Influence oía Liquid Diet and Meal Pattern onBody Weight and Body Fat in Rats1

AYSEL OZELCI, DALE R. ROMSOSANDGILBERT A. LEVEILLEFood Science and Human Nutrition Department,Michigan State University, East Lansing, Michigan 48824

ABSTRACT Rats weighing 110 to 150 g or 250 g initially were utilizedin five experiments to determine the effect of the form of the diet (dryversus liquid) and the pattern of feeding (meal-feeding, force-feeding,nibbling, or ad libitum) on body weight gain and on body fat. A high-carbohydrate, 20% casein or 20r/c lactalbumin diet was fed for 4 to 8weeks. Consumption of a diet mixed with an equal weight of water increased weight gain in one of three experiments. Body fat content of therats was not influenced by addition of water to the diet. Neither force-feeding nor meal-feeding influenced body fat gain provided the respectivecontrol rats were pair-fed during the initial adaptation period. Likewise,when rats were pair force-fed to ad libitum fed rats without an initialadaptation meal frequency did not influence body fat gain. When meal-fedrats were switched to ad libitum intake their food intake increased toequal that of rats which had been continuously fed ad libitum; however,the rats which had been switched gained more body fat than did ratscontinuously fed ad libitum. These results suggest that meal frequencymay have a minimal influence on body fat accumulation, but that a shiftto a higher level of food intake may cause an increased food efficiency andgreater rate of fat deposition than in rats continuously fed the higher levelof intake. J. Nutr. 108: 1128-1136, 1978.

INDEXING KEY WORDS•meal feeding •nibbling

body fat •force-feeding •liquid diet

Obesity is a serious public health problem in Western Societies ( 1 ) and the pattern of food intake, in addition to manyother factors, has been suggested to influence obesity in humans (1-4). Studieswith rats which were pair-fed with a stomach tube twice daily to ad libitum fed control rats supported the suggestion that mealpattern may affect body fatness. Cohn etal. (5-7) observed that after force-fed ratshad been gradually adapted to ingestingtwo meals daily for 1 week they depositedmore body fat than the ad libitum fed controls. On the other hand, rats trained toeat one meal per day (meal-eaters) consumed less food than the ad libitum fedcontrols but gained body weight at the

same rate as the ad libitum fed rats (8, 9).These meal-fed rats gained as much fat asthe ad libitum fed controls while consuming less food. Both the force-fed and themeal-fed rats consumed restricted amountsof food initially, but the food intake of control rats was not restricted. When the foodintake of the control rats was restrictedwith an automated feeding machine to thelevel consumed by meal-fed rats, no differences in body fat gain were observedbetween the meal-fed and continuously

Received for publication November 28, 1977.1 Supported In part by NIH AM 15847. DRR is the

recipient of Career Development Award KO4 AM00112. Michigan Agricultural Experiment StationJournal Article No. 8335.

1128

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

MEAL PATTERN AND BODY FAT 1129

pair-fed rats (10). These results suggestedthat the increased body fat accumulationobserved in force-fed rats (5-7, 11), whichhas been attributed to their altered mealpattern, may not occur when the feedingpattern of rats is altered by other mechanisms.

The present experiments were designedto determine the effects of two differencesin experimental design between previousforce-feeding and meal-feeding experiments on body weight gain and on bodyfat gain of rats. It is necessary to add waterto the diet before force-feeding rats; however, ad libitum fed control rats have routinely consumed a dry diet. Addition ofwater to the diet has been reported byothers ( 12, 13 ) to increase growth of rats;consequently, three experiments were conducted to evaluate the influence of feeding the diet in liquid form. Secondly, thelevel of food intake, ad libitum or restricted, may alter the response to the feeding pattern ( 14 ). Thus, in one experimentrats were force-fed ad libitum levels of intake without previous adaptation to force-feeding and in another experiment meal-fed rats were offered two meals per day,rather than a single meal per day, to induce them to increase their food intake.

MATERIALS AND METHODS

Five experiments were conducted utilizing male Sprague-Dawley rats 2 weighingapproximately 110 to 150 g or 250 g initially. Rats were individually housed instainless steel cages in a temperature regulated room (22 ±2°)with lights on from0700 hours to 1900 hours. Water was available ad libitum. Table 1 presents the composition of the two purified diets utilized.Diet 2 was utilized in experiments whererats were force-fed. Lactalbumin was substituted for casein because of the greaterwater solubility of lactalbumin as compared to casein. This diet was cellulosefree and contained 35.05 parts glucose and35.05 parts corn starch to duplicate the dietused in Cohn's (5, 6) force-feeding experi

ments. Food intake ( dry weight basis ) wasrecorded daily and body weights were recorded twice a week.

Rats were killed by decapitation; epi-didymal fat pads, the liver (experiment 1

TABLE 1Composition of diets

Diet

CaseinLactalbuminBasal1CelluloseGlucoseCorn

starch920.0—9.94.066.1——20.09.9—30.0030.00

1The basal mix contained (in g/9.9 g) : methio-niiie, 0.3; vitamin mix, 0.4 (see Ref. 15); mineralmix, 4.0 (see Kef. 16) ; choline chloride, 0.2 and cornoil, 0.0.

only ) and stomach contents were removed.Carcass weight represented the weight ofthe rat minus blood loss and minus weightof the stomach contents. Carcasses werefrozen until analyzed. Body fat was determined gravimetrically after chloroform:methanol (3:2 v/v) extraction of the carcasses which had been dried and groundor homogenized with an equal amount ofwater. Duplicate samples of the warmhomogenate were taken immediately afterblending. All the data were evaluated statistically by the Student's f-test or by analy

sis of variance. Treatment differences werecompared by the least significant rangeprocedure for a posteriori tests (17).

Experiment 1. Rats weighing approximately 250 g initially, were allotted tothree treatment groups. Meal-eaters werefed for 2 hours a day (0900 hours to 1100hours) and nibblers were continuouslypair-fed with an automated feeding machine as described earlier (18). The thirdgroup was fed the same amount of foodmixed with an equal weight of water;these rats were fed at 0900 hours daily.The liquid-fed rats consumed about 50%of the diet by 1900 hours and the remaining 50^ was consumed during the night.Blood samples were taken on each of 3days during the sixth week to determineblood glucose levels. At 0800 and 1200hours rats were bled by clipping the tail.

2Obtained from Spartan Research Animals. Inc..Ilnslett. Michigan (experiments 1-3 and 5) and fromHarían Industries, Inc.. Cumberland, Indiana (experiment 4).

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

1130 OZELCl, ROMSOS AND LEVEILLE

Blood (0.1 ml) was collected into hepa-rinized capillary tubes. Protein was precipitated with Ba(OH)2 and ZnSO4 andthe samples were centrifuged. Blood glucose levels were determined in the supernatant by the glucose oxidase procedure.3

At the termination of the experiment,in vivo rates of fatty acid synthesis weremeasured in liver and adipose tissue bythe amount of tritiated water incorporatedinto fatty acids. Each rat was injected in-traperitoneally at 1200 hours with 2 mCi3H2O ìin 0.8 ml physiological saline; ratswere killed exactly 15 minutes later.Plasma samples were collected at the timeof killing to determine the plasma specificactivity. The liver and epididymal fat padswere removed and homogenized in equalvolumes of water. Aliquots of the homoge-nates were saponified in 30% ethanol-KOH. The mixture was then acidified andextracted three times with petroleum ether.Radioactivity was determined in a scintillation counter after adding scintillationcocktail (4.0 g scintillant •'dissolved in230 ml absolute ethanol and toluene to 1liter) to the extracted fatty acids. Bodywater specific activity was used as an indexto calculate nmoles of 3H incorporatedinto fatty acids per g tissue per minute(19).

Carcasses were dried to constant weightand ground for body composition analyses.Carcass moisture was calculated as thedifference between wet and dry weights.Carcass protein was calculated as nitrogenX 6.25 after nitrogen determination by thesemi-micro Kjeldhal procedure (20).

Experiment 2. This experiment was designed to investigate the effect of feedingthe diet in liquid form as compared toforce-feeding the diet. Rats (140 g) werefed diet 2 for 4 weeks. One group was fedthe diet in dry form whereas anothergroup received the diet mixed with anequal weight of water. Force-fed rats weregiven the liquid diet with a stomach tube 6twice daily. Food intake was gradually increased during the first week to preventfood shock. During the first week rats fedthe liquid and dry diets were pair-fed tothe force-fed group. Starting from the second week, rats consuming the diet in dryform were fed ad libitum and the other

two groups of rats were pair-fed to theserats for three additional weeks. Food consumption patterns were similar for rats fedliquid and dry diets. Since the food intakewas restricted during the first week, bothgroups consumed all the food in 1 to 2hours. During the second week, bothgroups consumed about 40% to 50% oftheir food during the day. After the second week the ad libitum fed and the pairliquid-fed rats consumed less than 40% oftheir food during the daylight hours.

The rats were killed at the end of 4weeks. The carcasses were blended withan equal weight of water. Aliquots (about1 g) were taken for body fat analyses.

Experiment 3. Both meal-fed and force-fed rats require an initial adaptation beforefood intake approaches ad libitum levels.Meal-eating rats consume small quantitiesof food initially. Average food consumption of meal-eating rats (one 2 hour mealper day) plateaus at about 70% to 80%of ad libitum intake after 7 to 10 days (8 ).Force-fed rats have been given smalleramounts of food during the first days oftubing to prevent food shock (5, 6). Inthis experiment, the control rats consumedfood ad libitum and the liquid fed ratswere pair-fed to examine the influence ofthe liquid diet on body weight and bodyfat in rats consuming food at levels equalto ad libitum intake throughout the experiment. Rats weighing 250 g initiallywere either fed a 20% casein, high-carbohydrate diet ad libitum or pair-fed dailythe diet mixed with an equal weight ofwater. The food consumption pattern ofof the liquid-fed group was very similarto the pattern observed in ad libitum fedrats. The experiment lasted 8 weeks. Atthe end of the experiment rats were killedby decapitation and stomach contents wereremoved. Carcasses were blended with anequal weight of water and the body fatwas determined.

Experiment 4. In previous force-feedingstudies (5, 6), food intake of rats has been

"Glucostat, Worthington Biochemical Corp., Freehold. New Jersey.

1 New England Nuclear, 575 Albany Street, Boston,Massachusetts.

5Omniflour, New England Nuclear, Pilot ChemicalsDivision, 575 Albany St., Boston, Massachusetts.

«Feeding tube, size S French, C. R. Bard, Inc.,Murray Hill, New Jerse'y.

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

MEAL PATTERN AND BODY FAT 1131

initially restricted to adapt the animals tostomach distention. It was suggested earlierthat the initial food restriction might alterthe metabolic efficiency of the force-fedrats (10). Consequently, the experimentwas designed to examine the influence offorce-feeding, without initially restrictingfood intake, on body weight and on bodyfat. Rats, weighing approximately 150 ginitially, were fed ad libitum or were pairfed a 20% lactalbumin, high-carbohydratediet with a stomach tube twice daily for4 weeks. On the first day of the experiment, force-fed rats were fed three timesto reduce the stress of stomach load. However, 20% of the force-fed rats still diedduring the first week. At the end of thefirst week, 10 ad libitum fed rats and 12force-fed rats were killed. Remaining ratsfrom both groups were killed at the end of4 weeks. Carcasses were homogenized withan equal weight of water. Carcass fat wasdetermined. Carcass energy was determined with an Adiabetic Bomb Calorimeter. Fat energy was calculated by multiplying the g of body fat by 8.5 kcal/g fat.This value was the average of several determinations of carcass fat extracts. Leanenergy was calculated as the difference between the total carcass energy and the fatenergy.

Experiment 5. In this experiment ratswere trained to eat two 1-hour meals perday for several reasons: a) this two mealper day pattern was similar to the feedingpattern utilized in previous (5, 6) force-feeding experiments; b) rats trained toconsume two 1-hour meals per day havebeen reported to consume more food thanrats restricted to one 2-hour meal per day(21) and c) by switching rats trained tomeal-eat from meal-eating to nibbling itwas possible to compare the two feedingpatterns without a reduction in food intake.The feeding pattern of nibblers was controlled with an automated feeding machine(18).

Five groups of rats were used. Onegroup was fed ad libitum for 5 weeks andthen killed. Four other groups of rats werefed two 1-hour meals per day (from 0900hours to 1000 hours and again from 1600hours to 1700 hours ) for the first 2 weeks;one group was killed at this time, a second

group continued to meal-eat for an additional 3 weeks, a third group was fed thesame amount of food eaten by the meal-eaters with an automated feeding machine(nibblers) for the last 3 weeks and the remaining group was fed ad libitum for thelast 3 weeks of the 5 week study. Diet 1was fed.

Body fat was determined at the end of2 and 5 weeks. Ad libitum fed rats werekilled at 5 weeks, their body fat content at2 weeks was estimated. The estimate wasbased on the body fat content observed inrats in our laboratory of similar age andweight fed the same diet ad libitum. Fatgain of the ad libitum fed rats was calculated as the difference between the finalbody fat and the calculated body fat at 2weeks.

RESULTSExperiment 1. Meal-fed, nibbling and

liquid-fed rats consumed an average of14.7 g food per day (table 2). Meal-eatersgained about 20 g less weight than eithernibblers or liquid-fed rats; however, thedifferences were not significant. Carcassweight and carcass composition, at the endof the 8 week experiment, were also similar in the three groups. Carcass proteindid not differ among the meal-fed, nibblingor liquid-fed rats. Carcass fat content wasslightly, but not significantly, less in meal-eaters than in either nibblers or liquid-fedrats. All three groups had essentially thesame carcass moisture content.

Blood glucose levels at 0800 hours weresimilar in the three groups of rats (table2). From 0800 hours to 1200 hours, bloodglucose levels increased in meal-eaters, inagreement with previous studies (22); thecirculating level of glucose also increasedin liquid-fed rats but not in nibblers. Consequently, at 1200 hours meal-fed rats hadsignificantly higher blood glucose levelsthan the nibblers, and the blood glucoselevels of the liquid-fed rats were intermediate.

The rates of fatty acid synthesis in liver,as measured by the incorporation of triti-ated water into fatty acids, were similaramong the three groups. In the adiposetissue (epididymal pad) however, meal-eaters incorporated 3H2O into fatty acids at

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

1132 OZELCI, ROMSOS AND LEVEILLE

TABLE 2Food intake, weight gain, carcass composition, blood glucose level and fatty acid synthesis in meal-eating

(one 2-hour meal daily), nibbling and liquid-fed rats (experiment I)1

Parameter Meal-eater Nibbler Liquid-fed

Food intake(g/day)Weightgain(g)Carcassweight(g)Carcassprotein(g)Carcassfat(g)Carcassmoisture(g)Blood

glucose (mg/100ml)0800hours1200

hoursFattyacidsynthesis'LiverEpididymal

fat pad14.7±

0.6°77±9°325±11«53.

1±2.8°45.3±2.1«213.9±6.8°79

db3°109±6«-2709

±70°1453±178"14.7±

0.6°97±10°343±11«50.9±2.5°53.7±4.7°212.9±

8.3°80

±3°86±2*661

±127°756±209*14.7±

0.6°93±8«340±9°51.9±

3.0«52.7±1.9°222.8

±7.3«77

±2°99±3°.».*625

±58°1017±199«-*

1Values represent mean±snMfor eight rats, weighing 253 ±2 g initially. The rats were fed for 8 weeks.Treatment means not sharing a common superscript letter differ significantly (P < 0.05). ìSignificantdifference (P < 0.05) between samples obtained at 0800 hours versus those obtained at 1200 hours.3nmoles 3HjO incorporated into fatty acids/g tissue/minute. Liver weights averaged 11.8, 12.6, and 12.1 gand epididymal fat pad weights averaged 5.4, 5.4, and 6.9 g for meal-eaters, nibblers, and liquid-fed rats,respectively. Tissue weights were not influenced (P > 0.05) by dietary treatment.

twice the rate observed in the nibblers.Fatty acid synthesis in adipose tissue of therats consuming the diet in liquid form occurred at an intermediate rate (table 2).These results confirm previous observations(8) that rates of fatty acid synthesis areelevated to a greater extent in epididymaladipose tissue than in liver of meal-fedrats, even when the control nibblers arepair-fed. This greater rate of fatty acidsynthesis in the epididymal adipose tissueshortly after consumption of the high-carbohydrate meal did not result in increased body fat.

Experiment 2. Food intake, weight gain,carcass weight, and carcass fat of the rats

TABLE 3Food intake, weight gain and carcass fat of force-fed,

pair-fed and ad libitum rats (experiment %)1

ParameterForm

ofdietFoodintake(g/day)Weight

gain(g)Carcassweight(g)Carcassfat (g)Force-fed(0-4)'Liquid20.8

±0.7°149±4«278

±5«51.1±1.1"Pair-fed(0-4)Liquid18.2

±1.9°186±3'315±4'57.5

±2.8°Pair-fed

(0-1)andadlibitum(2-4)Dry20.9

±0.6°164±3'295

±5"51.1±3.2°

1Values represent mean±sEU for 10 rats weighing 140±1g initially and fed for 4 weeks. Treatment means on the same linenot sharing a common superscript letter differ significantly(P < 0.05). ' Numbers in parentheses represent the weeksof the experiment.

are shown in table 3. All three groups consumed essentially the same amount of food.Force-fed rats gained less body weightthan either of the other two groups; likewise, the carcasses of the force-fed groupweighed less than carcasses of the othertwo groups. However, there were no differences in the amount of carcass fatamong the three groups of rats.

Experiment 3. Table 4 presents the foodintake, body weight gain, carcass weightand fat of rats fed ad libitum or pair-fedthe same diet in liquid form. Both groupsconsumed similar amounts of food duringthe 8-week experiment. Neither weightgain nor carcass weight differed betweenthe ad libitum and liquid-fed rats. Carcassfat content was slightly greater in the

TABLE 4Food intake, weight gain, carcass weight and carcass

fat of rats fed ad libitum or pair-fed a liquid diet(experiment 3)1

Parameter Ad Libitum Liquid-fed

Food intake (g/day)Weight gain (g)Carcass weight (g)Carcass fat (g)22.6±

0.7217 ±12453 ±1166.2±4.322.4

±0.5206 ±10442 ±871.4±5.2

1Values represent mean±SEMfor eight ratsweighing 250±3g initially and fed for 8 weeks.Treatment effects were not significantly different(P > 0.05).

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

MEAL PATTERN AND BODY FAT 1133

TABLE 5Food intake, weight gain, carcass weight and carcass energy of rats fed ad libitum

or pair force-fed for 4 weeks (experiment 4)-1

Ad libitum fed Pair force-fed

Parameter End of week 1 Week 2 to week 42 End of week 1 Week 2 to week 42

Number of ratspertreatmentTotal

food intake(g)Weightgain(g)Carcassweight(g)Energy(kcal/rat) :

TotalcarcassFat4Lean51095

±146±4180

±3250

±8102±10148

±1010431±

8133±6134

±5380

±31259±44121±271293

±031±P168

±23255±5117±7138±420433

±0137±2137±

3436

±32341±2695

±24

1Values represent mean±sKMfor rats weighing 146±2g initially and fed for 4 weeks. Rats were killedat the end of 1 and 4 weeks. 2Food intake and body weight gain expressed as totals for the 3-week period,whereas carcass weight and carcass energy were expressed as increases during the 3-week period. Increase incarcass weight and energy during the 3-week period calculated by subtracting the carcass weight and energyin rats killed after 1 week from values obtained at 4 weeks. 3Significantly different (P < 0.05) from thead libitum fed group at the end of the first week. 48.5 kcal/g fat X g body fat. »Total carcass energyminus carcass fat energy.

liquid-fed rats but the difference was notsignificant. Based on the results of the firstthree experiments, we concluded that feeding the diet in liquid form did not contribute to the greater body fat content previously observed in force-fed rats (5, 6, 11).

Experiment 4. Rats were killed after 1and 4 weeks. During the first week, bothad libitum and force-fed rats consumed anaverage of about 16 g food per day (table5). Ad libitum fed rats had gained morebody weight than the force fed rats at theend of the first week. Carcass weight at theend of 1 week, was also greater for the ad

libitum fed rats than for the force-fedgroup. Total carcass energy did not differbetween the two groups at the end of 1week.

At the end of the experiment, bothgroups had consumed about 520 g foodper rat ( approximately 19 g per day ). Thebody weight gained in the last 3 weeks wassimilar for ad libitum and force-fed rats.Changes in carcass weight were essentiallythe same for both groups. Force-fed ratsgained slightly, but not significantly, moretotal carcass energy during the last 3 weeksthan did the ad libitum fed rats. The

TABLE 6Food intake, weight gain, carcass weight, and carcass fat of rats (experiment 5)1

ParameterNumber

ofratsFoodintake(g/dayjWeight

gain(g)Carcasawcigbt(g)Carcassfat(g)Fat

gain (g)Meal

fed(0-2)!1012.1±1.350

±3156±315.5±1.5—Ad-libitum(0-2)617.3

±0.593±4—25.0«—Meal-fed

(0-2)andMeal-fed

(3-5)«1018.9

±0.7°118±4«272

±4«39.1±2.2«23.6

±2.2«Meal-fed

(0-2)andNibbler

(3-5)«1018.8

±0.7«125±2«278±3«42.6±2.9«27.1

±2.9«Meal-fed

(0-2)andAd-lib

(3-5)'1023.7

±0.3'164±6'314±5*57.2±3.8'41.7±3.8»Ad-lib

(0-2)andAd-lib

(3-5)»623.4

±0.5*122±8«314

±8»52.9±3.9«.*27.9

±3.9«

1Values represent mean ±SEMfor rats weighing 114±0.4g initially and fed for 5 weeks. Meal-eater = two 1-hour meals per day.Treatment means from weeks 3 to 5 not sharing a common superscript letter differ significantly at P < 0.05. ' Numbers in parentheses represent weeks of the experiment. «Food intake and weight gain values in these columns represent data for weeks 3 to5 of the experiment. Carcass weight and carcass fat was determined at the end of the 5 week experiment. Fat gain was determinedby subtracting the estimated body fat at 2 weeks from the values obtained at 5 weeks. <Estimated from values of fat content ofad libitum fed rats of similar weights in previous experiments.

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

1134 OZELCI, ROMSOS AND LEVEILLE

energy gain contributed by fat was notgreater in the force-fed rats than in the adlibitum fed rats. Thus, we conclude thatforce-feeding has only a minimal influenceon body fat when the force-fed rats arepair-fed throughout the entire experiment.

Experiment 5. During the first 2-weekperiod of meal-feeding (two 1-hour mealsper day), rats consumed about 12 g offood in contrast to consumption of about17 g food by the ad libitum fed rats ( table6). The ad libitum fed group gained nearlytwice as much weight compared to themeal-fed rats during this period.

During the last 3 weeks ad libitum fedrats consumed about 25% more food thandid the meal-fed rats or the pair-fed nib-biers. Body weight gain was similar formeal-fed, nibbling and ad libitum fed rats;however, rats which had been switchedfrom meal-feeding to ad libitum feedinggained more weight than the other threegroups during the last 3 weeks of the experiment. Carcasses of meal-fed and nibbling rats killed at the end of the 5-weekexperiment weighed significantly less thandid carcasses of the other two groups.

At the end of the 5-week experiment,there was no difference in the carcass fatcontent between meal-eating and nibblingrats. Thus, provided rats are pair-fedthroughout the experiment, meal frequencymay have only a minimal influence onbody fat. Both the ad libitum fed groupsof rats had significantly greater body fatthan observed in the meal-fed and nibblergroups. Fat gain (weeks 3 to 5) wasgreater in rats switched from meal-eatingto ad libitum intake than in any of theother groups. There was no difference infat accumulation among meal-fed, nibbling and continuously ad libitum fed rats(table 6). These results suggest that ratsswitched from a restricted food intake toad libitum eating were more efficient inconverting food energy to body fat thanrats fed ad libitum throughout the experimental period.

DISCUSSION

Addition of water to a dry diet has beensuggested to have a growth promotingeffect (12, 13). It is possible that additionof water to the diet of force-fed rats may

have contributed to the increased body fatof these rats relative to rats fed the dry dietad libitum (5, 6, 11). Consequently, in thefirst three experiments of this study we investigated the effects of feeding the diet inliquid form as compared to meal-feedingthe dry diet, force-feeding the liquid dietor feeding the dry diet ad libitum. Meal-feeders require an initial adaptation tomeal-feeding and food intake reaches onlyabout 80% of that expected for ad libitumfed rats (8). Force-fed rats in experiment2 were also gradually adapted to the feeding pattern during the first week but during the remainder of the experiment theywere pair-fed to the ad libitum fed group.In all three experiments the liquid diet waspair-fed; consequently, several of the comparisons involved restricted food intakeand others involved ad libitum intake.

When the rats were fed the diet in liquidform, the weight gain was not differentfrom that of meal-fed or nibbling rats(expt. 1 ). Also, pair-feeding rats the liquiddiet with a group of rats which had beenfed ad libitum throughout the experimentdid not change the weight gain. However,rats allowed to consume the diet in liquidform gained more weight than the force-fed rats or the rats which had been restricted for the first week and then fed adlibitum for 3 weeks (experiment 2). Thus,it appears that feeding a high-carbohydratediet in liquid form may increase bodyweight gain in some, but not all, situations.

Consumption of the liquid diet did notincrease body fat content of the rats. Carcass protein and moisture content of theliquid-fed rats were also similar to themeal-fed or nibbling counterparts. It appears very unlikely that the addition ofwater to the diet contributed to the increased body fat previously observed inforce-fed rats (5, 6, 11). In agreementwith our previous report (10), meal pattern did not influence body fat when pair-fed meal-eaters and nibblers were compared. Similarly, force-feeding did not increase body fat when control rats werepair-fed during the first week.

Since feeding the diet in liquid form didnot result in greater fat content in theserats, we designed two more experiments toinvestigate the influence of meal pattern

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

MEAL PATTERN AND BODY FAT 1135

on body composition of rats. In these experiments treatment and control rats werepair-fed and attempts were made to havethe rats consume food at or near ad libitumintake throughout the experiment.

Rats were pair force-fed to ad libitumfed controls without an initial adaptation.In agreement with results obtained whenboth force-fed and control rats were fedrestricted amounts of food during the firstweek (experiment 2), force-feeding didnot influence body fat content. The designof these experiments differs from previousreports (5-7) in one aspect; Cohn and co-workers gave force-fed rats limited amountsof food during the first week whereas control rats consumed food ad libitum duringthis period. Our present results suggestthat feeding pattern may have only a minimal influence on body fat accumulation inrats pair force-fed during the entire experiment.

In the last experiment rats were fed two1-hour meals for 2 weeks; at this time therats were consuming 20% less food thanad libitum fed controls. One group of ratswas then switched from meal-eating toautomated feeders which continuously delivered food to the cage. This switch inmeal pattern did not result in an initialdecrease in food intake as would have beenobserved if continuously-fed rats had beenswitched to meal-eating. Meal-fed rats didnot accumulate more body fat than thepair-fed nibblers. This observation is inagreement with our previous results utilizing meal-fed rats and pair-fed nibblers(10) as well as with the results of thepresent study with force-fed rats. Underthe conditions of these experiments inwhich rats were pair-fed throughout theentire experiment, including the initialadaptation period, meal pattern did notinfluence body fat.

When previously trained meal-eaterswere allowed to consume food ad libitumtheir food intake equalled that of ratswhich had been fed ad libitum throughoutthe experiment. Under these conditionsrats switched from meal-eating to ad libitum food intake gained more body fatthan ad libitum fed rats, even though theydid not consume more food than the adlibitum fed rats. These results are in agree

ment with a previous report (23) and suggest that a shift from one level of food intake to a higher level might cause a metabolic adaptation and increased food efficiency in rats. It remains to be establishedto what extent the initial food restrictionutilized to adapt rats to force-feeding contributed to the increased body fat reportedin force-fed rats.

LITERATURE CITED

1. Fabry, P. (1969) Feeding frequency andnutritional adaptation—general conclusionsand perspectives. In: Feeding Patterns andNutritional Adaptations. (Fabry, P., ed.),pp. 147-152, Publishing House of Czechoslovakia Academy of Sciences, Prague.

2. Fabry, P. & Tepperman, J. (1970) Mealfrequency—A possible factor in human pathology. Am. J. Clin. Nutr. 23, 1059-1068.

3. Fabry, P., Fodor, J., Hejl, Z., Braun, T. &Zvolankova, K. (1964) Frequency of meals:Its relation to overweight, hypercholesterol-emia, and decreased glucose tolerance. Lancet2, 614-615.

4. Gordon, E. J., Goldberg, M. & Chassy, G. J.( 1963) New concept in treatment of obesity.J. Am. Med. Assoc. 186, 50-60.

5. Cohn, C. & Joseph, D. (1959) Changes inbody composition attendant on force-feeding.Am. J. Physiol. 196, 965-968.

6. Cohn, C. & Joseph, D. (1963) Feedingfrequency and protein metabolism. Am. J.Physiol. 205, 71-78.

7. Cohn, C., Shrago, E. & Joseph, D. (1955)The effect of food administration on weightgains and body composition of normal andadrenalectomized rats. Am. J. Physiol. 180,503-507.

8. Leveille, G. A. (1970) Adipose tissuemetabolism: Influence of periodicity of eating and diet composition. Federation Proc.29, 1294-1301.

9. Leveille, G. A. (1975) Effect of periodicityof eating and diet composition on enzymaticadaptations. Nutr. Proc. 9th Int. Cong. Nutr.Mexico, 1972, 1, 13-21.

10. Ozelci, A., Romsos, D. R. & Leveille, G. A.(1977) Influence of diet composition onnitrogen balance and body composition inmeal-eating and nibbling rats. J. Nutr. 107,1768-1774.

11. Han, K. (1973) Effect of frequency ofmeal on the growth rate, nutrient digestibility,body composition, nitrogen retention andheat production of rats. Nutr. Rep. Int. 7,9-18.

12. Keane, K. W., Smutko, C. J., Krieger, C. H.& Dentón, A. E. (1962) The addition ofwater to purified diets and its effect upongrowth and protein efficiency ratio in therat. J. Nutr. 67, 18-22.

13. Hopkins, D. T. & Steinke, F. H. (1976)Effect of water of hydration on the measure-

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from

1136 OZELCI, ROMSOS AND LEVEILLE

ments of the protein eflBciency ratio of caseinand soybean protein with rats. J. Nutr. 106,1438-1446.

14. Cohn, C. & Joseph, D. (1968) Caloric intake, weight loss and changes in body composition of rats as influenced by feedingfrequency. J. Nutr. 96, 94-100.

15. Yeh, Y. Y. & Leveille, G. A. (1969) Effectof dietary protein on hepatic lipogenesis inthe growing chick. J. Nutr. 98, 356-366.

16. Leveiue, G. A. & O'Hea, E. K. (1967)

Influence of periodicity of eating on energymetabolism in the rat. J. Nutr. 93, 541-545.

17. Sokal, R. R. & Rohlf, F. J. (1969) Biometry.W. H. Freeman & Company, San Francisco,pp. 237-238.

18. Romsos, D. R. & Leveille, G. A. (1974)Effect of meal frequency and diet composition on glucose tolerance in the rat. J. Nutr.104, 1503-1512.

19. Jungas, R. L. (1968) Fatty acid synthesis

in adipose tissue incubated in tritiated water.Biochem. 7, 3708-3716.

20. Horwitz, W. (1960) Official Methods ofAnalysis of the Association of Official Agricultural Chemists, pp. 643-644.

21. Muiruri K. L. & Leveille, G. A. (1970)Metabolic adaptations in meal-fed rats: Effects of increased meal-frequency or ad libitum feeding in rats adapted to a single dailymeal. J. Nutr. 100, 450-459.

22. Gasquet, P. de & Pequignot, E. (1973)Changes in adipose tissue and heart lipo-protein lipase activities and in serum glucose,insulin and corticosterone concentrations inrats adapted to a daily meal. Horm. Metab.Res. 5, 440-443.

23. DeBont, A. J., Romsos, D. R., Tsai, A. C.,Waterman, R. A. & Leveille, G. A. (1975)Influence of alterations in meal frequency onlipogenesis and body fat content in the rat.Proc. Soc. Exp. Biol. Med. 149, 849-854.

by guest on January 4, 2012jn.nutrition.org

Dow

nloaded from