyte number than animals of the OM strain? Is there any way to alter this trend? Since the pups of the S strain have smaller and fewer adipocytes at 24 days of age, are born in smaller litters, weigh less at birth, and are supplied with a naturally less calorically dense milk than OM pups it is likely that the former are undernourished both postpartum and in utero rela- tive to the OM. To see if altering prenatal nourishemnt can affect the above, one horn of the 5 strain female uterus could be ligated. Fewer pups per pregnancy would result and they might therefore receive more nourishment. Investi-. gation of the question of caloric density of milk could pro- ceed by raising pups from each strain by mothers of the opposite strain from birth. Given that animals of the 5 strain are smaller at weaning than animals of the OM strain, why do they continue to be smaller in size? Work in our laboratory has shown that the 5 strain eats less food and conVerts food calories to body energy less efficiently than the OM strain (Schemmel gt_gl. 1972). Do they also have tighter appetite control or 84 85 regulation than the 0M strain and why should this be the case? What controksthe size of adult fat depot adipocytes? Research in our laboratory is underway concerning the quantity of fecal fat excreted in each strain. The S 5B/Pl strain may well be less able to absorb dietary fat than the 0M strain. This would account, at least in part, for the reduced efficiency of the S relative to the OM to convert dietary calories to body energy. Biochemical studies might clarify why the 5 strain is a less efficient converter of dietary energy to body energy too. When fed the high fat diet the activity of lipoprotein lipase would give information concerning availability of circulating triglycerides to fat cell absorption. The activities of acetyl coenzyme A car- boxalase and of the fatty acid synthetase complex would yield similar information for de novo synthesis of trigly- cerides. Microscopic Versus Coulter Counter Sizing The data collected on the 105 day OM strain raised on the high fat diet should be checked since such adipocytes are large and more susceptible to rupture by collagenase. This could be done by the method of Hirsch and Gallian (1969). An attempt was made to fix the cells with osmium tetroxide and size them microscopically. This was futile due to the irregular shape of the resulting cells. 86 Proliferation Pattern How does the cellular pattern of adipose tissue deve- lopment differ between the two strains? The 5 strain had fewer cells than comparably treated OM rats at 24 days of age. At 105 days of age all animals of both strains, except the OM rats fed the high fat diet, had approximately the same number of cells in their whole left inguinal depots. Is the pattern of adipose tissue proliferation over time altered in overfed animals? Knittle and Hirsch (1968) have 7 recorded data which indicate that proliferation of the number of adipocytes might continue beyond the normal fifteen week limit in epididymal tissue. Overfeeding was accomplished, in their study, by altering litter sizes. The length of time over which proliferation continues as well as the re- sulting final number of adipocytes formed would shed further light on the results of overfeeding the very young. Sex Difference bf Are females of either strain more susceptible to hyper- plastic obesity than males? Lemonnier (1971) has observed that females are more susceptible to hyperplastic obesity than males. Work in our laboratory has shown that, in the Osborne Mendel strain, females accumulate more genital fat while males accumulate more inguinal fat (Schemmel gt_§l. 1970b). How would the females of both strains react, on the cellular level in adipose tissue, to high and low fat diets? LITERATURE CITED LITERATURE CITED Asher, P. 1966. Fat babies and fat children, the prognosis of obesity in the very young. Arch. Dis. Childh. 41: 672. Barboriak, J.J., W.A. Krehl, G.R. Cowhill and A.D. Whedon. 1958. Influence of high-fat diets on growth and de- velOpment of obesity in the albino rat. J. of Nutr. 64:241. Beal, V.A. 1957. On acceptance of foods, and other food patterns, of infants and children. Ped. 20:448. Bjarntorp, P. and L. Sjbstrbm. 1971. Number and size of adipose tissue fat cells in relation to metabolism in human obesity. Metab. 20(7): 703. Booth, M.A., M.J. Booth and A.W. Taylor. 1974. Rat fat cell size and number with excercise, training, detraining and weight loss. Fed. Proc. 33:1959. Bray, G.A. 1969. Studies on the composition of adipose tissue from the genetically obese rats. Proc. of the Soc. for Exp. Biol. Med. 131:1111. , 1970. Measurement of subcutaneous fat cells from obese patients. Ann. of Inter. Med. 73:565. Brook, C.G.D. 1971. Composition of human adipose tissue ' from deep and subcutaneous sites. Br. J. Nutr. 25:377. Brook, C.D.G., J.K. Lloyd and O.H. Wolff. 1972. Relation between age of onset of obesity and size and number of adipose cells. Br. Med. J. 3:25. Brook, C.D.G. and J.K. Lloyd. 1973. Adipose cell size and glucose tolerance in obese children and effects of diet. Arch. Dis. Childh. 48:301. Chou, Ya-Iun. 1963. Statistical Analysis with Business and Economic Applications. Holt, Rinehart and Winston, New York. 87 88 Di Girolamo, M., S. Mendlinger and J.W. Fertig. 1969. Abstract. The role of adipose cell size, dispersion and number in the enlargement of the epididymal fat pads in rat, hamster and guinea pig. Clin° Res. 17:22. Di Giroalmo, M. and S. Mendlinger. 1971. Role of fat cell size and number in enlargement of epididymal fat pads in three species. Amer. J. of Phys. 221(3):859. Di Girolamo, M., S. Mendlinger, and J.W. Fertig. 1971. A simple method to determine fat cell size and number in four mammalian species. Amer. J. of Phys. 221(3):850. Dymza, H.A., D.M. Czajka, and S.A. Miller. 1964. Influence of artificial diet on weight gain and body composition of the neonatal rat. J. of Nutr. 84:100. Emery, R.S., J.D. Benson and H.A. Tucker. 1971. Dietary and hormonal effects on extended lactation and lipid metabolism in rats. J. of Nurt. 101:831. Enesco, M. and C.P. Leblond. 1962. Increase in cell number as a factor in the growth of the organs and tissues of the young male rat. J. of Emborl. Exp. Morphol. 10:530. Eid, E.E. 1970. Follow-up study of physical growth of children who had excessive weight gain the first six months of life. Br. Med. J. 2:74. Eden, A.N. 1975. Growing up thin. Amer. Baby. July, p. 36. Fomon, S.J., L.J. Filer,jr., and L.N. Thomas. 1969. Rela- tionship between formula concentration and rate of growth in normal infants. J. of Nutr. 98:241. Fomon, S.J. 1974. Infant Nutrition. W.B. Sanders Company, Philadelphia, Pa. Goldrick, R.B. 1967. Morphological changes in the adipocyte during fat deposition and mobilization. Am. J. Phys. 212(4):777. Greenwood, M.R.C. and J. Hirsch. 1974. Postnatal development of adipocyte cellularity in the normal rat. J. of Lipid Res. 15:474. Guthrie, H.A. 1966. Effect of early feeding of solid foods on nutritive intake of infants. Ped. 38:879. Hammond, J. 1955. Deposition of fat in farm animals. Adv. in Sci. p. 126. 89 Han, P.W., C-H. Lin, K-C. Chu, J-Y. Mu and A-C lui. 1965. Hypothalamic obesity in weanling rats. Am. J. of Phys. 209(3):627. Han, P.W. and A-C Lui. 1966. Obesity and impaired growth of rats force fed 40 days after hypothalamic lesions. Am. J. of Phys. 211(1):229. Heggness, F.W., D. Bindschadler,J, Chadwick, P. Conklin, S. Hulnick, and M. Oaks. 1961. Weight gains of over- nourished preweanling rats. J. of Nutr. 75:39. Hellman, 8., I-B. Taljedal and S. Westman. 1962. Morpho- logical characteristics of the epididymal adipose tissue in normal and obese-hyperglycemic mice. Atca Morph. Neer. Scand. 5:182. Herberg, L., M. Bergmann, A. Hennings, E. Major and F.A. Gries. 1970a. Influence of diet on the metabolic syndrom of obesity. Israel J. Med. Sci. 8(6):822. Herberg, L., F.A. Gries, C. Hesse-Wortmann. 1970b. Effect of weight and cell size on hormone-induced lypolysis in New Zealand obese mice and American obese-hyperglycemic mice. Diabetolgia. 6:300. Herberg, L., W. Doppen, E. Major, and F.A. Gries. 1974. Dietary-induced hypertrophic-hyperplastic obesity in mice. J. of Lipid Res. 15:580. Hirsch, J., J.L. Knittle and L.B. Salans. 1966. Abstract. Cell lipid content and cell number in obese and non- obese human adipose tissue. J. of Clin. Invest. 45:1023. Hirsch, J. and E. Gallian. 1968. Methods for the deter- mination of adipose cell size in man and animals. J. of Lipid Res. 9:110. Hollenberg, C.H., A. Vost, and R.L. Patten. 1970. Adipose mass: control of fat cell development and lipid content. Recent Prog. in Horm. Res. 26:463. Johnson, M.L., B.S. Burke, and J. Mayer. 1956. The pre- valence and incidence of obesity in a cross-section of elementary and secondary school children. Am. J. Clin. Nutr. 4:231. Johnson, P.R., L.M. Zucker, J.A.F. Cruce and J. Hirsch. 1971. Cellularity of adipose depots in the genetically obese Zucker rat. J. of Lipid Res. 12:706. Johnson, P.R. and J. Hirsch. 1972. Cellularity of adipose depots in six strains of genetically obese mice. J. of Lipid Res. 13:2. 90 Johnson, P.R., J.S. Stern, M.R.C. Greenwood, L.M. Zucker and J. Hirsch. 1973. Effect of early nutrition on adipose cellulatity and pancreatic insulin release in the Zucker rat. J. of Nutr. 103:738. Kennedy, G.C. 1955. The central nervous control of food intake. Adv. of Sci. 12:123. Knittle, J.L. and J. Hirsch. 1968. Effect of early nutrition on the develOpment of rat epididymal fat pads: cel- lularity and metabolism. J. of Clin. Invest. 47:2091. Knittle, J.L. 1972a. -Meternal diet as.a factor in adipose tissue cellularity and metabolism in the young rat. J. of Nutr. 102:427. , 1972b. Obesity in childhood: 3 problem in adipose tissue cellular development. J. Ped. 181(6):1048. Kovanen, P. T., E. A. Nikkila and T. A. Miettinen. 1975. Regulation of cholesterol synthesis and storage in fat cells. J. of Lipid Res. 16:211. Lemonnier, D., J. Winand, J. Furnelle and J. Christophe. 1971. Effect of a high fat diet on obese-hyperglycemic and non-obese Bar Harbor mice. Diabetologia. 7:328. Lemonnier, D. 1971. Sex difference in the number of adipose cells from genetically obese rats. Nature. 231:50. . 1972. Effect of age, sex and site on the cellularity of the adipose tissue in mice and rats rendered obese by a high fat diet. J. of Clin. Invest. 51:2907. Lemonnier, D., J-P. Suquat, R. Aubert and G. Rosselin. 1973. Long term effect of mouse neonate food intake on adult body composition, insulin and glucose serum levels. Horm. Metab. Res. 5:223. - Lemonnier, D. and A. Alexiu. 1974. Nutritional, genetic and hormonal aspects of adipose tissue cellularity. The Regulation of Adipose Tissue Mass, ed. J. Vague and J. Boyer, American Elsevier Co., Inc., New York, p. 158. Mickelsen, 0., S. Takagashi and C. Craig. 1955. Experimental obesity. I. Production of obesity in rats by feeding high fat diets. J. Nutr. 57:541. Micklejohn, A.P. 1955. The assessment of obesity. Adv. of Sci. 12:120. Narins, D.M.C. and J. Hirsch. 1974. Supplementary feeding during the preweaning period, effect on carcass compo- sition and adipose tissue cellularity of the rat. Biol. Neonate. 25:176. 91 Parkes, A.S. 1926. Growth of young mice according to size of litter. Ann. of Applied. Biol. p. 374. Peckham, S.C., C. Entemann and H.W. Carroll. 1962. The influence of a hypercaloric diet on gross body compo- sition and adipose tissue cellularity of the rat. J. 0f Nutr. 77:187. Robertson, A.F. and G.H. Lowrey, 1964. Overweight children. ‘ Mich. Med. p. 629. Rodbell, M. 1964. Metabolism of isolated fat cells. I. Effects of hormones on glucose metabolism and lypolysis. J. of Biol. Chem. 239:375. Salans, L.B., E.S. Horton and E.A.H. Sims. 1971. Experi- mental obesity in man: cellular character of the adipose tissue. J. of Clin. Invest. 50:1005. Salans, L.B., M.J.Zarowski and R. Segal. 1972. Effect of insulin upon the cellular character of rat adipose tissue. J. of Lipid Res. 13:616. Salans, L.B., S.W. Cushman and R.E. Weismann. 1973. Adipose cell size and number in nonobese and obese patients. J. of Clin. Invest. 52:929. Schemmel R., 0. Mickelsen and J.L. Gill. 19703. Dietary obesity in rats: body weight and body fat accretion in seven strains of rats. J. of Nutr. 100:1041. Schemmel, R., 0. Mickelsen and U. Mostosky. 1970b. Influence of body weight, age, diet and sex on fat depots in rats. Anat. Rec. 166:437. Schemmel, R., 0. Mickelsen and K. Motawi. 1972. Conversion of dietary to body energy in rats as affected by strain, sex and ration. J. of Nutr. 102:1187. Schemmel, R., 0. Mickelsen and L. Fisher. 1973. Body com- position and fat depot weights of rats as influenced by ration fed dams during lactation and that fed rats after weaning. J. of Nutr. 103:477. Schemmel, R.,and 0. Mickelsen. 1974. Influence of diet, strain, age and sex on fat depot mass and body compo- sition of the nutritionally obese rat. The Regulation of Adipose Tissue Mass, ed. J. Vague and J. Boyer, American Elsevier Co., Inc., New York, p. 238. Scott, J.A. 1961. Report on the heights and weights (and other measurements) of school pupils in the County of London in 1956. London, London County Council. 92 Shukla, A., H.A. Forsyth, C.M. Anderson and S.M. Marwah. 1972. Infantile overnutrition in the first year of life: a field study in Dudley, Worcestershire. Br. Med. J. 2:507. Shier, P.D. and R. Schemmel. 1975. Effects of diet, age, strain and anatomical site on fat depot triglyceride and fatty acid content in rats. Proc. of the Soc. for Exp. Biol. and Med. l49:864. Sinclair, H.A. 1955. Obesity-introduction. Adv. of Sci. 12:115. Sjostrom, L., P. Bjorntorp and J. Vrana. 1971. Microscopic fat cell size measurements on frozen—cut adipose tissue in comparison with automatic determinations of osmium fixed fat cells. J. of Lipid Res. 12:521. Smith, U. 1971. Morphologic studies of human subcutaneous adipose tissue in vitro. Anat. Rec. 169:97. Tanner, J.M. 1955. Obesity and the classification of body build. Adv. of Sci. 12:;;6. Tanner, J.M. 1966. Standards from birth to maturity for height, weight, height velocity, and weight velocity: British children 1965. Arch. Dis. Childh. 41:613. Ten State Nutrition Survey, 1968-1970. DHEW publication No. (HSM) 72-8131. III. AnthrOpometry Dental. Dept. HEW Health Services and Mental Health Administration, Atlanta. Tepperman, J. 1958. Etologic factors in obesity. Per- spectives in Bio. and Med. 1:293. Thiel, A., R. Schemmel O. Mickelsen, J.T. Johnson and J.L. Gill. 1972. Free choice intakes by two strains of rats offered high fat and high carbohydrate rations. Nutr. Reports Intrnat. 5(2):101 Thomas, D.W. and J. Mayer. 1973. Psych. Today. 7:74. Wetzel, N.C. 1941. Physical fitness in terms of physique, development and basal metabolism. J.A.M.A. 116:1187. Widdowson, E.M. and R.A. McCance. 1960. Some effects of accelerating growth, I general somatic development. Proc. of Roy. Soc., "B". 152:188. Wilkinson, J.F, 1975. The special hazards of childhood obesity. Parents Magazine. August. p. 34. 93 Winick, M. and A. Nobel. 1966. Cellular response in rats during malnutrition at various ages. J. of Nutr. 89: 89:300. , 1967. Cellular response with increased feeding in neonatal rats. J. of Nutr. 91:179. Wolff, O.H. 1955. Obesity in childhood, a study of the birth weight, the height and the onset of puberty. 0. J. of Ped. 24:109. Zinder, 0., R. Arad and B. Shapiro. 1967. Effect of cell size on the metabolism of isolated fat cells. Israel. J. Med. Sci. 3(6):787. Zingg, W., A. Angel and M.D. Steinberg. 1962. Studies on the number and volume of fat cells in adipose tissue. Can. J. of Bioc. and Phys. 40:437. HICHIGRN STQTE UNIV. LIBRQRIES 293101725152