CONTENTS

No. 1 JANUARY 1956

James M. Orten. Biography of Eugen Baumann (1846 -1896) with frontis­ piece ...... 3

W. A. H ardison, R. W. E ngel, W. N. L inkous, H. C. Sweeney and.G. C Graf. Eecal chromic oxide concentration in 12 dairy cows as related to time and frequency of administration and to feeding schedule. One f-gure 11

E lwood F. E eber, C. C. Morrill, Horace W. N orton and Harry E. Rhoades. Studies of the effects of ehlortetraeycline, vitamin E and vita­ min K in the nutrition of the rat. I. Growth. One figure...... 19

W erner J. M ueller. Feasibility of the chromic oxide and the lignin indi­ cator methods for metabolism experiments with chickens...... 29

W. J. M ueller, R. V. B oucher and E. W. Callenbach. Influence of age and sex on the utilization of proximate nutrients and energy by chiekens 37

Edwin T. Mertz and J ulius Golubow. Effect of optical antipodes of 6- hydroxy norleucine (hexahomoserine) on growth and hematopoiesis in rats ...... 51

R uth M. L everton, M ary R. Gram, Marilyn Chaloupka, E ileen B rod­ ovsky and A my M itchell. The quantitative amino acid requirements of young women. I. Threonine. One figure ...... 59

R uth M. L everton, Mary R. Gram, E ileen B rodovsky, Marilyn Cha loupka, A my Mitchell and N orma J ohnson. The quantitative amino acid requirements of young women. II. Valine. One figure ...... 83

A rthur H. Smith, M ax K leiber, A rthur L. B lack and Glen P. L ofgreen. Transfer of phosphate in the digestive tract. III. Dairy cattle...... 95

A nthony M. A mbrose and Dorothy J. R obbins. Studies on comparative absorption and digestibility of acetoglycerides ...... 113

Marjorie M. N elson, Catherine D. C. BaIrd, H oward V. W right and Herbert M E vans Multiple congenital abnormalities in the rat re­ sulting from riboflavin deficiency induced by the antimetabolite galaeto- flavin ...... 125

L uther R. R ichardson and R uth B rock. Studies of reproduction in rats using large doses of vitamin B,, and highly purified soybean proteins . . 135

iii CONTENTS

No. 2 FEBRUARY 1956

Donald G. W alker and Zolton T. W irtschafter. Resorption of embryos in rats on Lathyrus odoratus diet. Seventeen figures ...... 147

Donald G. W alker and Zolton T. W irtschafter. Estrogenic inhibition of fetal resorption in lathyrism. Sixteen figures ...... 161

D. M..H egsted, J. J. V itale and Helen M cGrath. The effect of low temperature and dietary calcium upon magnesium requirement. Four ■figures ...... 175

Leslie P. Drtden, J oseeu B. F oley, Paul F. Gleis and ARTHUt M Hart­ man. Experiments on the comparative nutritive value of Cutter and vegetable fats. One figure ...... 189

Helen L. Mayfield and Richard R. R oehm. The influence of ascorbic acid and the souree of the B vitamins on the utilization of carotene. One figure ...... 203

Ruth M. Leverton, N orma J ohnson, Jean Pazur and Joan Ellison. The quantitative amino acid requirements of young women. III. Tryptophan. One figure ...... 219

W. T. W ong and B. S. Schweigert. Role of vitamin B^ in nucleic acid metabolism. I. Hemoglobin and liver nucleic acid levels in the rat .... 231

E dward J. T hacker. The dietary fat level in the nutrition of the rabbit . . 243

M argaret L. Ross and Ruth L. P ike. The relationship of vitamin B, to protein metabolism during pregnancy in the rat ...... 251

Herbert R. Heinicke, A lfred E. Harper and G. A. Elvehjem. Protein and amino acid requirements of the guinea pig. II. Effect of age, po­ tassium and magnesium and type of protein. One figure ...... 269

Dorothy A. E hmke, Bessie L. Davey and E. N eige Todhunter. The in­ fluence of age and diet on ascorbic acid metabolism in rats ...... 281

Ellen H. Morse, Martha P otc-ieter and Georgianna R. W alkeb. Ascorbic acid utilization by women. Response of blood serum and white blood cells to increasing levels of intake. One figure ...... 291

No. 3 MARCH 1956

J. A. Goyco. A study of the relation between the liver protein regeneration capacity and the hepatic neerogenie activity of yeast proteins ...... 299

A. J. M cGinnis, R. W. Newburgh and V ernon H. Chsldelin. Nutritional studies on the blowfly, Phormia regina (Meig ). One figure ...... 309

W infrey W ynn, J ohn Haldi, K atherine Dickey Bentley and Mary L ouise L aw . Dental caries in the albino rat in relation to the chemical composition of the teeth and of the diet. II. Variations in the Ca/P ratio of the diet induced by changing the phosphorus content...... 325

Leslie P. Dryden, Paul F. Gleis and A rthur M. Hartman. Reproduction and lactation of rats fed glyceryl trilaurate-containing diets ...... 335 CONTENTS

R uth M. L everton, N orma J ohnson, J oan Ellison, Dovina GesLhwender and Florence Schmidt. The quantitative amino aeid requirement? of young women. IV. Phenylalanine, with and without tyrosine. Two figures ...... 341

Ruth M. Leverton, J oan Ellison, N orma J ohnson, J ean P azur, Florence Schmidt and Donna Geschwender. The quantitative amino acid re­ quirements of young women. V. Leucine. One figure ...... 355

Otto A. B essey M. K. Horwitt and Ruth H. Love. Dietary deprivation of rifc flavin and blood riboflavin levels in man. One figure...... 367

Leon Swell, T. A. B oiter, Henry F ield, Jr and C. R. Treadwell. The absorption or plant sterols and their effect on serum and liver sterol levels 385

J. P. F easter, Sam L. Hansard, J. C. Outler and G. K. Davis. Placental transfer of cahium45 in the rat ...... 399

J ames H. Meyer. Influence of dietary fiber on metabolic and endogenous nitrogen excretion. Two figures ...... 407

C. F. Gessert and P. H. P hillips. Protein in the nutrition of the growing dog ...... 418

C. F. Gessert and P. H. P hillips. Adverse effects of some amino acid sup plements in low-protein diets for growing dogs ...... 423

No. 4 APRIL 1956

J ordan G. Lee, Harold P. Dupuy and Herman E. Rolfs. Dietary protein and the development of rat lathyrism ...... 433

George S. Sharp, Sven L assen, S. Shankman, A. F. Gebhart, J r. and J ohn W. Hazlet. Studies of protein absorption using nitrogen15 as a tag ...... 443

Hans F isher, R. C. Salander and M. W ight Taylor. Growth and creatine biosynthesis in the chick as affected by the amino acid deficiencies of casein. Three figures ...... 45

Ruth Okey and Marian M. L yman. Protein intake and liver cholesterol: Effects of age and growth of the test animal. One figure...... 471

Helen L. Mayfield and Richard R. R oehm. Carotene utilization as influ­ enced by the addition of vitamin Bu to diets containing yeast or a syn­ thetic vitamin mixture ...... 483

Marian E. Swendseid, I ona W illiams and Max S. 'Dunn. Amino acid re­ quirements of young women based on nitrogen balance data. I. The sulfur-containing amino acids ...... 495

Marian E. Swendseid and Max S. Dunn. Amino acid requirements of young women based on nitrogen balance data. II. Studies on isoleucine and on minimum amounts of the eight essential amino acids fed simultaneously 507

F umito Taketa, Marguerite A Constant, Henry S. P erdue and P aul H. P hillips. Maternal liet and resistance to dental caries in the cotton rat 519 VI CONTENTS

Herbert P. Sarett and uawrence P. Snipper with the technical assist­ ance oe D. L. Schneider. Effect of adding' carbohydrate to milk diets I. Growth and body composition...... 529

Herbert P. Sarett and L awrence P. Snipper with the technical assist­ ance op d. l. Schneider. Effect of adding carbohydrate to milk diets. II Water restriction. One figure ...... 543

Taketami Saktjragi and Fred A. K itaImerow. The biological utilization of various rat-soluble esters of pyridoxine and 4-desoxypyridoxine by rats . 557

Herbert C. T idwell. Effect of choline, methionine and ethionine on fat ab­ sorption ...... 569 EUGEN BAUMANN (1846-1896) £ . & EUGEN BAUMANN

(December 12, 1846 - November 3, 1896)

The reinankable development of physiological chemistry and its related science, nutrition, during the period from-1870 to 1895 was due to a considerable extent to the exceedingly fruitful activity of two men, Felix Hoppe-Seyler and Eugen Baumann. The work of these two scientists, the teacner and his student;, was closely interrelated during this period of 25 years, according to A. Kossel, one of Baumann’s colleagues in Strassburg. Hoppe-Seyler was trained in the more medical and biological sciences and Baumann in the field of structural chemistry and its pharmaceutical applications. In addition to an extraordinary intelligence, Baumann possessed a deal insight into the practical situations of life, a hearty humor, unusual kindness, simple likes, and an attractive personality which drew many friends and students to him. Eugen Baumann was born December 12, 1846 in Cannstatt (Cannstadt), Germany, the second son of an apothecary, J. Baumann. He had 7 brothers and sisters. Eugen attended primary school in Cannstatt and preparatory school in Stutt­ gart. He then became an apprentice in his father’s pharmacy in Cannstatt. During this time he went almost daily to the technischen Hochschule in Stuttgart to hear the lectures of Hermann v. Fehling and to work in his laboratory. His intel­ ligence and industry soon attracted Fehling’s attention and a close friendship developed between the two, which endured until Fehling’s death. In 1867, Baumann went to Lübeck for further apprentice­ ship training in an apothecary shop, and after a year to Gothenburg where 100 years earlier the great chemist Scheele had started his career as an apothecary. Two years later, fortunately, a pharmacist friend persuaded Baumann to go

3 Copyright 1956 The Wistar Institute of Anatomy and Biology All rights reserved 4 EUGEN BAUMANN to the University of Tübingen to prepare for the apothecary, examination. During this time he attracted the attention of Felix Hoppe-Seyler, who was the Professor of Applied Chem­ istry and Examiner in Toxicology. Impressed by the unusual knowledge and ability of the young candidate, he offered Baumann an assistantship in his laboratory. Hoppe-Seyler’s Institute was located in the Tübingen Castle, in a room which was formerly the duke ’s kitchen. The fire-place, large enough to roast an entire ox on a spit, was now used for a chemical hood! A large number of enthusiastic young students had as­ sembled in the improvised castle laboratory. Baumann soon became absorbed in his work with Hoppe-Seyler ancx received his doctoral degree in 1872. His doctoral dissertation was titled, “ lieber einge Vinylverbindungen.” Shortly after receiving his degree, Baumann transferred with Hoppe-Seyler to the University of Strassburg, in Alsace- Lorraine which had just been acquired by Germany following the Franco-German war of 1870. With characteristic zeal and enthusiasm, the two re-established the famous chemical lab­ oratory at Strassburg as an outstanding training center. Four years of fruitful research followed, climaxed by Baumann’s discovery of “ ethereal sulfates” in the urine. Baumann also began many lifelong friendships at Strassburg, including that with J. von Mering. In 1876, the medical faculty at Strass­ burg conferred on Baumann the honor, medicinischen Doc- tonviirde honoris causa The same year, 1876, Baumann was appointed director of the chemical division of the newly-constructed Physio­ logical Institute of du Bois-Beymond in Berlin. Here Bau­ mann again found conditions favorable for vigorous research and development and his laboratory attracted many young scientists who later became well known. These included A. Baginsky, Brieger, G. Hopp^-Seyler, Ledderhose, Preusse, Rohmann, and Schiffer. The first 8 volumes of Hoppe- Seyler’s Zeitschrift attest to tire enthusiasm and activity of Baumann’s group in Berlin. Baumann soon received the title BIOGRAPHY 5

Professor and in March 1882 he was appointed Eztraorcli- narius on the medical faculty. After 7 years in Berlin, Baumann accepted a call as Pro fessor of Chemistry on the medical faculty of the University of Freiburg. The same year, 1883, he married Therese Kopp, the daughter of the celebrated Heidelberg chemist. Five children were born of this marriage and Baumann’s home and family were a source of continued happiness to him. As in Berlin, Baumann’s laboratory soon became filled with students and vigorous research and teaching activities developed. Baumann’s research investigations at Freiburg were culminated in the discovery of iodine in the thyroid gland. The transfer to Freiburg also brought Baumann close to his teacher, Hoppe-Seyler, and re-established the active collaboration between the two, which continued until Hoppe- Seyler’s death in 1895. The University of Strassburg tried in vain to obtain Baumann as a successor to his teacher. How­ ever, the University of Freiburg was likewise not to retain the services of its distinguished and beloved teacher much longer. On October 30,1896, he became ill with an unsuspected heart ailment while he was teaching, and succumbed on No­ vember 3, of a coronai’y stricture, a month and 9 days before his 50th birthday, to the deepest sorrow of his family, his friends, and his many students. Baumann’s scientific work covered a wide range of sub jects. His complete bibliography, numbering some 106 pa­ pers, is given in a memorial article written by Kossel in Hoppe-Seylet’s Zeitschrift, 23, 1 (1897). His first work after completing his doctoral degree with Hoppe-Seyler in 1872 was concerned with the chemistry and preparation of cyan- amid, sarcosin, and methylhydantoin acid. In 1876 he began a series of studies on ethereal sulfates in urine which later proved to be of great fundamental importance. Hoppe-Seyler had previously shown that phenol appears in the urine in an altered form, “ phenol substanz,” which could be hydrolyzed by boiling with acid. Baumann demonstrated that one hydro­ lytic product was sulfuric acid and that it always appeared in 5 BUGEN BAUMANN a definite quantity. Subsequently, he prepared the substance in pure form and determined its structure to be that of an othereal sulfate of phenol. His subsequent studies demon strated the fact that other aromatic substances, including cresol pyrocatechol, resorcinol, hydfoquinone, and indole like­ wise were excreted in the urine as conjugated ethereal sul­ fates. Baumann postulated, as a result of studies of the action of certain bacteria on proteins, that the foregomg aromatic substances originated from protein putrefaction in the in­ testine, particularly of tyrosine and tryptophan. Indole derived from the latter, he believed, was converted to in- doxyl in the body, then conjugated with sulfate, and excreted in the urine as “ indican.” This work covering a 4-year period was summarized in the Zeitschrift fur pliysiologische Chemie, 10, 123 (1886). The work of Baumann and his students in Berlin thus clearly established the importance of ethereal sulfates in a detoxication mechanism for certain aromatic compounds in the animal organism. It is interesting to note that as a result of Baumann’s work, Glauber’s salt (sodium sulfate) was erroneously suggested at that time as an antidote for carbolic acid poisoning! Baumann’s interest then turned to a related topic, namely, the appearance of bromophenylmercapturic acid in the urine following the administration of brombenzene. Cysteine, de­ rived from cystine, was shown to be used in the formation of bromphenylmercapturic acid and thus the importance of cystine as a detoxifying agent was demonstrated. Baumann and his students also discovered two diamines, cadaverine and putrescine, in the urine of a case of cystinuria and spec­ ulated as to their possible origin from bacterial action in the intestine. This discovery has received some support from recent studies on cystinuria, showing the presence of relatively large amounts of lysine and arginine in the urine of cystinurics. The availability of a case of alcaptonuria led Baumann’s group to investigate this metabolic anomaly. The group was able to prepare pure homogentisic acid from the urine of BIOGRAPHY 7 this patient and to determine its structure. "Baumann be­ lieved that this substance was derived from tyrosine, possibly as a result of the action of special bacteria in the intestine It was noted that the darkening of the urine of the alcap- tonuric on standing was due to the change (oxidation) of homogentisic acid. He also devised a method for estimating homogentisic acid. Another series of studies, which proved to be of great practical importance in therapy, began with an investigation of the reaction of various organic sulfur compounds with certain aldehydes and ketones. The reaction of mercaptans with pyruvic acid led to Baumann’s discovery of the widely- used hypnotics, “ sulfonal, ” “ trional,” and “ tetranal. ” An­ other important pharmaceutical substance, “ phenacetin was developed in his laboratory. The foregoing major investigations were typical of Bau­ mann’s research endeavors, an initial observation serving as the starting point for a series of brilliant studies in a logical sequence. Important as his other contributions were, Baumann is, of course, best known for the discovery of iodine in the thyroid gland. His interest in this field was stimulated by his colleague, Kraske, and, in turn, by the first report in 1894 by Emminghaus and Beinhold (Munch, med. Wochn- schr., Nr. 31, 1894) that the administration of thyroid gland was effective in the treatment of goiter in man. In 1 ecping with his life-long interest in medical-chemical problems and perhaps also stimulated by the invaluable contributions of his first chief and teacher, Hoppe-Seyler, on the chemical constitution of hemoglobin some 25 years earlier, Baumann at once set about to determine the chemical nature of the therapeutically active constituent of the thyroid. At that time the therapy of goiter was still empirical and still con­ troversial even though the value of iodine-containing sub­ stances had been known for many centuries and, indeed, the effectiveness of iodine itself had been known since Coindet’s work in 1820. In the relatively short period of one and one- 8 EUGEN BAUMANN half years, Baumann’s work was crowned with success and in 1895 his now classical paper on “ thyroiodin” appeared in Hoppe-Seyler’s Zeitschrift fur physiolog'ische Chemie, 21, 319 (Dec. 1895). In a logically-planned series of studies, Baumann first devised a procedure for obtaining a purified preparation from sheep thyroid glands. With careful check­ ing of each step of purification with tests for activity in goitrous patients and metabolism studies in do^s, Baumann found- that the active constituent could be extracted from fresh minced glands with boiling 10% sulfuric acid. Upon cooling, an active flocculent precipitate settled out. The active substance was extracted from the precipitate by boiling 85% alcohol and the residue of the alcohol extract was treated with petroleum ether to remove adhering fat. The remaining substance was easily dissolved by 1% sodium hydroxide and reprecipitated by dilute sulfuric acid, filtered, washed, and dried. In Baumann’s words, an uninteresting appearing gray- brown, flaky material resulted in a yield of 0.2 to 0.5% of the original fresh glands. However, the substance rvas extremely active, containing approximately the same activity as the original quantity of fresh gland! Baumann named the active substance “ thyroiodin” and next sought to determine some­ thing of its chemical constitution. Upon heating it evolved an odor similar to that of a pyridine base. It was insoluble in water but somewhat soluble in alcohol. It dissolved easily in dilute alkali and was reprecipitated with dilute acid. It gave no protein reaction but contained about 0.5% of phos­ phoric acid in stable combination. Then came the discovery of iodine. After ashing in the then usual manner with caustic soda and saltpeter, Baumann noticed that a yellow color appeared after acidification of a solution of the alkali melt. Upon shaking the solution with chloroform a violet color ap­ peared in the chloroform layer indicating the presence of iodine. Furthermore, acidification of the warm alkali melt resulted in the evolution of a light violet vapor of iodine. With characteristic cautiousness, Baumann at first attributed the results to iodine contamination of his reagents but repe­

1 0 EUGEN BAUMANN burg. It is interesting to note .that a memorial biography of Hoppe-Seyler written by Baumann and Kossel appeared in the same volume of the Zeitschrift with Baumann’s now class­ ical paper relating the discovery of iodine in the thyroid gland as described above. Baumann was thus a remarkably productive investigator both from a fundamental and practical standpoint. His work emphasized the great significance of physiological chemistry and its allied science, nutrition, to the entire field of medicine. In addition, however, he was a warm, kindly person — an un­ derstanding teacher, and a sincere friend. In the words of his life-long friend and colleague, Kossel:

“ Allen, die ihm nahetraten, musste es als ein erstrebenswerthes Ziel erscheinen, so zu denken und so zu schaffen, wie er.’ ’

J ames M. Orten FECAL CHROMIC OXIDE CONCENTRATION IN 12 DAIRY COWS AS RELATED TO TIME AND FREQUENCY OF ADMINISTRATION AND TO FEEDING SCHEDULE

W. A. HARDISON, R. W. ENGEL, W. N. LINKOUS, H. C. SW EENEY AND G. C. GRAF Virginia Agricultural Experiment Station, Blacksburg

ONE FIGURE

(Received for publication July 28, 1955) Since its introduction (Edin, ’18), Cr203 has been used extensively in digestibility studies with several animal specieo. In most of these investigations the indicator was determined in a composite sample representing the total production of feces during periods of varying length It is obvious that the use of the indicator in this manner offers no advantage over the conventional digestion trial. However, if it were possible to predict accurately dry matter digestibility or fecal outgo from a partial or “ grab” sample of feces, a great advantage would be gained in terms of time and expense. It has been shown that when Cr203 is administered once daily in capsule form (Hardison and Reid, ’53) or mixed with the concentrate and given twice daily (Kane et ah, ’52), the excretion of the indicator is not constant but fluctuates greatly over a 24-hour period. This has recently been con­ firmed by Smith and Reid ( ’55) working with grazing dairy cows. Thus, great care must be taken in the sampling of feces with respect to the time of day or else some way must be found to reduce the variation in the excretion of the indicator. It is not clearly known how the excretion pattern of Cr203

i i 12 W. A. HARDISON AND OTHERS is affeeted by such factors as type of ration fed, time and frequency of administering the indicator, level of milk pro­ duction and activity of the animal. Neither is it known what effect age of the animal has on the excretion of Cr203. The over-all objective of this study was to investigate further the reliability of Cr2Os for estimating the dry matter digestibility and fecal output of lactating dairy cows under barn-feeding conditions. To determine the reliability of the indicator under these conditions, the following were studied: (a) the influence of time and frequency of Cr203 administra­ tion on the excretion of the indicator and (b) the relationship, if any, between the fecal concentration of the indicator and time of feeding. # In addition, the amount .of variation in dry matter digestion coefficients contributed by cows and by days was determined.

EXPERIMENTAL PROCEDURE Twelve lactating Holstein cows were used. Each treatment group consisted of one first-calf heifer and three older animals. The animals were grouped on the basis of production and body weight into three replicates and then allotted to the following treatment groups : group I was given 15 gm of Cr203 at the morning feeding ; group II, 7.5 gm of the indica­ tor at the morning feeding and 7.5 gm at the evening feeding; group III, 15 gm midway between the morning and evening feedings ; group IV, 7.5 gm midway between the morning and evening feedings and 7.5 gm midway between the evening and morning feedings. The Cr203 was administered in gelatin capsules and all the cows were fed twice daily at 6 :00 A.M. and 2:30 P.M. on grain, alfalfa hay and corn silage. The collection of feces was preceded by a 7-day preliminary feeding period during which time the animals received the indicator according to the above schedule. Following the preliminary period all feces were collected over a 5-day period. Fecal sub-samples were obtained from each animal at two-hour intervals over the entire collection period. These EXCRETION OF CHROMIC OXIDI 13 sub-samples were employed in an analysis of the Cr203-ex- cretion pattern of the animals.

RESULTS AND DISCUSSION An examination of the data revealed no differences in the excretion of the indicator attributable to the time of day the indicator was given (at-feeding vs. between-feeding). Neither was there any essential difference between the heifers and

FECAL SAMPLING TIME Fig 1 Average chromic oxide excretion-time patterns for cows administered chromic oxide in one and two doses daily.

the older animals in the excretion of the indicator. Thus, the data were analyzed on the basis of only two treatments, once- daily vs. twice-daily administration. The average Cr203-excretion patterns of the animals are shown in figure 1. Expression of the amount of Cr203 in the partial samples as a percentage of the concentration of Cr203 in feces compounded from the total collection during the entire 5-day period allows the Cr203 excretion rates of animals consuming different quantities of feed to be compared. 14 W. A. HARDISON AND OTHERS

A careful examination of the individual excretion, patterns revealed that the excretion rates of two of the animals (one in the “ onee-daily,” one in the “ twice-daily” administration group) were distinctly different from the excretion patterns of the other animals in their respective groups. Since these two animals stood side-by side in the barn, it was strongly suspected that an error was made in dosing these animals one day during the trial. Thei efore, the data for these animals were eliminated and the curves in figures 1 are the average of 5 animals in each case instead of 6. The excretion patterns observed when the indicator was given once daily were of the same general form as those obtained by Kane et al. ( ’52) as reported by Hardison and Reid ( ’53). The average excretion pattern in figure 1 shows that the animals excreted at certain periods of the day from 91 to 111% as much Cr203 per unit of dry matter as was found in feces representative of those produced during the 5-day period. Although the variability in the excretion of the indicator was less than has been reported for grazing animals (Hardison and Reid, ’53; Smith and Reid, ’55), it is sufficiently great to preclude random sampling of feces. When the indicator was given twice daily the variability in the excretion rate was reduced substantially (fig. 1). In this instance the recovery of Cr203 ranged from 97 to 103%. Theoretically, this variability could be reduced still further by Inore frequent dosing of the indicator. It would appear from these data that random sampling of feces could be followed if the indicator were given twice daily. It is apparent from the excretion patterns in figure 1 that there v7as no significant relationship between the excretion of Cr203 and the time at vdiich the animals "were fed. The data were treated statistically and a brief report of the statistical analysis will be given here (Analysis of Variance, Snedecor, ’46). A study -was made of the accuracy with which fecal output could be predicted from the fecal concentration of Cr,03 when the indicator was given once or twice daily and when partial EXCRETION OF CHROMIC OXIDH 15

samples were taken twice daily and composited on an equal- weight basis. In the analysis of the data the percentage deviation from the measured output of fecal dry matter as determined by the following formula was used:

Estimated D.M. Output — Measured D.M. Output ------— x 100 Measured D.M. Output The results of this analysis arc shown in table 1. When the indicator was administered twice daily and partial fecal samples taken at 6 a.m . and 4 p .m . the random error was about 1/16 of the error for the same sampling times but for once-daily administration. In the case of the 6 a.m . and 6 p .m sampling times the random error for the once-a-day

TABLE 1 Variance analysis of the estimated daily fecal dry matter output

SAMPLING TIME 6 A .M . - 4 P.M. 6 A .M . - 6 P .M .

Cr203 FREQUENCY (DAILY) IX 2X IX 2X

SOURCE OP VARIATION D.P. Mean Squares

Cows 4 367 84 575 132 Days 4 95 82 925 135 Cows X days (error) 16 467 28 231 47 administration of the indicator was about 5 times greater than the random error resulting from the twice-a-day administra­ tion. This may be considored to be at variance with a recent report by Smith and Reid ( ’55). These workers were able to estimate the fecal output of grazing dairy cows just as accurately when Cr203 was administered once daily as when given twice daily. Lastly, the data in table 1 show no cow or day differences in contribution to variance when the indicator was given twice daily. However, this was not true for once-a-day administration. The data collected in this experiment afforded an excellent opportunity to study the variance contributed by cows and by days to the true or measured dry matter digestion coef­ ficients. (Coefficients calculated from total feed intake and 16 V. A. HARDISON AND OTHERS total fecal output). The calculation of dry matter digestibility by cows revealed a mean for the 5-day period of 65.31 with values for individual animals ranging from 61.10 to 68.33 and a standard deviation of 2.23. The calculation by days revealed a mean for the 12 animals of 65.31, as before, but values tor individual days ranged from 63.66 — 66.36 (S.D. — 0.94) or a lesser range and a smaller standard deviation than was observed for individual animals. It is evident that there was more variability among animals than among days. In table 2 is given the variance analysis of the measured dry matter digestibility for cows and for days. This analysis reveals that the variability among cows was highly significant

TABLE 2

Variance analysis of the measured dry matter digestibility for cows and for days

SOURCE OP VARIATION D.F. M.S. F Cows i i 24.96 2.58** Days 4 13.54 1.40 Cows X days (error) 44 9.67 while the variability among days was not significant. Thus, it is indicated that in digestibility studies more reliable data may be obtained by making observations on a large number of animals over several days. In view of these observations, data from three digestion trials conducted in this laboratory over the past three winter seasons were similarly analyzed. In two of the three trials, involving 4 cows in 10-day collection periods, cews contributed significantly more variability thah did days while in the third trial cows were again more variable than days but the dif­ ference was not significant.

SUMMARY A study was made of the effect of certain factors on the excretion pattern of Cr203, namely, frequency and time of administration. Twelve lactating cows were employed in the study. EXCRETION OP CHROMIC OXID 17

From the results of this experiment the following conclu­ sions have been drawn: 1. Variability in the excretion of Cr,03 was much greater when the Cr203 was given once daily than when it was given twice daily. When the indicator was adminh tered twice daily the random error was reduced 5- to 16-fold as compared with the error with single daily administration. There was no essential difference noted between at-feeding or between- feeding administration of the Cr203. 2. No significant relationship was observed between fecal concentration of the indicator and time of feeding under the conditions of this experiment. 3. Since the amount of variation in the measured dry matter digestion coefficients contributed by cows was greater than that contributed by days, it is indicated that in digesti­ bility studies more reliable data may be obtained by making observations on more animals over a fewer number of days. 4. It appears that one of the great contributions that indicators can make to such studies is that through their use data may be collected on many animals rather than having to draw conclusions from the use of a limited number of animals. 5. It is recommended that the indicator, Cr203, be given in two doses daily when it is employed for the estimation of fecal output or dry matter digestibility in mature animals.

LITERATURE CITED

Edin, H. 1918 Ohenterande Forsok Over en pa Ledkropprincipen Grundad Metod att Bestamma en Foderblandnings Smaltbarhet Meddelande N : r 309 fram Centralanstalten for Forsoksvasendet pa Jordbruksomradet, Husdjursavdelningen N : 4 50, Stockholm. Hardison, W. A., and J. T. R eid 1953 Use of indicators in the measurement of the dry matter intake of grazing animals. J. Nutrition., 51: 35. K ane, E. A., W. C. Jacobsck and L. A. Moore 1952 Diurnal variations in the excretion of Cr203 and lignin. Ibid., 47: 263. Smith, A. M., and J. T. Reid 1955 Use of chromic oxide as an indicator of fecal output for the purpose of determining the intake of pasture herbage by grazing cows. J. Dairy Sci., 38: 515. Snedecor, G. W. 1946 (Statistical Methods The Iowa State College Press, Ames. STUDIES OF THE EFFECTS OF CHLORTETRACYCLINE, VITAMIN E AND VITAMIN K IN THE NUTRITION OF THE RAT

I. GROWTH

ELWOOD E. REBER, C. C. MORRILL, HORACE W. NORTON AND HARRY E. RHOADES Department of Veterinary Physiology and Pharmacology and the Department of Veterinary Pathology and Hygiene, College of Veterinary Medicine and Agricultural Experiment Station, University of Illinois, Urban a

ONE FIGURE

(Received for publication August 10, 1955)

The role of vitamin E has been discussed recently in a symposium (Herriott, ’53). Particularly pertinent to this paper was the discussion of the relationship of Chlortetra­ cycline (aureomycin) to vitamin E. Although Chlortetra­ cycline and vitamin E prevent hepatic necrosis the basic re­ lationship between these two factors has not been elucidated. Furthermore, there is little known of the relationships of fat- soluble vitamins with antibiotics. Daft and Schwarz ( ’52) reported that rats fed a ration deficient in riboflavin or panto­ thenic acid were protected by the addition of Chlortetracycline. An examination of the chemical structures of vitamins E and K and Chlortetracycline reveals a common characteristic, that is, the possibility of their participation in oxidation-reduction reactions. An investigation was made to determine the effect these substances and their combinations would have upon the growth of rats. 19 2 0 REBER MORRILL, NORTON AND RHOADES

PROCEDURE A factorial design was used consisting of 8 treatments. The variables and the respective designations for the rations are as follows: menadione (K) was included in rations K, KE, KA and KAE ; vitamin E (E) was included in rations E, EK, EA and E AK ; chlortetracy line (A) was included in rations A, AK, AE and A EK ; and one treatment (0 ) was deficient

TABLE 1 Composition of rations

RATION 1 2 % 1° Casein, vitamin-free 20.0 20.0 Lard 38.0 38.0 Salts, U.S.P. X IV 4.0 7.0 Sodium chloride 2.5 Cod liver oil (1000 U.S.P., A : 100 U.S.P., D/gm ) 2.0 2.0 Vitamin mix 1 2.4 2.4 Menadione1 2 3 0.02 0.03 dl-alpha Tocopherol acetate 1 0.02 d-alpha Tocopheryl acetate 2,3 0.03 Aureomycin - HC12 0.02 0.03 Starch to 100 to 100

1 The following amounts of vitamins were added per kilogram of ration: thia­ mine, 16 mg; riboflavin, 16 mg; pyridoxine, 10 mg; Ca pantothenate, 200 mg; choline, 6gm ; nicotinic acid, 16 mg; inositol, 2gm ; p-aminobenzoic acid, 0.6 gm; biotin, 500 fig; folic acid, 1 mg. 2 Variables in the different treatments employed. 3 Myvamix, a vitamin E feed supplement is a soybean meal base containing 14.71 gm of d-alpha tocopheryl acetate (20,000 I.U. vitamin E) per pound. Distil­ lation Products Industries, Rochester 3, N. Y. in vitamins E and K and lacking in chlortetracycline. The experiment consisted of three replications. Replication 1 was fed ration 1 (table 1). Pour weanling female rats were used in each treatment. Replications 2 and 3 were fed ration 2 (table 1). Replication 2 consisted of 5 weanling female rats per treatment. Replication 3 consisted-of 5 weanling male rats per treatment. An additional treatment was added to replication 3 by feeding 5 rats a commercial laboratory CHLORTETRACYCLINE AND VITAMIN E 21 ration.1 Rats of the Sprague-Dawley strain were used in this experiment. Replications 2 and 3 were distributed into the rat rack space so that each of the 8 treatments occurred once in each of the 5 levels, permitting measurement of the influence of level in the rack upon the response and elimina­ tion of level as a source of experimental error. The compositions of the rations are shown in table 1. Ra­ tion 1 contained the variables at a level of 0.02%. The vita­ min E requirement of the rat has been described as 1 to 3 mg per day (Farris and Griffith, ’49). Ration 1 supplied 2 mg of vitamin E per 10 gm of ration. The vitamin K requirement for the rat has not been established. A ration containing 0.02% of chlortetracycline has been fed by Daft and Schwarz ( ’52). The vitamin K and chlortetracycline were added at the 0.02% level in order to compare their effects with that of vitamin E on an equal weight basis. Ration 2 contained the same variables at a level of 0.03%. Higher levels of the variables were used to supply vitamin E at the rate of 3 mg per 10 gm of ration and to test a higher level of antibiotic. In addition a more stable form of vitamin E 2 was used in ration 2. The water-soluble vitamins were added to each ration at a higher level than recommended (Farris and Griffith, ’40). Two kilograms of each ration were prepared at one time and refrigerated until fed to the rats. The rats were fed fresh ration three times each week. Food consumption data were obtained for the first 30-week growth period. The rats were weighed at weekly intervals. The rats of replications 1 and 2 (females) were continued on their respective rations for reproduction studies. Those of replication 3 (males) were sacrificed at the end of the 10th experimental week and the internal organs weighed. Sections of the internal organs and muscle were taken for histopatho- logical examination. The tissues were stained with hema­ toxylin and eosin. Sections of the liver were analyzed for

1 Purina Lab Chow. 2 Myvamix, Distillation Products Industries, Rochester, New York. 2 2 REBEB MORKILL, NORTON AND RlSOADES their vitamin A content by the ahtimony trichloride method (Johnson, ’48). The entire cecal contents of two male rats fed each diet were combined and placed in a sterile weighed jar containing glass beads. The jars representing each diet were reweighed and sterile physiological saline added to make a 1 to 40 dilu­ tion. Plate counts were made using the dilutions 1 X 10~4, 1 X 10 J 1 X 10“6, 1 X 10“7 and 1 X 10-8. Duplicate plates were poured with MacConkey’s agar and with a medium selec­ tive for Lactobacilli reported by Morrison, Mitchell and Wise­ man ( ’51). The plates were incubated for 48 hours at 37°C. The Lactobacilli medium plates were incubated under 10% carbon dioxide. Representative types of colonies on the Lactobacilli medium were stained by the Gram method and the number of each type were recorded.

RESULTS AND DISCUSSION All of the rats in the study survived and there was no ap­ parent disease or unusual incident during the 10-week growth period. The position of the rat in the rack in replications 2 and 3 had no statistically significant effect upon growth or food consumption for either males or females. The average weight gained and food consumed per gram of gain for the rats in the three replications are presented in table 2. The differences in initial weights of the rats were not perceptibly reflected by final weights or food consump­ tion. There was a difference in response between replications 1 and 2 with respect to body weight and food consumption. The final weight in replication 1 was 7.2 gm higher than in repli­ cation 2, but not significantly so. This difference is 1.7 times its standard error, hot when adjusted to equal food consumption it falls to less than 0.5 standard error. Food consumption in replication 1 was greater than that in replica­ tion 2 to attain the same body weight. Food consumption in replication 2 was lower (significant at 2% level) than in OHLORTETBACYCLINE AND VITAMIN E 23 replication 1 to the extern of 23.2 gm. These differences should be considered with the fact in mind that the levels of E, K, and A were each 0.02% in replication 1 hut 0.03% in replication 2. The effects of the respective treatments in growth and food consumption in replications 2 and 3 were approximately equal. The average daily food consumption daring the 10-week test period was 7 to 8 gm per day in replications 1 and 2.

TABLE 2

The average weight gained and food consumed per gram o f gain for the female (1, 2) and male (S) rats

AV. WEIGHT GAINED AV. FOOD CONSUMED/WT. GAINED t r e a t m e n t 1 1 2 2 3 1 2 3 gm gm gm 0 144 143 201 3.89 3.45 2.77 K 162 147 242 3.19 3.32 2.59 E 180 151 243 3.03 3.40 2.74 A 168 159 246 3.21 3.23 2.57 EK 167 160 250 3.30 3.33 2 . 6 6 AK 177 146 254 3.24 3.46 2.55 EA 173 180 256 3.19 3.08 2.51 EAK 157 173 254 3.25 3.17 2.55 ' 0 indicates treatment deficient in vitamins E and K and lacking in clilortetra- cycline. K = Menadione. E = Vitamin E. A = Chlortetracycline. 2 Replications.

This average food intake supplied an average daily intake of 1.5 mg and 2.1 mg of the variables, vitamins E and K and chlortetracycline, to the female rats in replications 1 and 2 respectively. The difference in levels of the variables fed was taken into account, on the assumption that the response was proportional to the dose, as it appeared, and all three replications were pooled to evaluate the treatment effect. The difference in levels of E, A and K for replications 1 and 2 did not cause a significant decrease in body weights with or without adjustment to equal food consumption. The re­ 24 KEBEK, MORRILL, NORTOX AND RHOADES gression of final weight on food consumption was 0.389 gm per gram, with standard error of 0.038. Rats fed rations 1 or 2 which were deficient in vitamins E and K and lacking in chlortetracycline had the poorest growth. Vitamin E and chlortctracycline each had a significant (1% level) beneficial effect upon, growth. The average effect of vitamin K was not significant, but the interaction of vitamins Iv and E was significant at the 5% level. Hence, the situation is not clear, but the 4 relevant treatment means, given in table 3, suggest that each was of benefit but no added benefit resulted from both. It should be mentioned that, in replica­ tions 2 and 3 combined, the interaction of vitamin Iv and chlortetracycline was significant at the 5% level but this

TABLE 3 Average weight gains,1 in grams, as affected by vitamins K and E, replications 1, 2, and 3 combined

V IT A M IN K

Absent Present-

VITAMIN Absent 228.9 249.3 K. Present 240.6 243.7 1 Standard error, 3.5 gm.

was not confirmed by replication 1. No other combination of vitamins E and K and chlortetracycline had a significant effect on growth. These data indicate that chlortetracycline and, to a lesser extent, vitamin Iv are capable of partially relieving a dietary deficiency of vitamin E. The various diets fed the male rats appeared to produce no differences in the liver, thyroid, spleen or kidney weights when compared on the basis of percentage of body weight. Histopatliological examination revealed some moderate de­ gree of degeneration and desquamation of thyroid alveolar epithelium in rats fed the ration deficient in all variables. Table 4 presents the average weights of the testes ex­ pressed as percentage of body weight. The testes weights of those rats fed the basal ration (0 ) were lower than those CHLORTETRACYCLINE AND VITAMIN E 25 of rats receiving the dietary variables. Analysis of variance on testes weights suggests fairly clearly that K had an effect, and less clearly that vitamin K interacted with both vitamin E and chlortetracycline. In both cases, each additive ap­ peared beneficial alone, but no increased benefit was received from both. Differences in the development of the testes due to the treatments probably existed which were not fully re­ flected in weight. Histopathological examination revealed clearly significant lesions only in the testes of rats in the

TABLE 4 The average weights of the testicles and vitamin A content of the livers of male rats

TESTICLES LIVER TREATMENT 1 Av. percentage Standard Av. vitamin A Standard of body \vt. deviation per gin liver deviation I.TJ. 0 0.79 0.25 260 80.4 K 1.10 0.05 374 57.3 E 0.97 0.07 1039 539.0 A 0.97 0.07 942 354.5 KE 0.99 0.08 1206 340.0 KA 1.00 0.09 1877 651.4 EA 1.03 0.07 2811 1230.7 KEA 1.01 0.07 2409 1110.4

1 0 indicates treatment deficient in vitamins E and K and lacking in chlortetra­ cycline. K = Menadione. E = Vitamin E. A = Chlortetracycline.

group fed the ration deficient in all variables. The lesions in the testes consisted of progressive degeneration and reduc­ tion in the number of cells lining the seminiferous tubules. In some cases, only a few Sertoli cells remained (fig. 1C). Multinucleated giant cells appeared especially early in the degenerative process (fig. IB). The degenerative process appeared to involve chromatolysis and fusion of sperma­ tozoa, vésiculation and disintegration of spermatids, and de­ generation of spermatocytes, spermatogonia and even some of the Sertoli cells (Follis, ’48). REBER, MORRILL, NORTON AND RHOADES

Fig. 1 Sections from rat testes: A, normal seminiferous tubules: B, semi­ niferous tubules showing early degeneration with formation of numerous multi- nucleated giant cells; C, advanced degeneration of seminiferous tubules. Hema­ toxylin and eosin. 200 X. CHLORTETRACYCLINE AND VITAMIN E 27

Table 4 also presents the vitamin A contents of the livers and their respective standard deviations. The addition of either vitamin E or Chlortetracycline to the basal ration resulted in an increase in the concentration of vitamin A in the liver which was significant at the 0.1% level. Vitamin K, alone or in combination with the other variables, did nod increase the .concentration of liver vitamin A significantly Table 5 presents the bacteria and yeast counts on cecal contents of male rats at the end of the 10th experiment week. Chlortetracycline and vitamin K apparently reduced the num- oer of coliform type of bacteria Chlortetracycline was ef-

TABLE 5 Bacteria and yeast counts of the colon contents of male rats 1

TREATMENT 2 COLIEORM LACTIC YEAST Lab Chow 9.1 1350 none E 8.6 294 none EK 0.9 660 300 EA 2.6 none 109 EAK 3.5 none 24

1 All figures are in millions. 2 E = Vitamin E. K = Menadione. A = Chlortetrac} eline. fective in eliminating the lactic acid bacteria. The presence of Chlortetracycline and vitamin K in the diet appeared to be conducive to the growth of yeast in the cecum.

SUMMARY A factorial design was used to ascertain the effects of vitamins E and K and Chlortetracycline upon rats during their growth period. The variables were fed at levels of 0.02% or 0.03%. The addition of vitamin E or Chlortetracycline to a vitamin-E-deficient diet increased growth significantly (1% level). There was some indication of an interaction of vita­ mins K and E. While the testicles of rats fed the basal ration showed severe degeneration on histopatliological examina­ 28 REBER, MORRILL, NORTON AND RHOADES tions. those of animals fed rations containing chlortetrh- cycline, vitamin K or vitamin E showed no snch degenera­ tion. Vitamin E or Chlortetracycline but not vitamin K increased the vitamin A concentration in the liver. Chlortetra­ cycline rednced the nnmber of the coliform and lactic acid bacteria. Vitamin K reduced the number of coliform bacteria. Chlortetracycline and vitamin K were conducive to the growth of yeast in the colon.

ACKNOWLEDGMENTS The authors express their appreciation to : Dr. C. E. Schoettle and Mr. R. T. Hamilton, College of Veterinary Medicine, University of Illinois, for technical assistance, Dr. L. Michaud, Veterinary Research Department, Merck and Company, Rahway, New Jersey for B-eomplex vitamins; Dr. T. Jukes, Lederle Laboratories, American Cyanamid Com­ pany, New York, for aureomyein; Mr. J. Byers for statistical assistance; and The Research Laboratories, Swift and Com­ pany, Chicago, Illinois, for a research grant which partially supported this work.

LITERATURE CITED

D a f t , F. S., a n d K. S c h w a r z 1952 Prevention of certain B vitamin de­ ficiencies with ascorbic acid or antibiotics. Fed. Proc., 11: 200. F a r r is , E. J., a n d J. 0. G r if f it h , J r . 1949 The Rat in Laboratory In­ vestigation. J. P. Lippineott Co., Philadelphia. F o l l is , R. H., J r . 1948 The Pathology of Nutrition Disease. C. C Thomas, Springfield, Illinois. H errio tt, R. M. 1953 Symposium on Nutrition. The Physiological Role of Certain Vitamins and Trace Elements. The Johns Hopkins Press, Baltimore. Experimental Muscular Dystrophy, by C. G. MaeKenzie; Vitamin E in Early Life, by K. E. Mason; Some Metabolic Effects of Vitamin E, by P. Gyorgy. J o h n s o n , B. C. 1948 Methods of Vitamin Determination. Burgess Pub­ lishing Company, Minneapolis, Minn. M orrison, R., J. Mitchell and R. W iseman 1951 A selective medium for the isolation and enumeration of oral and fecal Lactolacilli. J. Baet., 62: 132-133. FEASIBILITY OF THE CHROMIC OXIDE AND THE LIGNIN INDICATOR METHODS FOR METABO­ LISM EXPERIMENTS WITH CHICKENS

WERNER J. MUELLER2 Swise Poultry Breeding School, ZolWkofen-Bern and Department of Animal Nutrition, ETH, Zurich

(Received for publication July 11, 1955)

One of the principal difficulties of metabolism experiments is the identification of the excreta that correspond to a spe­ cific food intake. In experiments which employ the conventional time-collection method a representative output of excreta during the experimental period is attempted by feeding the same ration during both a preliminary, and the subsequent experimental period. The necessary measurements of diet intake and output of excreta make metabolism experiments according to this method very laborious. Recent improvements in indicator methods promise a con­ siderable simplification of metabolism experiments. These methods use as an indicator of digestibility a substance that is not absorbed or altered during its passage through the digestive system. Digestion coefficients are calculated by means of the following formula: a — b digestion coefficient = 100 a Where : a = grams of nutrient per gram of indicator in diet b = grams of nutrient per gram of indicator in excreta In experiments with poultry two substances have been used as indicators of digestibility : barium sulfate by Whitson 1 Taken from a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Technical Sciences to the Swiss Federal Institute of Technology (ETH). “Present address: Department of Poultry Husbandry, The Pennsylvania State University, University Park. 29 30 WERNER J. MUELLER

et al. ( ’43), and chromic oxide hy Olsson ( ’50) and Dansky and Hill ( ’52). Except for critical studies on the chromic oxide method by Olsson and Kihlen ( ’48), and Dansky and Hill ( ’52), no information is available on the accuracy of indicator methods in experiments with chickens. The experiments reported here were designed to test two indicators, chromic oxide and lignin, and to .obtain more general information on the accuracy of indicator methods as compared with the time-collection method.

EXPERIMENTAL The accuracy of the chromic oxide method was studied in 6 series of digestion experiments (C-l to C-6) with a total of 16 male chickens of various breeds and ages (table 2). All chickens were kept in individual cages. The study of the lignin method was based on lignin determinations on samples of diet and excreta obtained in 5 of the 6 series of experi­ ments on the chromic oxide method (C-2 to C-6). Each digestion experiment consisted of a preliminary period of at least 5 days and an experimental period of 8 days dui ing which diet consumption and excreta output were recorded quantitatively. The experimental period was divided into two equal sub-periods of 4 days each. In each of these sub­ periods the excreta voided during the daytime (8 a.m . to 5 p .m .) and those voided during the night (5 p .m . to 8 a.m .) weie collected and analyzed separately. An all-masn diet, satisfying the known nutrient require­ ments of growing and adult chickens was used as a carrier for the chromic oxide. The diet was ground coarsely for experiment C-l and to a particle size of less than 1 mm for all other experiments. For experiments C-l to C-3 the chromic oxide was mixed directly into the diet, whereas for experi­ ments C-4 to C-6 it was first baked With wheat flour. The chromic oxide content of the diet was approximately 2% in experiments C-l to C-5 and approximately 1% in experiment C-6. During each preliminary and subsequent experimental period the same quantity»of diet was fed daily. The daily FEASIBILITY OF INDICATOR METHODS 31 rations were mixed with a quantity of water corresponding to half their weight and fed in two equal portions at 9 a .m . and at 2 p .m . Scattered feed was recovered and fed after the main part of the ration had been consumed. Chromic oxide was determined according to Schlirch et al. ( ’50), and lignin according to Ellis et al. ( ’46).

RESULTS Chromic oxide. The rate at which chromic oxide appeared in the excreta after the first feeding of a chromic oxide-con­ taining diet was determined in the preliminary periods of experiments C-3 and C-4. Two days after the first feeding

TABLE 1 Cr202 excretion following the first feeding of a Cr.,03-containing diet

DAYS AFTER FIRST FEEDING OF A Cr20 3-CONTAINING DIET

EXPERIMENT 1 2 3 4 5 Cr203 in 1 gm of air-dry excreta mg mg mg mg mg 0-3 45.3 51.1 52.8 50.4 52.6 C-4 51.2 53.2 53.7 51.5 51.4 of a chromic oxide-containing ration the indicator content of the excreta reached a fairly constant level (table 1). This must be a result of the usual rapid passage of feed through the digestive system of chickens, a condition verified as follows. When 20 gm of the experimental diet, to which small celluloid particles had been added as a marker, were fed to 4 8-week-old Rhode Island cockerels, 70% of the marker was recovered in the excreta within 9 hours after feeding the test ration. The rapid appearance of chromic oxide in the excreta in experiments C-3 and C-4 and the marker test described above indicate that in experiments employing the chromic oxide indicator method a preliminary period of 5 days is sufficient. With regard to the experimental period it was found that there were no significant differences between the chromic oxide content of the excreta voided in corresponding 4-day 32 WERNER J. MUELLER sub-periods. Significant differences were found between the chromic oxide content of day (8 a .m . to 5 p .m .) and night (5 p .m . to 8 a.m .) excreta (table 2). These differences could not be reduced significantly by varying the method of adding chromic oxide. The day excreta of growing cockerels con­ tained more chromic oxide than the corresponding night ex­ creta, whereas in the experiments with adult cocks the reverse was true. In an nnreported experiment with 4 8-week-old Single-comb White Leghorn cockerels it was found that the differences between the chromic oxide content of day and night excreta probably has physiological causes. The con­ tents of the ceca, which in this experiment were always voided between 5 p .m . and 8 a .m ., contained less chromic oxide than the corresponding contents of the intestines. Calculation showed that the difference between the chromic oxide content of the excreta from ceca and intestines was the main reason for the daily variation in chromic oxide content of the excreta. From the data in table 2 and other observations during the experiments reported in this paper it seemed probable that in the adult cocks the contents of the ceca were, for unknown reasons, richer in chromic oxide than the corresponding con­ tents of the intestines. In general the results of these experiments indicate that in the case of poultry, aliquot samples of the excreta voided during entire 24-hour periods are required for accurate di­ gestibility determinations with the chromic oxide method. With the exception of experiment C-l, where chromic oxide was mixed with a coarsely ground diet, the chromic oxide recovery was good (table 2). Neither baking the chromic oxide with wheat flour (experiment C-4) nor a reduction of the indicator content of the diet from approximately 2% to ap­ proximately 1% (experiment C-6) significantly increased the indicator recovery as compared with an addition of approxi­ mately 2% chromic oxide to a finely ground diet. Lignin. The differences between the lignin content of day and night excreta were smaller than the differences between the chromic oxide content of corresponding samples (table 2). TABLE •«a ^3 'Q • ^3 Si «3 Si si si <3 «3 Si Si Q Si SI MO g§ < fc Q 3 *3 M fc >< METHODS INDICATOR OF FEASIBILITY w £ +1 £ # +1 # o P y-1>vi tf r M% pH H3 o O * P © © *> * © 1 i h ^ ¡S § h

O l v C O +1 + 1 P * § h o .S 510 »o p=< © o io o i* S - +1 +1 +1 II -1 © - p © a a ap s ^ * -s r* £ o E +1 6 tf r II o b p* © £> © &D5 £ h 33

» SP«» n3 rtf O F +1 •H © © o o 3 <2 * h 2 m S 1 rtf o Cl si .s CO o +1 +1 +1 +1 o 0 © © OD O ©

Mean ± standard error. The values are averages of two sub-periods. 33

FEASIBILITY OF INDICATOR METHODS 35

Accuracy of indicator methods. If the diurnal variation of the indicator content of the excreta is excluded as a source of error by appropriate sampling, the accuracy of indicator methods depends principally upon the recovery of the indi­ cator intake. The relationship between the true digestion coefficient of a nutrient (DC), the digestion coefficient’ according to the indicator method (DCr), and the indicator loss in percentage of the indicator intake (IL) may be expressed by the following equation: IL (100 — DC) DC — DC, = ------100 — IL This formula shows that the difference between the true digestion coefficient and the digestion coefficient found with an indicator method (called “ error” in the following) de­ pends not only on the size of the indicator loss but also on the digestibility of the nutrient under examination. The error caused by an indicator loss of 10% e.g. will be 9% for a nutrient with a digestibility of 20% and only 2% for a nutri­ ent with a digestibility of 80%. In the experiments reported, the chromic oxide losses varied between — 1.2% and -f- 4.5% when a finely ground diet was used as a carrier for the indicator. For this range the error is small even if the digestibility of the nutrient under exami­ nation is low. Therefore, if the sampling is appropriate, the chromic oxide method should give good results. Nevertheless, it has to be expected that, due to small indicator losses, the digestion coefficients found with the chromic oxide method will generally be somewhat too low.

SUMMARY The feasibility of the chromic oxide and the lignin indi­ cator methods for metabolism experiments with chickens was examined in 6 series of digestion trials employing the con­ ventional time collection method. 36 WERNER J. MUELLER

In all experiments the chromic oxide content of the day excreta differed significantly from the chromic oxide content of the night excreta. An average of 95.3% of the chromic oxide intake was recovered in the excreta. It was found that the error caused by an indicator loss of 5% is, especially in highly digestible nutrients, relatively small. The differences between the lignin content of day and night excreta were much smaller than these between the chromic oxide content of corresponding samples An average of 99.1% of the dietary lignin was recovered in the excreta. It was found, however, that lignin recovery is greatly affected by small variations in the concentration of the sulfuric acid used for lignin determinations.

ACKNOWLEDMENTS The interest and valuable advice of Dr. E. Crasemann, Dr. H. Engler and Dr. A. Schiirch are gratefully acknowledged.

LITERATURE CITED

D a n s k y , L. M., a n d F. W. Hill 1952 Application of the chromic oxide indi­ cator method to balance studies with growing chickens. J. Nutrition, 47: 449. E l l is , G. H., G. M a t r o n e a n d L. A. M a y n a r d 1946 A 72 per cent H,S04 method for the determination of lignin and its use in animal nutrition studies. J. of Animal Science, 5: 285. M u e l l e r , W. 1953 Die Brauchbarkeit von Indikatormethoden bei Verdauungs- versuchen mit Gefluegel. Diss. Zuneh. O l s s o n , N. 1950 Smaltbarhetsforsok med fjtiderfa. Statens Husdjursforsok, Meddelande Nr. 43, Norrtalje. O l s s o n , N., a n d G. K i h l £n 1948 Edins indicator method in digestibility ex­ periments on poultry. Proc. 8th World’s Poultry Congress: 225. S c h u r c h , A. F., L . E. L lo y d a n d E. W. Cr a m p t o n 1950 The use of chromic oxide as an index for determining the digestibility of a diet. J. Nu­ trition, 41: 629. W h it s o n , D., C. W. Ca r r ic k , R. E. R oberts a n d S. M. H a u g e 1943 Utilization of fat by chickens —- a method for determining the absorption of nutrients. Poultry Sci., £S: 137. INFLUENCE OF AGE AND SEX ON THE UTILIZATION OF PROXIMATE NUTRIENTS AND ENERGY BY CHICKENS 1

W. J. MUELLER, R. V. BOUCHER AND E. W. CALLENBACH Departments of Agricultural and Biological Chemistry and Poultry Husbandry, The Pennsylvania State University, University Parle

(Received for publication July 11, 1955)

The general opinion that digestibility of nutrients is not affected by age is based on results of experiments with sheep and swine after weaning. Wolf (1873) found that at 6 months of age sheep digested their feed as well as 8 months later. In more recent studies with swine, Watson et al. ( ’43) and Crampton and Whiting ( ’43) found no differences between the digestibility of various feedstuffs for ages from three months to 168 days and live weights from 40 to 140 kilograms. Chickens grow more rapidly than sheep or swine and do not nurse during the first period of postnatal life. These differences make questionable any direct application to young chickens of the results of digestion experiments with sheep and swine of post-weaning age. There is little information in the literature on the utilization of nutrients by growing chickens. In an experiment on the digestion of different levels of fat by Barred Plymouth Rocks, Whitson et al. ( ’43) found a significant increase of fat absorption from 8 to 10 and 12 weeks of age, if diets contain­ ing three to 9% of fat were fed, but not if the diet contained 20% fat. Fraps and Carlyle ( ’39) reported that chicks fed 1 Authorized for publication on May 18, 1955 as paper No. 1983 in the Journal series of the Pennsylvania Agricultural Experiment Station. This investigation was supported in part by a research fellowship from the Cooperative Grange League Eederation Exchange, Inc., Ithaca, New York.

37 38 MUELLER, BOUCHER AND CALLENBACH ad libitum a diet containing approximately 19% of protein retained 42.5% of the protein consumed during the period of approximately one to 4 weeks of age, 38.1% during the period from approximately one to 7 weeks of age, and 33.1% during the period from approximately one to 13 weeks of age. This paper reports the digestion and retention of nutrients by New Hampshire chickens at two, 4, 7, 11 and 26 weeks of age. EXPERIMENTAL Digestibility and retention of nutrients were determined on 5 groups of 15 male and 5 groups of 15 female New Hampshire chicks at the ages of 2, 4, 7 and 11 weeks, re­ spectively. Losses occurring between experiments were re­ placed with chicks from the same hatch, of the same sex, and of approximately the same weight. These replacements had also received the same diet and were raised in the same environment. Thus at the time of any digestion experiment each group numbered 15 chicks. For the experiment at 26 weeks of age only the 5 groups of pullets were used and the number of pullets per group was reduced to 10. For the first 12 weeks the experimental and reserve groups were held in growing batteries located in a room with a constant temperature of approximately 75°F. and light from 6 a.m. to 10 p.m. At 12 weeks all cockerels were discarded and the pullets were distributed at random in two 10’ X 12’ floor pens. When the pullets were 25 weeks old the original groups were reconstituted and, after reducing the number of chickens per group to 10, placed in batteries. The composition of the experimental diet (table 1) was the same during the entire investigation. Two different mixes of the experimental diet were prepared, one for the experi­ ments at 2, 4, 7 and 11 weeks of age, the other for the experiment at 26 weeks of age. To reduce feed loss and to assure a uniform intake of all components of the diet both mixes were pelleted. In the experiment at 26 weeks of age limestone grit was fed ad libitum in addition to the diet. NUTRIENT UTILIZATION BY CHICKENS 39

Chromic oxide was used as an indicator of digestibility. To avoid a sedimentation of the relatively heavy indicator during feed handling and pelleting, the chromic oxide mixture was prepared as follows: 10 pounds of chromic oxide C.P. were mixed with 40 pounds of cornmeal. The mixture was soaked with water, kneaded, baked at a temperature of approximately

TABLE 1 Composition of the experimental diet1

POUNDS Ground yellow corn 521.8 Wheat middlings 50.0 Ground heavy oats 50.0 Soybean oil meal 115.0 Fish meal 110.0 Alfalfa meal, dehydrated 25.0 Dried whey 30.0 Brewers’ dried yeast 24.0 Ground limestone 10.0 Steamed bonemeal 10.0 Iodized salt 2.5 Manganese sulfate (technical) 0.2 »-activated animal sterol (3000 I.C.U. per gram) 0.3 Riboflavin concentrate (8 mg per gram) 0.2 Stabilized vitamin A concentrate (10,000 I.U. per gram) 1.0 Chromic oxide mixture2 50.0

1000.0

1 This diet contained: moisture 9.24%, protein (N X 6.25) 20.82%, ether extract 4.06%, fiber 3.22%, ash 6.90%, nitrogen-free extract 55.76% and gross energy (calculated according to Fraps et al., ’40) 4.04 Cal. per gram. 2 Containing 10 lb. of chromic oxide, baked with 40 lb. of cornmeal.

190° F. and ground to a particle size of less than 1mm diameter. Digestion experiments of 5 days duration were carried out at two and 4 weeks of age, while 4-day experiments were used with older chickens. Since the same diet was fed throughout the entire investigation and feed intake was unrestricted, no preliminary period was necessary except for the 26-week old pullets which were removed from floor pens to batteries 40 MUELLER, BOUCHER AND CALLENBACH at 25 weeks of age. To avoid errors caused by the diurnal variation in chromic oxide content of the excreta (Mueller, ’56) representative samples of all excreta voided during the entire experimental periods of 4 or 5 days were used for chemical determinations. ‘ Droppings were collected on glass plates and transferred at 8 a.m. and 4 p.m. to procelain dishes after- feathers and the larger skin particles had been removed. The samples were then dried at a temperature of approximately 160° F. Kjeldahl-nitrogen determinations on wet and dry excreta showed that nitrogen losses during drying were negligible. After cooling, the excreta samples were ground and stored in glass jars. Kjeldahl-nitrogen, ether extract, fiber, ash and dry matter were determined according to the official A.O.A.C. methods ’( ’50). Chromic oxide was determined by the method described by Schurch et al. ( ’50) as modified by Dansky and Hill ( ’52). Nitrogen content of the urine-free feces was determined according to the method of Stotz ( ’31). A direct method was selected to avoid errors caused by eventual changes of the urine composition with increasing age. For calculation of the organic substance (dry matter minus crude ash) of the urine-free feces the following formulae were used: organic substance of feces = organic substance of excreta — organic substance of urine, where: organic sub­ stance of urine — 3.0 X nitrogen of urine, according to Diakow ( ’32). The fat content of the urine-free feces was calculated by the following formulae: fat of feces = fat of excreta -— fat of urine, where: fat of urine = 0.032 X organic substance of urine, according to Engler ( ’33).

RESULTS AND DISCUSSION Digestibility of principal nutrients Analysis of variance showed that for all nutrients the differences between sexes were small and insignificant. There­ NUTRIENT UTILIZATION BY CHICKENS 41 fore, the results of male and female groups were pooled for further study of the influence of age on digestibility and, if not mentioned otherwise, “ average digestibility” designates in the following the average of the pooled male and female groups. The greatest differences between digestion coefficients at different ages.were found for fiber and protein (table 2).

TABLE 2 Digestion coefficients for principal nutrients and metabolizable energy at different ages1

COMPOSITE COEFFICIENT OF VARIATION FOR MEANS AGE IN WEEKS —------2, 4, 7, 11 26 Weeks "Weeks

2 4 7 11 26 % % Organic substance 76 78 76 75 73 0.33 0.46 Protein 87 89 86 84 80 0.42 C.59 Ether extract 92 94 94 93 87 0.44 C.63 Fiber 4 3 — 6 — 14 — 8 32 45 Nitrogen-free extract 74 78 76 75 75 0.35 0.49 Metabolizable energy in percentage of gross energy of diet 74 77 74 72 70 0.32 0.46 Metabolizable energy on a maintenance basis in percentage of gross energy of diet2 71 74 72 71 69 0.28 0.40

1 The values fcr two to 11 weeks of age! are the sex-weighted averages of 5 groups of males and 5 groups of* females. Those for 26 weeks are the averages of 5 groups of females. 2 For explanation, see text, page 44.

The fiber digestion coefficients were positive at two and 4 weeks of age and negative at 7, 11 and 26 weeks. As shown earlier (Mueller, ’56) the accuracy of the chromic oxide indicator method increases with the digestibility of the nu­ trient under examination. It is also possible that the nega­ tive fiber digestibility at 7, 11 and 26 weeks of age was caused to some extent by the following: (a) a partial neutral­ 42 MUELLER, BOUCHER AXD CALLENBACH ization of the sulfuric acid used in the fiber determinations by the increasing amount of ash in the excreta or (b) adhesion of dust to the excreta. Protein digestibility increased from two to 4 weeks of age and gradually declined thereafter. A statistical analysis showed that the average protein digestibility at two weeks of age was not significantly different from the average pro­ tein digestibility at 7 weeks. With this exception, the average protein digestibility at any age was significantly different (5% level) from the average protein digestibility at all other ages. A critical study of different methods used for the deter­ mination of nitrogen in urine-free poultry feces by Thomann ( ’52) showed that of all methods examined the one developed by Stotz ( ’31) gave the highest recovery. Howmver, as found by Stotz, this recovery amounts to only 86 to 95% of the urine-free fecal nitrogen. On the other hand, Stotz ( ’31) found that during the nitrogen determination according to his procedure uric acid is completely removed from the sample. On the basis of these studies it seems probable that the protein digestion coefficients presented here are generally somewhat too high. Since the same method was used for all experiments this should not affect the differences between digestion coefficients for various ages. Based on the studies mentioned above, it is also improbable that the decrease in protein digestibility with advancing age was caused by an incomplete removal of the increasing amounts of urinary waste products (table 3). It is possible that the decline in protein digestibility with advancing age was to some extent only “ apparent,” i.e., caused by an increased excretion of metabolic nitrogen. Since, as stated by Maynard ( ’47), the excretion of metabolic nitrogen is most appropriately assessed against digestion, this possibility should not affect the inter­ pretation of our results to any appreciable extent. For the calculation of digestibility of the ether extractable material two different methods were used. For the values in table 2 corrections were made for the amount of such material NUTKIENT UTILIZATION BY CHICKENS 43 in urine, using formulae given before. Corresponding un­ corrected values, arranged in order of increasing age, were : 89, 91, 90, 88 and 82%. The trend in the two sets of data is similar. A statistical analysis showed that only the average ether extract digestibility at two weeks of age and that of females at 26 weeks of age were significantly different from the values for other ages. No significant differences were found between the ether extract digestibilities at 4, 7 and 11 weeks of age. These results are in contradiction to those of Whitson et al. ('43) who found a significant increase in ether extract

TABLE 3 Nitrogen balance and crude ash retention at different ages 1

AGE IN WEEKS

2 4 7 1 1 2 6 Percentage of nitrogen intake found in feces 13 il 14 16 20 found in urine 36 37 54 63 67 retained 51 52 32 21 13 Percentage of crude ash intake retained 32 32 16 3

1 The values for two to 11 weeks of age are the sex-weighted averages of 5 groups of males and 5 groups of females. Those for 26 weeks are the averages of 5 groups cf females.

digestibility from 8 to 12 weeks of age. As mentioned by these authors, it is possible that the differences found in their experiments were at least partially caused by some variation of environment. The digestibility of nitrogen-free extract showed a trend similar to the digestibility of the other nutrients. However, only the average digestibility at 4 weeks of age was signifi­ cantly different from the average at other ages. The digestibility of the organic substance increased from two to 4 weeks of age and decreased thereafter. All of the differences between the average digestion coefficients of suc­ ceeding ages were significant. 44 MUELLER, BOUCHER AND CALLËXBACH

Metabolisable energy The portion of the gross energy intake that was metaboliz­ able at various ages was calculated from the digestible nutrients using factors given by Fraps et al. ( ’40). As was shown by these authors there is a very close agreement between the metabolizable energy thus calculated and the values found by actual determination in the bomb calorimeter. Thus, for maintenance, one gram of digestible protein has a metabolizable energy of 4.4 Cal. This value is found by subtracting from the gross energy of one gram of protein the energy content of the corresponding urinary waste prod­ ucts. During growth however only part of the digestible protein is used for energy production, the remainder being used for tissue formation. From this latter fraction very little, or no, waste products are obtained. Therefore accord­ ing to a proposition of Fraps et al. (’40), the two following factors were used for calculating the metabolizable energy of digestible protein: 5.7 Cal. per gram of digestible protein retained within the body and 4.4 Cal. per gram of digestible protein used for energy production. Data on protein retention found in the present experiments are given in table 3 and will be discussed later. To show the influence of tissue formation on metabolizable energy, table 2 presents, besides values for metabolizable energy, also those for “ metabolizable energy on a maintenance basis” (the amount of metabolizable energy found if a factor of 4.4 Cal. is used for all digestible protein). As for the digestibility of the principal nutrients, only insignificant differences were found between the values for metabolizable energy of male and female groups. On the other hand, the average metabolizable energy at any age was, except for the comparison of two with 7 weeks, significantly different from the average metabolizable energy at all other ages. If the metabolizable energy was reduced to a mainte­ nance basis by using a factor of 4.4 Cal. for all protein, much smaller differences between ages were found. This seems NUTRIENT UTILIZATION BY CHICKENS 45 to indicate that the retention of protein for tissue formation was a main cause of the differences in metabolizable energy among various ages.

Protein retention The percentages of nitrogen intake that were retained within the body at different ages, as well as the percentages excreted as waste products in urine and feces, are given in table 3. The nitrogen excretion in feces at various ages reflects the

TABLE 4 Gain in weight, feed consumption, protein and crude ash retention per day, and protein content of gain in weight

AGE IN WEEKS SEX 2 4 7 11

Gain in weight per males 9 .6 1 2 .5 2 9 .9 1 8 .8 chick per day, gm females 9.0 10.6 21.9 9.9 Feed consumption per males 18.1 28.1 76.9 90 0 chick per day, gm females 17.2 26.4 56.1 66.5 Protein retention per males 2.0 3.3 5.4 4.4 ehick per day, gm females 1.9 2.8 4.0 2.8 Protein content of males 20.8 26.4 18.1 23.4 100 gm gain, gm females 21.1 26.4 18.3 28.3 Crude ash retention per males 0.31 0.54 0.63 0.28 chick per day, gm females 0.30 0.42 0.52 0.05 variation in protein digestibility discussed before. The per­ centage of nitrogen intake excreted as urinary waste products increased with advancing age, whereas with the exception of a small increase from two to 4 weeks, the percentage of nitrogen intake retained within the body decreased steadily with advancing age. Table 4 presents data on the quantity of protein retained per day and the protein content of the gain in weight. The quantity of protein retained per day increased gradually from two to 7 weeks and decreased from 7 to 11 weeks. As may 46 MUELLER, BOUCHER AUD CALLENBACH be seen from table 4 the protein retention per day and the gain in weight per day followed a similar pattern. The protein content of the gain, however, was not constant but varied rather irregularly from experiment to experiment. This is probably explainable by the complex composition of gain in weight. Turner (’47) and Chueng-Shyang Ma ( ’54) suggested that the male hormones may inciease nitrogen retention within the-body. In the experiments reported here the differences between percentages of feed protein retained by male and female chickens were not significant. This observation seems to indicate that the faster growth rate of male chicks was made possible by a greater feed consumption and not by a better utilization of dietary protein. The fact that, irre­ spective of rather great differences between body weight of male and female chickens, the protein utilization at a definite age was approximately the same for both sexes raises some interesting questions regarding the mechanisms controlling protein utilization. An attempt was made to use the data on protein retention at different ages to estimate the minimum protein requirement for the type of growth found in these experiments. The main difficulty consisted in estimating the quantity of protein used for maintenance. Mitchell et al. ( ’31) estimated the protein maintenance requirements during the earlier stages of growth by extrapolating a curve for endogenous nitrogen excretion given by Ackerson et al. ( ’23) for 6- to 12-month-old Rhode Island capons. Present estimates were based on the finding of Smuts ( ’35) that, regardless of species, 2 mg of endogenous nitrogen are lost daily for every calorie of basal heat. As found by Ashworth and Brody ( ’33) in experiments with rats and by Sorg-Matter (’28) in experiments with chickens, age does not affect this relationship between en­ dogenous nitrogen and basal heat. The basal heat production at various ages was calculated from the average weight of our experimental groups and the lowest values for basal NUTRIENT UTILIZATION BY CHICKENS 47 energy production at the nearest age as found by Barott and Pringle (’46). The results of the estimates are given in table 5. The quantity of crude protein required per day increased quite rapidly from two to 7 weeks of age and decreased slightly thereafter. The somewhat higher value for pullets at 26 weeks ot age -was probably caused by an average egg produc-

TABLE 5 Estimated protein requirement of male and female New Hampshire chichens at various ages and body weights

PROTEIN REQUIREMENT AGE AV. WEIGHT ■ PER CHICK Mainte­ Digestible nance Growth Total protein 1 Crude protein in % wk. gm gm gm gm gm gm/day of diet intake Males 2 109 0.20 1.98 2.18 3.11 3.57 19.7 4 33C 0.59 3.31 3.90 5.57 6.26 22.3 7 954 1.42 5.36 6.78 9.69 11.31 14.7 11 1746 2.04 4.37 6.41 9.16 10.93 12.1 Females 2 104 0.19 1.91 2.10 3.00 3.46 20.1 4 294 0.52 2.85 3.37 4.81 5.43 20.6 7 767 1.14 3.99 5.13 7.33 8.44 15.0 11 1291 1.51 2.84 4.35 6.21 7.36 11.1 26 2340 2.00 3.042 5.04 7.20 9.02 11.3

1 Estimated productive value of protein = 0.7, according to Eraps and Carlyle ( ’39). 2 Growth plus egg production.

tion of about 40%. When the protein requirements per day were expressed in percent of the feed consumption per day a slight increase from two to 4 weeks of age and a rapid decline from 4 to 11 weeks of age were found. The data on the estimated protein requirement for growth and maintenance to the age of 7 weeks agree fairly well with the values found by Mitchell et al. ( ’31) for White Plymouth Rocks of corresponding live weight. For older chickens the values found in the present experiments are lower, probably 48 MUELLER, BOUCHER AND CALLENBACH because of the different methods used to estimate the main­ tenance requirements.

Crude ash retention The data on crude ash ref ention by male and female chickens at various ages are summarized in tables 3 and 4. As for protein, the quantity of crude ash retained per day increased from two to 7 weeks of age. At 11 weeks ‘of age the crude ash’ retention was very small. Also the data on the percentage utilization of the crude ash intake show a pattern similar to that of protein utilization (table 3). Again the main difference consists in the much more rapid decline of crude ash retention. At 11 weeks of age crude ash intake and crude ash excretion were nearly in balance. This seems to indicate that at this age bone formation was practically complete and that minerals were needed principally to replace the endogenous excretion.

SUMMARY Digestibility and retention of nutrients were determined on 5 groups of 15 male and 5 groups of 15 female New Hampshire chicks at two, 4, 7 and 11 weeks of age. At 26 weeks of age an additional experiment was carried out with the 5 groups of females after the number had been reduced to 10 pullets per group. Chromic oxide was used as an indicator of digestibility and a single diet, containing approx­ imately 21% of protein, was fed throughout the entire in­ vestigation. Only insignificant differences between male and female chickens were found for the digestibility of organic substance, protein, fat, fiber and nitrogen-free extract. On the other hand, the differences in digestibility of these nutrients at various ages were in most instances significant. The pattern was similar for all nutrients: after a slight increase from two to 4 weeks of age the digestibility generally declined steadily. Only the variation of protein digestibility may be of NUTRIENT UTILIZATION BY CHICKENS 49 some practical importance. For all other nutrients tlic dif­ ferences between digestibility at various ages were too small to be important for the formulation of feeding standards. For metabolizable energy a slight increase from two to 4 weeks of age and a steady decrease from 4 to 11 or 26 weeks of age were found. It appeared probable that this variation was principally a reflection of differences in protein retention at various ages. With the exception of a slight increase from two to 4 weeks of age, the percentage of nitrogen retention steadily decreased with advancing age. The differences between percentage nitrogen utilization of males and females were insignificant. The quantity of protein retained per day increased gradually up to the age of 7 weeks and decreased thereafter, showing a variation similar to that of the gain in weight per day. On the basis of the experiments reported the following crude protein requirements were estimated: at two weeks 20%, at 4 weeks 21%, at 7 weeks 15%, at 11 weeks 12%, and at 26 weeks 11%. The percentage retention of crude ash declined rapidly after 4 weeks of age. At 11 weeks crude ash intake and crude ash excretion were nearly equal.

LITERATURE CITED

A c k e r s o n , C. W., M. J. B u s h a n d F. E. M u s s e h l 1923 The endogenous metabolism of hens and eapons. Poultry Sci., 2: 189. A s h w o r t h , U . S., a n d S. B ro d y 1933 Growth and development with special reference to domestic animals. XXVII. Endogenous urinary nitrogen and total creatinine excretion in rats as a function of dietary protein level, time on nitrogen-free diet, age, body weight and basal meta­ bolism. Missouri Agr. Exp. Sta. Res. Bull., 189. A. O. A. C. 1950 Official methods of analysis of the Association of Official Agricultural Chemists. Seventh ed.

B ar o tt, II. G., a n d E. M. P r in g i.e 1946 Energy and gaseous metabolism of the chicken from hatch to maturity as affected by temperature. J. Nutrition, 81: 35. Cr a m p t o n , E. W., a n d F. W h it in g 1943 The digestibility of typical eastern Canadian feeds by market bacon hogs. Sci. Agr., 28: 518. 50 MUELLER, BOUCHER AND CALLENBACH

Dansky, L. M,, and F. W. Hill 1952 Application of the chronic oxide in­ dicator method to balance studies with growing chickens. J. Nutrition, 47: 449. D i a k o w , M. J. 1932 Untersuchungen ueber Verdaulichkeit, Stoff-und Energie­ wechsel bei Huehnern als Grundlage fuer die rationelle Fuetterung des Gefluegels. Archiv fuer Tiei ernaehrung und Tierzucht, 7: 571. E ngler, H. 1933 Quantitative Verdauugsversuche am Haushuhn. Die Tierer- naehrung, 5: 329. F kaps, G. S., and E. C. Carlyle 1939 The utilization of the energy of feed by growing ehiekens. TexaS Agr. Exp. Sta. Bull., 571. F raps, G. S., E. C. Carlyle and J. F. F udge 1940 Metabolizable energy of some chicken feeds. Texas Agr. Exp. Sta. Bull., 589. Chueng-Shyang M a 1954 Mechanism of difference in growth rate between cockerels and pullets Poultry Sei., 33: 1028. Maynard, I,. A. 1947 Animal nutrition, McGraw-Hill Book Co. Inc., 2nd edition, page 105. M itchell, H. H., L. E. Card and T. S. Hamilton 1931 A technical study on the growth of White Leghorn chickens. Illinois Agr. Exp. Sta. Bull., 367. Mueller, W. 1956 Feasibility of the chromic oxide and lignin indicator method for metabolism experiments with chickens. J. Nutrition, 58: 29-36. Schürch, A. F,, L. E. Lloyd and E. W. Crampton 1950 The use of chromic oxide as an index for determining the digestibility of a diet. Ibid., 41 : 629. Smuts, D. B. 1935 The relation between the basal metabolism and the endogenous nitrogen metabolism with particular reference to the estimation of the maintenance requirement of protein. Ibid., 9: 403. Sorg-Matter, H. 1928 Loi quantitative de la dépense azotée minima des homéothermes : validité intrnspecifîque. Arch, internat, physiol., 30: 126. Stotz, H., 1931 Ein neues Verfahren zur Bestimmung der Verdauungskoef­ fizienten fuer Rohprotein beim Gefluegel. Archiv fuer Tierernaehrung und Tierzucht, 7: 29. Thomann, W. 1952 Verdauungsversuche mit Kleegras und Luzerne beim Haushuhn. Einfluss des Wachstumsstadiums und der Aufbereitung. Diss. Zollikofen-Bcrn. T u r n e r , C. W. 1947 Dried lactating cow manure in the ration of growing chickens. Poultry Sei., 36: 143. W atson, C. J., J. A. Campbell, W. M. Davidson, C. H. R obinson and G. W. Muir 1943 Digestibility studies wdth swine. 1. The digestibility of grains and concentrates at different stage of the growing and fattening period. Sei. Agr., 33: 708. W hitson, D., C. W. Carrick, E. E. R oberts a n d S. M. Hauge 1943 Utilization of fat by chickens — a method for determining the absorption of nutrients. Poultry Sei., 33: 137. W olf, E. 1873 Landwirtschaftliche Jahrbuecher, s : 221. EFFECT OF OPTICAL ANTIPODES OF 6-HYI)EO^Y NORLEUCINE (HEXAHOMOSERINE) ON GROWTH AND HEMATOPOIESIS IN RATS 1

EDWIN T. MERTZ AND JULIUS GOLUBOW2 Department of Biochemistry, Purdue University, Lafayette, Indiana

(Received for publication September 6, 1955)

INTRODUCTION Iii a previous report (Mertz et al., ’50) it was shown that the growth rate, but not the red blood cell count, of young white rats on a lysine-supplemented zein diet, was reduced Avhen the diet contained 0.08 to 0.33% of 6-hydroxy-DL-norleu- cine (hexahomoserine). At a level of 1%, however, growth was almost completely inhibited and the animals developed an orthochromic type of anemia (Merck Manual, ’40). In order to determine whether the growth-inhibiting and anemia- producing effects of the racemate are caused by both enanti- omorphs, 6-hydroxy-DL-norleucine has been resolved and the isomers fed to young white rats. The isomers were found to have different effects.

EXPERIMENTAL Resolution of Ivydroxy-DL-norleucine. The asymmetric en­ zymatic hydrolysis devised by Greenstein et al. ( ’53), was used in the studies reported here. The synthesis used by Birnbaum and Greenstein ( ’53) for the chloroacetyl deriva- 1 Journal Paper no. 889, Purdue Agricultural Experiment Station. The experi­ mental data in this paper are taken from a thesis submitted by Julius Golubow in partial fulfillment of the requirements for the degree of Master of Science in Biochemistry, and served as the basis for his receipt of the 1955 Purdue Biological Society Graduate Student Award. 2 Present Address: Department of Biochemistry, University of Pittsburgh.

51 52 EDWIN T. MERTZ AND JULIUS GOLUBOW five of DL-liomoserine was employed. One hundred and forty- seven grams of 6-hydroxy-DL-norleucine, synthesized from 5-S-hydroxybutyl liydantoin 3 by the method of Gaudry ( ’48), was treated with chloroacetyl chloride and sodium hydroxide in the cold to give crude chloroacetyl-6-hydroxy-DL-norleucine. It is our experience that the crude acyl derivative must be recrystallized from boiling ethyl acetate within 24 hours or no satisfactory product can be obtained. A n average yield was 82 gm of pure chloroacetyl-6-hydroxy-DL-ncrleucine (36.8% of theory). The melting point was 94°C. (corrected); calculated for C8HU04NC1: N, 6.3%. Found: N, 6.4%. Preparation of the enantiomorphs of hydroxy norleucine. Forty-one and one-tenth grams of chloroacctyl-6-hyd roXy-DL- norleucine were treated with 60 mg of acylase (Birnbaum et al., ’52) until the Van Slyke ninhydrin-carbon dioxide titri- metric test (Van Slyke et al., ’41) revealed 50% hydrolysis.4 Ilydroxy-L-norleucine was precipitated from a concentrated aqueous solution by the addition of alcohol. A total of 9.4 gm of 6-hydroxy-L-norleucine was obtained (63%' yield); [a] for a 2% solution in 6 N HC1 was + 2 5 ° , melting point, 265°C. (corrected). The chloroacetyl-6-hydroxy-D-norleucine was chemically hy­ drolyzed to give 6-hydroxy-D-norleucine (Berlinguet and Gaudry, ’52). A total of 6.0 gm of 6-hydroxy-D-norleucine was obtained (41% y ield ); [a]2^ for a 2% solution in 6 A HC1 was — 24.5°, melting point, 258°C. (corrected). Berlinguet and Gaudry ( ’52) obtained a specific rotation of — 23.2° (at 23°C.) and a melting point of 248°C. for the D-isomer, and a specific rotation of + 23.7° at 27°C. for the n-isomer. Optical purity of the enantiomorphs. The amino acid oxi­ dase method of Meister et al. ( ’51), was employed to test the enantiomorphs of 6-hydroxy norleucine for possible contam- 3 Supplied through the courtesy of E. I. du Pont de Nemours and Company. 4 One millilitter aliquots were taken, from the digestion mixture at various times, the reaction being stopped by the addition of 7 ml of tungstic acid so­ lution. The hydrolysis was determined on 3 ml of the resulting solution. A zero time aliquot served as the control. GROWTH ON HYDROXY NORLEUCINE 53 ination with the opposite isomer. Naja naja venom 5 was Used as the source of L-amino acid oxidase, and a hog kidney ex­ tract (Krebs, ’35, and Krebs and Bender, ’50) served as the source of D-amino acid oxidase. When 1C00 pM of the l - or D-isomer were used, no contam­ ination with the enantiomorph was observed. Test vessels to which 1 ¡jM of- the labile isomer was added showed a definite oxygen uptake. Therefore, the enantiomorphs contained less than 0.1% of the optical antipode. Feeding experiments. Weanling white female rats from the stock comny were divided into 8 experimental groups of 4 each. Each litter was distributed as evenly as possible, one rat to each group. The rats were housed in individual cages with wire screen floors in an air-conditioned room maintained at 23 °C. and 40% relative humidity. Only 4 rats were used in each lot because of low yields of the optical antipodes which limited the amounts available for study. Lots I, II and III duplicated a previous feeding- test (see footnote 6) with con­ firming results. The rats were fed a ration containing, per 100 gm, casein 19.0, sucrose 29.8, dextrin 37.75, minerals (Heinemann et al., ’46) 5.2, cottonseed oil 5.0, cod liver oil (Nopco X X ) 1.0, cellu- flour 2.0, vitamin B12 2.2 gg (0.1% trituration with mannitol), and vitamin supplement (Manna and Hauge, ’53) 0.25. The ration contained 1.2% of L-lysine based on microbiological as­ say of the casein. Hydroxy norleucine replaced an equivalent amount of dextrin in 7 experimental groups. The control group received no hydroxy norleucine. Food and water were given ad libitum. The animals were weighed every 7th day. On the 28th day, red blood cell counts were made on blood from the tip of the tail, and the experi­ ment was terminated.

RESULTS The effect of the racemate and its enantiomorphs on the rate of growth and red blood cell count of weanling rats is sum- s Ross Alien Reptile Institute, Silver Springs, Florida. 54 EDWIN T. MERTZ AND JULIUS GOLUBOW

aSi- CO » « S f « fc «S Ph o ° ^

e H O g ta

S3 O O iO Cd o rJH CO CO bß t H CO CO Cd CO CO O O o cd o CD o o

© ?=i SSg bß Ci I>- Ci 00 Cd r—' CO ci 00 ci oc td vd ci Düäcg f c H H* 00 Cd Ht- S3W « E

TtH Ci CO iD cd rH - 7 3 ) 1 <¡<3 2 I 1 T T3 « b 2 Cd Cd o o § BOS cd vo vo o CO « ► Bö range in total gains. <3 CO vq vq vd ts Ci id ö CD vo CO fro

5=1

S3 S3 g Q G G Q Q SI S3 g f g 5 o LO t-i o g g q q vq q q q bß cd ci ci o cd g w | r"'1 r_i h K SS

rOg £ CO£ i> î> i> H 0 GROWTH ON HYDROXY NORLEUCINE 55 marized in table 1. The control animals made an average total gain (134 gin) which was double that of the control animals receiving a zein diet in the previous publication (Mertz et al., ’50). The control diet has therefore been improved. The in­ corporation of 0.5% of hydroxy-DL-norleucine in the casein diet reduced the» rate of gain to about 70% of that of the con­ trol group, without affecting food efficiency, and with only a minor drop in red blood cell count. W hen the racemate was increased to a level of 1.0% , the average total gains and aver­ age total red blood cell counts dropped to approximately one- half of that of the controls. A further increase of the racemate to the 2% level reduced average weight gains and red blood cell counts, but not to the extent expected. Unpublished data 6 indicate that the toxic effects of hydroxy-DL-norleucine level off between the 1 and 2% concentrations when a casein diet is used. This is in contrast to our previous findings with a zein diet, where a level of 1% of the racemate caused almost complete inhibition of growth (Mertz et al., ’50). The nature of the diet is therefore an important factor in determining the degree of growth inhibition caused by this amino acid. The incorporation of 0.5% of hydroxy-D-norleucine in the casein diet, caused less growth inhibition than an equal amount of the racemate, had no affect on food efficiency, and had an insignificant effect on the red blood count. When the level of the D-isomer was raised to 1 % , however, average weight gains were reduced sharply to about one-fourth of that of the controls, and there was a drastic reduction in food effi­ ciency. In spite of these marked changes in weight gains and food efficiency, no significant drop in red blood cell count oc­ curred. A level of 0.5% of hydroxy-L-norleucine in the casein diet had a more profound effect on growth, food efficiency, and red blood cell count than the same level of either the racemate or the D-isomer. Raising the level of the L-isomer further to 1.0% did not reduce the average total gains or the food effi­ ciency below those of the group receiving 0.5% of the L-isomer, 6 Mertz, E. T., and Golubow, J., unpublished data. 56 EDWIN T. MEETZ AND JULIUS GOLUBOW but did cause a drastic reduction in the average red blood cell count. Since the average weight gain of the animals on the 1% level of L-isomer was 47% of that of the controls at the end of the first week (14 versus 30 gm), the reduction in growth rate could not be a secondary effect of the anemia, which develops more slowly. Inasmuch as the d - and L-isomers of hydroxy norleucine produce different responses, which are related to the specific levels fed, one must conclude that the response to the race- mate must be the final result of the separate actions of the d - and L-isomers when fed simultaneously. However, the final result obtained with the racemate cannot be predicted by simply averaging the experimental values obtained with the separate isomers at their respective levels in the racemic mixture. If this were the case, 1.0% of hydroxv-DL-norleucine should reduce the average gain per day to 67% of the con­ trols, and the average red blood cell count to 83% of the controls, instead of the observed values of 48.7 and 47.2% respectively. It is evident that the d - and L-isomers of hydroxy norleucine affect metabolism in different manners when fed together, and when fed separately. The finding that the D-isomer of hydroxy norleucine strongly inhibits growth at the 1 % level, but, unlike the l - isomer, does not decrease the red blood cell count to a sig­ nificant extent, was unexpected, and has some challenging implications. It is possible that the d - and L-isomers display different degrees of reactivity towards enzymes (or sub­ strates) concerned with the growth process and red blood cell production. This might account for the differences in growth inhibition potency with increasing concentration, and also for the differences in ability of the two isomers to reduce the red blood cell level. Inasmuch as the major toxic factor in deaminized casein is apparently 6-hydroxy-L-norleucine (Gingras et al., ’47), the pioneer findings of Hogan and CO workers (Hogan et al., ’34, ’36, ’41, and Guerrant and Hogan ’39) on “ deaminized casein anemia” in rats may be explained largely on the basis GROWTH OK HYDROXY XOBLEUCIXE 57 of the action of 6-hydroxy-L-norleucine as a growth depres­ sant and inhibitor of red blood cell formation. The toxic effects of 6-hydroxy-L-norleucine could also account for the nausea, dizziness, hypersensitivity to metallic sounds, and sharp rise in non-ketone urinary organic acids observed by Albanese et al. ( ’43) in human subjects who received a lysine- deficient diet containing hydrolyzed deaminized casein as the source of the protein. According to.Rose et al. ( ’55), these symptoms do not appear when humans are fed a lysine- deficient diet containing a mixture of purified amino acids as the protein source.

SUMMARY

6-Hydroxy-DL-norleucine (hexahomoserine) was resolved and the optical isomers fed to weanling rats. At the highest level fed (1% ), the D-isomer reduced the average weight gain to about one-fourth of that of the controls, without causing a significant reduction in the average red blood cell count. At this same level, the L-isomer reduced the average weight gain to one-half of that of the controls and the average red blood cell count to about one-third of that of the controls. It is concluded that the D-isomer is a specific growth depres­ sant, whereas the L-isomer depresses both growth and red blood cell production.

LITERATURE CITED

A m a n e s e , A. A., L. E. H o l t , J r ., J. E. F r a n k s t o n , C. N. K a j d i, J. E. B r u m - b a c k , J r . a n d D. M. W a n g e r in 1943 A biochemical lesion of lysine deficiency in man. Proe. Soc. Exp. Biol. Med., 52: 209-211. B e r l ix g u e t , L., a n d R. G au d r y J.952 Enzymatic resolution of DL-a-amino-e- liydroxy-n-eaproic acid and DL-a-aniino-5-hydroxy-n-valeric acid. J. Biol. Chem., 198: 705-769. B ir n b a u m , S. M., a n d J P. G r e e n s t e in 1953 Optical enantiomorphs of homo- serine, and liomocystine. Arch. Biochem. and Biophys., 42: 212-218. B ir n b a u m , S. Ml, L. L eyin to ay, R. B. K in g s l e y a n d J. P. Gr e e n s t e in 1952 Specificity of amino acid acylases. J. Biol. Chem., 194: 455-470. Gaudry, R. 1948 The synthesis of DL-a-amino-e-hydroxycaproic acid and a new synthesis of DL-lysine. Can. J. Research, B26: 387-392. G in g r a s , R., E. P age a n d R. G a u d r y 1947 The anemia-producing properties of a-amino-e-h"droxy-caproic acid. .1. Biol. Chem., 171: 831-832. 58 EDWIN T. MEKTZ AND JULIUS GOLUBOW

G r e i Ns t e in , J. P., S. M. B ir n b a u m a n d M. C. Ot e y 1953 Optical and enzy­ matic characterization of amino acids. J. Biol. Chem., 204: 307-320. G u e r r a n t , R . E., a n d A . G. H ogan 1939 Effect of amino acids on anemia caused by deaminized casein. J. Biol. Chem., 128: 363-372. H e in e m a n n , W. M ., M. E . E n s m in g e r , T. J. C u n h a a n d E . C. M cCu l l o c h 1946 The relation of the amount of thiamine in the ration of the hog to the thiamine and riboflavin content of the tissue. J. Nutrition, 31: 107-125. H o g a n , A. G., R. E. G u e r r a n t a n d W. J. R it c h ie 1936 -Additional observa­ tions on the anemia caused by deaminized casein. J. Biol. Chem., US: 659-672. H o g an , A. G., E. L. P o w e l l a n d R. E. G u e r r a n t 1941 Anemia from lysine deficiency in deaminized casein. Ibid., 137: 41-49. H o g an , A. G., a n d W. J. R it c h ie 1934 An anemia caused by deaminized casein. Ibid., 107: 179-189. K rebs, H. A . 1935 CXCVII. Metabolism of amino acids. III. Deamination of amino acids. Bioehem. J., 29: 1620-1644. K r e b s, H. A., a n d A. E. B en d er 1950 The oxidation of various synthetic a-amino acids by mammalian D-amino-acid oxidase, L-amino-acid oxidase of cobra venom and the L- and D-amino acid oxidases of Neurospora crassa. Ibid., 46: 210-219. M a n n a , L., a n d S. M. H au g e 1953 A possible relationship of vitamin B13 to orotic acid. J. Biol. Chem., 202: 91-96. M e is t e r , A., L. L e v in t o w a n d J. P. Gr e e n s t e in 1951 Optical purity o f amino acid enantiomorphs. Ibid., 192: 535-541. M e r c k M a n u a l , T h e 1940 Merck and Co., Rahway, N. J., 7th ed., p. 71. M e r t z, E. T., R, H. W a l t z , J r ., D . C. S h e l t o n , L. P . D o y l e a n d A . L. D elez 1950 Effect of different levels of hexahomoserine on growth and hematopoiesis in rats. Proc. Soc. Exp. Biol. Med., 73: 75-77. R o se, W. C., A. B o r m a n , M. J. Coon a n d G. P. L a m b e r t 1955 The amino acid requirements of man. X. The lysine requirement. J. Biol. Chem., 214: 579-587. V a n S l y k e , D. D., D. A. M a c F a y d e n a n d P. H a m il t o n 1941 Determination of free amino acids by titration of the carbon dioxide formed in the reaction with ninhydrin. J. Biol. Chem., 141: 671-680. THE QUANTITATIVE AMINO ACID REQUIREMENTS OF YOUNG WOMEN

i. t h r e o n i n e 1

RUTH M. LEVERTON,2 MARY R. GRAM,3 MARILYN CHALOUPKA,4 EILEEN BRODOVSKY 5 AND AMY MITCHELL “ Nebraska Agricultural Experiment Station, University of Nebraska, Lincoln

ONE FIGURE

(Received for publication May 25, 1955)

The present status of our knowledge of the qualitative and quantitative aspects of the amino acid requirements of man has been reviewed by Albanese ( ’50), Mitchell ( ’50), Rose ( ’52) and Leverton ( ’54). Qualitatively, it is well established that valine, leucine, isoleucine, threonine, methionine, lysine, phenylalanine and tryptophan must be supplied preformed to maintain the body in nitrogen equilibrium, a fact which has earned for them the classification of essential or indispensable amino acids. Other amino acids can be synthesized by the body in amounts sufficient to meet its needs and hence are classified as nonessential or dispensable. Quantitative requirements have been studied most extensively by Rose and his co-

1 Published with the approval of the Director as Paper no. 705, Journal Series, Nebraska Agricultural Experiment Station. 2 Present address: Division of Home Economics, Oklahoma A. & M. College, Stillwater. 3 Present add ress: Department of Home Economics, University of California, Berkeley. 4 Present address: School of Home Economics, University of Wisconsin, Madison. 5 Present address: Department of Biochemistry, School of Medicine, Vander­ bilt University, Nashville, Tennessee. “ Present address: University of Nebraska Press, Lincoln, Nebraska. 59 60 RUTH M. LEVERTON AND OTHERS workers, and tentative values for tire requirement of young- men for the 8 essential amino acids have been published by Rose ( ’49, ’52). In 1951, studies of the quantitative requirement of young- women for 5 essential amino acrds, threonine, valine, trypto­ phan, phenylalanine and leucine, were begun at the Uni­ versity of Nebraska.7 Results from the study of the threonine requirement are presented m this paper; also included are the general plan and methods used throughout all of the studies. Findings with respect to each of the other 4 amino acids will appear in separate and subsequent papers.

PART I — GENERAL PROCEDURE The method used in these studies was based on the fact that a negative nitrogen balance results when any essential amino acid is ingested in an amount insufficient to meet the body’s need. To determine the quantitative requirement it was necessary to find the least amount of each of the amino acids studied, which in the presence of adequate amounts of all others, would just maintain nitrogen equilibrium. The experimental procedure for determining the require­ ment for an amino acid was to feed each subject a ration adequate in the known dietary essentials and generous in all of the essential amino acids until she had been in nitrogen equilibrium for at least 7 days. Then the intake of the amino acid to be measured was reduced stepwise until the subject went into negative nitrogen balance for at least 4 days. Usually the intake was then increased stepwise until nitro­ gen equilibrium was again established. The least amount which maintained a subject in nitrogen equilibrium was re­ garded as her requirement. Five separate studies were conducted between September 1951 and April 1954. The sequence of these and the number of subjects are recorded in table 1. The major part of each time period was devoted to work with one amino acid. Follow-

1 Supported in part by a U. S. Department of Agriculture contract sponsored by tlie Human Nutrition Research Branch, Agricultural Research Service. Sequence of studies of requirements for five amino acids ° £ ° ¿L “ is £ m > « o ^ 0.15 to H O 8, 1951 AMINO rH 4b O n 73 .M OP > ia rH OP C

CD EURMENTS REQUIREM ACID CO -M Feb. O t a - f § I 43 nd -t-s § cq n3O d « ft ft O ¡3 ¡>> ft d 7, 1953 H GO ^ cq oa -+-> O CP o. ¡5 o. 7^> £ OP rH OP id co E ■+o c3 a CP «5 « ° £ h - -

P ^ o h Q d CP'd • 1 M 5 « .5 o bjO ¡>> SJ r a H +3 CP CZ) -O 2 .a d H ft e b f>* o rcJ m I fi PI c c S3 rJ d ft ft op QÌ e be be 5 d d m

4 Part of group A in phenylalanine study. 61 5 Designated as group C in every study. 62 RUTH M. LEVERTON AND OTHERS ing this, except in the first study, there was an ancillary study to repeat some level of an amino acid previously studied, or to try a level of one which was to he quantified later. After the requirements for these 5 amino acids had been studied separately, they were all fed, together with the other acids in the amino acid ration, in the least amounts which had been found capable of maintaining nitrogen equilibrium in all of the subjects. This wTas called the Test Mix ration. In’ this way the adequacy of the minimum amount of each of the 5 amino acids measured separately was tested in the presence of minimum amounts of the other 4.

Subjects Thirty-five healthy college girls served as subjects; one of them served in two studies. They ranged from 19 to 26 years of age, from 157 to 174 cm in height, and from 45.6 to 74.7 kg in weight. The girls lived in their usual college residences — dormi­ tory, sorority house, or private home. The research labora­ tories which included a dining room and a living room were their daytime headquarters. Meals were served at the same hours every day, 6:45 a.m., 12 noon, and 6 p.m., and the girls ate together. Much time and effort were devoted to keeping the mechanics of the routine from becoming burdensome to the girls and to making them feel well cared for, highly ap­ preciated, and conscious of their obligations as contributors to science. At the beginning and end of the studies, each subject had a medical examination including routine laboratory tests of blood and urine. The basal metabolic rate of each subject was measured at the beginning and at the end of the first three studies. Since no change in rate had occurred, it was measured only at the beginning of the last two studies. All subjects had basal metabolic rates within the normal range for their ages. AMINO ACID REQUIREMENTS 63

Preliminary periocl During the first days of each study, the subjects were given a weighed diet composed of ordinary foods. The kind and amount of food were the same each day, except for the amounts of fat and sugar which were adjusted to meet the caloric requirements of the individual subjects. This period gave the subjects time to adjust physiologically and psychologically to the experimental set-up, which was to be­ come more rigid and increasingly tiresome. The subjects usually came into nitrogen equilibrium during this period of controlled food intake, which varied in length from 14 days in the first study to 8 days in the last two studies.

Experimental period The experimental period included all the days following the preliminary period until the end of the study. The diet was semi-purified and included a limited number of foods which were extremely low in nitrogen, mineral and vitamin supplements, and purified amino acids. Amino acids. The amounts of the 12 amino acids, which were fed daily in the same ratio as they occur in egg protein (Mitchell and Block, ’46), and the nitrogen supplied by them are shown in table 2. The amino acids were tested for purity by nitrogen analysis, optical rotation and micro­ biological assay. The total nitrogen intake was 6.2 gm for the first threonine study, 8.5 and 9.5 gm for the valine study and for repeating certain levels of threonine, and 9.5 gm for the last three studies. Glycine or glycine plus diammonium citrate was the chief source of nitrogen, and made up the difference between the nitrogen supplied by the 12 amino acids and the total nitrogen intake. All of the amino acids except cystine, tyrosine and the one being tested, were ground together in a ball mill. The mixture was then analyzed for nitrogen to check its total nitrogen content against the sum of the results of the nitrogen analysis of each component. 64 BUTH M. LEVEBTON AND OTHEKS

Each, subject’s daily portion of this amino acid mixture was divided into three portions and fed: 25% for breakfast, 37.5% for lunch, and 37.5% for dinner. The appropriate amount for each meal was weighed into a tumbler and dis­ solved in water prior to mealtime. The amino acid being tested was in a separate solution and was added to the other acids. The diammonium citrate solution was also added to the dissolved acids.

TABLE 2 The purified amino acids fed in all studies

AMOUNT PRESENT IN : AMOUNT AMINO ACIDS FED IN 1 5 gm 2 0 gm T E S T M IX egg protein 1 egg p ro te in 2

gm/person/day gm,l person/day gm,/ person/day L-Arginine-HCl 1.162 1.549 1.549 T.-Histidine-HCl 0.389 0.519 0.519 L Lysine-HC1 1.350 1.800 1.800 L-Tyrosine 0.675 0.900 0.900 T.-Phenylalanine 3 0.945 1.260 0.200 s-Tryptophan 3 0.225 0.300 0.150 L-Cystine 0.360 0.480 0.480 r,-Methionine 0.615 0.820 0.820 L-Threonine 3 0.735 0.980 0.184 l.-Leucine 3 1.380 1.840 0.600 DL-Isoleucine 2.400 3.200 3.200 l -V aline3 1.095 1.460 0.550 Nitrogen content 1.430 1.910 1.583

1 Used in the major threonine study only. 2 Used in all other studies. 3 The amounts of these acids were varied during the studies of requirement.

Because of their relative insolubility, cystine and tyrosine were mixed with a portion of the butterfat fed at each meal. Foods. The basal ration included: (1) white sugar, corn­ starch, especially prepared butterfat and corn oil made into wafers and pudding, (2) butterfat, corn syrup and jelly served as spreads on the wafers, and (3) white sugar and lemon juice mixed with the amino acid solutions. Pure sugar candies and a bottled soft drink were used for additional calories. AMINO ACID REQUIREMENTS 65

Wafers or pudding or both supplied one-fourth to one-half of the calories in this semi-purified diet. The ingredients — cornstarch, sugar, butterfat, corn oil and agar flakes,8 plus the special baking powder and part of the mineral supplement — could be mixed with a small amount of water and baked into a crisp wafer, or cooked with a large amount of water into a soft pudding. When a subject became iired of chewing the wafers, she could choose to have the caloric equivalent in pudding. Extra wafers and pudding without the baking powder and minerals were used to help raise caloric intakes to individual needs. Small amounts of a few other foods — canned orange juice, tomato juice, grape juice, applesauce, peaches, pineapple, and raw carrots and lettuce •— were added to improve the palatability of the diet. All of these were not fed in any one study but the foods (plus the liver extract) which were fed in each study were assayed microbiologicallv for the amino acid being measured. The auxiliary foods used with the Test Mix were assayed for all 5 of the amino acids being measured. Small amounts of decaffeinated coffee and weak green tea, which were low in nitrogen, were optional. The auxiliary foods used in each study supplied less than 5% of the total daily nitrogen intake. The individual foods as well as the chemicals were always analyzed for nitrogen before being fed. Supplements. The composition and daily intake of the mineral supplements are shown in table 3. Half of the daily portion of the mineral mixture was baked in the wafers or cooked in the pudding, and the other half was dissolved in the lemon juice. The baking powder was used to make the wafers porous but could be added to the pudding when no wafers were used. One multi-vitamin capsule 9 was given daily and supplied the following: vitamin A 5000 TJSP units, vitamin D 500 USP units and thiamine 3,- riboflavin 2.5, nicotinamide 20, pyri-

s Winthrop Stearns, Inc. 11 Abbott Laboratories. 66 RUTH M. LEVERTON AND OTHERS doxine 0.5, pantothenic acid 5 and ascorbic acid 50 mg respec­ tively. Further supplements included vitamin Bi2 9, 20 ng daily, added to the amino acid solution at each meal; liver extract10, 5 ml containing 10 USP anti-pernicious anemia units, was given every third day during the first 4 studies. During the leucine study, however, 0.5 gm of liver concentrate 1: 2010 was given daily. Mixed tocopherol acetates 11, 38 I.U., were given every third day during the last two studies.

TABLE 3 Composition of mineral supplement

AMOUNTS PER SUBJECT PER DAY SALTS Salt Element gm gm Mineral mix CaC03 1.362 0.548 Ca k h 2p o 4 3.307 0.298 P MgC03 • Mg (OH) 2 • 3H,0 0.800 0.199 Mg FeC,H50t-6H20 0.0948 0.015 Pe CuS 04-5H20 0.0078 0.002 Cu KI 0.0002 0.00015 I MnCl2-4H30 0.0072 0.002 Mn ZnCl2 0.0019 0.0009 Zn

Total 3.58 Baking powder 1 NaHCO, 2.05 Ca(H2P04)2-H20 2.86 0.455 Ca 0.703 P Cornstarch 2.69

Total 7.60

' Rose, Johnson, and Haines ( ’50).

Caloric intake Much attention was given to preventing the subjects from either losing or gaining weight. They were weighed daily before breakfast and caloric intakes adjusted upward or downward as seemed necessary. The caloric intake was kept

10 Wilson Laboratories 11 Lederle Laboratories. AMINO ACID REQUIREMENTS 67

s high as possible without having subjects gain weight. The ictivity of the subjects was similar, none of them having strenuous exercise. Calculated daily caloric intakes ranged from 2200 to 2900, or from 35 to 54 Cal. per kilogram of actual body weight. The mean daily caloric intake for the 36 subjects was 2500 Cal. or 42 Cal. per kilogram.

Metabolism study and equilibrium zone Urine and feces were collected for the entire period of study, from the first day of the preliminary period through the last day of the semi-purified diet. Urine was acidified, except during the tryptophan study, and refrigerated under toluene until a 24-hour collection was complete. Nitrogen and creatinine determinations were done on the urine within 4 hours after the completion of each 24-hour collection. Carmine was given to mark the feces when the level of amino acid intake was changed. When one level was fed for more than a week, carmine was given at the end of every 7 days as well as when the level was changed. The fecal collections were frozen until combined into a composite for a given period. Then a digest was made with 20% HC1 for nitrogen analysis. Fecal nitrogen values did not vary greatly from period to period for the same subject. Values for the subject with the lowest excretion varied from 0.31 to 0.46 gm of nitrogen daily; for the subject with the consistently highest excretion the values ranged from 0.70 to 0.78 gm daily. The two subjects with the greatest variability in fecal nitrogen through­ out their studies had mean daily values which ranged from 0.42 to 0.72 gm, and 0.35 to 0.65 gm of nitrogen. Nitrogen was determined by the macro-Kjeldahl method (AOAC, ;50) and the ammonia was collected in boric acid (Markley and Hann, ’25). Creatinine was determined by the Folin method using picric acid (AOAC, ’50), and the intensity of the color was measured photometrically. Different subjects were not always on each level of amino acid intake the same number of days, but in general for 6 68 RUTH M. LEVERTON AND OlHERS days or longer. The subjects who showed the greatest daily variation in urinary nitrogen had to he kept on each level longer than the subjects who showed little variation. The time had to he reduced to less than 6 days in cases where the intake of an amino acid was low and symptoms of ill-health in the subjects were suspected, or when severe negative nitrogen balances resulted. The nitrogen balance, or che intake minus the excretion, was calculated for each subject each day as the study pro­ gressed. The cumulative nitrogen balance of each subject, as she continued on a given intake, was reviewed daily and used as the basis for deciding when to change the level of amino acid intake, and whether to increase or decrease it. Because of the time lag in securing the values for the fecal nitrogen, the figures from the preceding period were used temporarily to add to the analyzed urinary values. Extensive experience with human metabolism studies con­ vinces one of the inherent variations, both human and mechan­ ical, which are involved in this research technique. These can be reduced by rigid control, but they cannot be eliminated completely. Therefore, it seems more reasonable to these authors to define nitrogen equilibrium as a zone in which the excretion and intake closely approximate each other, rather than a single point at which they are numerically identical. In these studies of amino acid requirements, therefore, nitro­ gen equilibrium is defined as the zone in which the difference between the intake and excretion does not exceed ± 5%, i.e. the excretion is within 95 to 105% of the intake. Nitrogen loss from the body, ordinarily referred to as “ negative balance,” is considered to exist when the excretion is more than 105% of the intake. Conversely, nitrogen retention or positive balance exists when the excretion is less than 95% of the intake. Studying the amino acid requirements from the standpoint of the minimum amounts needed for maintaining nitrogen equilibrium is not meant to imply that nitrogen storage may not be important for young women in the age range of these AMINO ACID REQUIREMENTS 69 subjects. It seemed, however, that nitrogen equilibrium would iome closer to being a common denominator to the nitrogen netabolism of all of the subjects than would the storage of litrogen.

PART II — THREONINE The essentiality of threonine for maintaining nitrogen equi­ librium in human subjects was first mentioned by Eose et al. ( ’43). Later these investigators ( ’51) reported in detail the procedure and results of the qualitative study made of two male subjects on diets with and without threonine. With­ drawal of threonine from the diet caused the subjects to go into negative nitrogen balance, and replacing it brought them into equilibrium. Eose ( ’49) has presented tentative values for the quantitative threonine requirement : 0.50 gm of L-thre- onine as the minimum daily requirement and 1.0 gm as the recommended intake. The latter value was found to be satis­ factory for nitrogen balance in 24 men who were subjects for studies of the requirement for amino acids other than thre­ onine. The term “ recommended intake” has recently been changed to “ safe intake” by Eose, Lambert, and Coon ( ’54).

EXPERIMENTAL PROCEDURE As indicated in table 1, three groups of subjects were used for this study of threonine requirement. The details of the intakes of nitrogen, threonine, and auxiliary foods are shown in table 4. The composition of the semi-purified diet and the management of the metabolism study have just been described.

RESULTS The mean daily nitrogen balances of each subject on the various levels of threonine intake are charted in figure 1, and the performance of each subject may be followed as her intake was changed from one level to another. The solid line joins the means of the balances of all the subjects on each level and emphasizes the increase in nitrogen loss 70 RUTH M. LEYERTON AND OTHERS with decreasing intakes of threonine. The zones of equilibrium for the two levels of nitrogen intake are shown by the shaded areas. The data have been summarized in table 5 to show the mean daily nitrogen balance, standard deviation and range, and the number t>f subjects in negative balance for e'ach group of subjects on each level of threonine. “ Total number of subject-days” refers to the total of the number

TABLE 4

Nitrogen, threonine and food intakes of subjects

INTAKE GROUP A GROUP B GROUP C

Nitrogen, gm 6.2 8.5 and 9.5 9.5 Chief source of nitrogen glycine glycine glycine plus diammonium citrate

Amino acids 15 gm O 20 gm C= 20 gm egg protein egg protein egg protein

Threonine, nig Pure acid, L-form 735, 367, 550, 275, 184 275, 184 184, 73, 0 DL-form 147

Auxiliary foods,1 mean 30 30 30 range 28-31 28-32 29-32

Auxiliary 100 gm tomato juice 100 gm tomato juice 100 gm reconstituted foods 2 60 gm grape juice Choice of any two: frozen orange juice 100 gm apple juice 150 gm apple juice 75 gm peaches 75 gm applesauce 75 gm grape juice 75 gm pineapple All omitted during 120 gm applesauce 25 gm grape juice threonine levels of or pineapple 0 and 147 mg DL 50 gm raw lettuce or raw carrots

Subject-code 1, 4, 6, 21, 22, 23, 24, A, B, C, D, numbers : 7, 8, 9 25, '27, 28 E, P, G, H

'Assayed microbiologically using Leuconostoc citrovorum 8081. 2 All foods canned unless otherwise noted. AMINO ACID REQUIREMENTS 71 of days that the subjects were on a given level of intake. As explained earlier, the number of days that different subjects were on the same intake was similar but not identical. The threonine intakes include the amount fed as the purifiej synthetic amino acid plus the 30 mg supplied daily by the

1.0

27 0.75

0.50 * o u A m 2 7 N. 8?ÌlL: :ill :ji .'lib ili till: 0.25 E 0) in SJ Z < _i < a -0 .2 5 Z U1 0 -0 .5 0 O a

Z -0 .7 5

-1.00

765L 5801. 397L 305L 2I4L I03L 0 I47DL THREONINE INTAKE m g/day Fig. 1 Threonine intakes and nitrogen balances of 21 subjects for 467 subject- days. Each subject is identified by her code which appears beside her retention for each level on which she was studied. X = mean balance of all subjects on each level of intake. Intake includes 30 mg/day supplied by auxiliary foods. 72 RUTH M. LEVERTON AND OTHERS auxiliary foods and unless described otherwise refer to th L-form. The 6 subjects in group A were able either to maintain nitrogen equilibrium (excretion within 95 to 105% of the intake) or to retain nitrogen when the intake of threonine \Vas reduced from 765 to 397 mg and then to 214 mg. When the daily intake was further reduced to 103 mg of threonine, only subjects 6 and 7 remained in equilibrium. None of the subjects maintained nitrogen equilibrium when threonine was

TABLE 5 Nitrogen balance of subjects on different levels of threonine intalce

NUMBER TOTAL NITROGEN BALANCE NUMBER OF LEVEL OF NUMBER OF OF !SUBJECTS Range SUBJECTS IN L-THREONINE 1 ON EACH SUBJECT- Mean S.D.2 NEGATIVE LEVEL DAYS Low High BALANCE s mg/day gm/day gm/day gm/day Group A 6.2 gm N/day 765 6 55 0.10 0.21 — 0.10 0.51 0 397 6 32 0.06 0.22 — 0.28 0.33 0 214 5 25 — 0.11 0.21 — 0.30 0.16 0 103 6 42 — 0.49 0.26 — 0.76 — 0.12 4 0 6 66 — 0.81 0.19 — 1.05 — 0.53 6 147 dl (73.5 L) 5 32 — 0.73 0.21 — 1.07 — 0.52 5

Group B 8.5 or 9.5 gm N/day 580 7 96 0.14 0.32 — 0.20 0.78 0 305 3 18 — 0.04 0.11 — 0.17 0.03 0 214 5 34 — 0,21 0.41 — 0.71 0.35 2

Group C Test Mix 9.5 gm N/day 214 8 56 — 0.15 0.42 — 0.88 0.25 2 305 2 11 — 0.20 0.28 — 0.40 0.00 0

Groups A, B, and G 214 18 115 — 0.15 0.35 — 0.88 0.35 4

Groups B and C 305 5 29 — 0.11 0.18 — 0.40 0.03 0

1 Includes 30 mg supplied by the auxiliary foods. 2 Standard deviation. ’ Nitrogen excretion more than 105% of the intake. AMINO ACID REQUIREMENTS 73

omitted entirely from the ration or when 147 mg of the DL-form of threonine were fed. Because 214 mg of threonine was the lowest level on which all of the subjects in group A had maintained equilibrium, it was checked as a possible minimum daily requirement using the subjects in groups B and C. Of the 13 subjects subse­ quently studied on this intake, however, 4 subjects, 21, 23, C, and F could not maintain nitrogen equilibrium. In group B the subjects remained in nitrogen equilibrium on daily intakes of 580 and 305 mg of threonine, but when the intake of 5 subjects was lowered to 214 mg of threonine two were in negative nitrogen balance. The Test Mix supplied 214 mg of threonine daily to the 8 subjects in group C. Two of the subjects, C and F, were in negative nitrogen balance. How­ ever, when their intake of threonine was increased from 214 to 305 mg daily, they came into nitrogen equilibrium.

TABLE 6

Test of significance of differences between nitrogen balances

LEVELS OS' MEAN DEGREES OF L-THREONINE INTAKE DIFFERENCE N EiCEEDOM mg/day gm N/ day 0 vs 103 0.32 12 10 2.4611 103 vs 214 0.34 24 22 2.125 1 214 vs 305 0.04 23 21 0.235 305 vs 397 0.17 11 9 1.417 397 vs 580 0.08 13 11 0.500 580 vs 765 0.04 13 11 0.267 147 DL vs 0 0.08 11 9 0.667

1 Significant at the 5% level of probability.

The results of the “ t” test are given in table 6. The difference between the mean nitrogen balances on intakes of 214 and 103 mg of threonine is significant at the 5% level of probability; the same is true for intakes of 103 and 0 mg of threonine. Of the 21 subjects for whom nitrogen balance data are shown in figure 1, 16 were on levels of threonine intake which appeared to approximate a minimum requirement under the 74

TABLE ns ns •§ •§ ss I Q £ © h 2 2 S •«!g3 *■ *3« £ *3“ BE. ! <5í } * 3 O ¡* o . O H O na W < H i LVRO AD OTHERS AND LEVERTON . M H T U R :c' ■ ^ C CO CO O O i—rH i i—I \ ^ '■$ «Ä O O OI O lO 1© CO O i— 1 ’ o LO C— LO 1© LO o o o o o rH © m CO^ O) CO o 4 3 4 ri H 04 03 04 t>- o © a co h i> m rH 04* b- rH 4 0 i t>! in © H > CO td 3 H id rH 03 CD lO C3 'So-’S i - r £ 'S r—i04 O CO rH 04 o o 04 rH id io co CD GO O 04 O O 04 ! O CO 05 04 05 C5 r 4 0 oq o |> CO rH N O [in 04 04 rH 04 CO rH i-l O rflCO rt< ^1 IO O IO tj 'S? -1 m ID re 3i rl l 0 0 i 0 0 04 04 04 i— I 04 I 04 r-l Ol 03 i— I .si rH rH b- i— 1 rH 03 rH 04* 04 © © o’ O CO r-i CO lO rH J3 o ’ o cq ? « ? top s rH 04 °q h t rH cá O CO CO O rH 00 4 4 4 40 0 04 04 04 04 04 , 04 04 ID rH rH 04* CO 05 4 O i t> Ci O CO 04 <©*iow q wio ft O O © O 1-Í o o’ O 04 o o’ O o O © O 04 4 4 4 4 4 04 04 04 04 04 rH 04 rH rH rH ■^H rH 1 q t q M o pq ft pq<1 4 d id id 04 ^ H HO rH ^ ^ - - - i>. © t- t- b- H rH 04 CO CO 04 rH O 04 04 04 04 O

'N itrogen excretion within 95-105% of the intake. AMINO ACID REQUIREMENTS 75 conditions cf this study. The least amount of threonine -which was fed and on which these 16 subjects maintained nitrogen equilibrium is shown in table 7. To consider differences due to metabolic size, the quantity of threonine is expressed as mg/kg3/4 actual weight/day as well as the total daily intake. Also shown for each subject is the mean daily nitrogen balance for that period, together with the mean calorie intake and the creatinine coefficient for the entire time on the semi- purified diet. The intake of threonine necessary to maintain nitrogen equi­ librium ranged from a low of 4.5 mg/kg3/4/day for subject 6 to a high of 14.9 mg/kg3/4/day for subject F. There seemed to be no relationship between the amount of threonine needed and the metabolic size or creatinine coefficient. Despite careful observation, the subjects neither displayed nor reported any acute or specific symptoms which suggested an inadequate intake of a dietary essential, even when 5 subjects had no threonine for 10 days and one subject had none for 16 days. The subjects did not know their exact threonine intakes but they watched their daily nitrogen balance data with interest. They did not attribute fatigue or other minor ailments to the diet. Occasionally appetites lagged, or the food was “ hard to get down,” or a subject felt nauseated, but these were not severe enough to interfere with consumption of the daily ration of food and chemicals.

DISCUSSION There was considerable yariability among individuals in re­ spect to the minimum amount of threonine needed for nitrogen equilibrium. On a total daily basis the amount required varied from 103 to 305 mg, and on a weight basis, from 4.5 to 14.9 mg/kg:!/4/day. »Such a range suggests large individual differences in requirement, probably due to physiological and psychosomatic influence . Although no correlation was found between threonine requirement and metabolic size the pos­ sibility of a relationship still exists. 76 RUTH M. LEVERT0N AND OTHERS

In the present study it is possible that the level daily nitrogen intake may have had some influence on certain subjects ’ requirements for threonine for nitrogen equilibrium. Subjects 6 and 7 who needed only 103 mg of threonine daily for this purpose, were on the lowest daily nitrogen intake, 6.2 gm. Subjects C and F who needed the most, 305 mg daily, and subjects 21 and 23 who needed more than 214 mg of threonine had nitrogen intakes of 9.5 gm daily. It must not be-overlooked, however, that there were subjects on the lowest nitrogen intake and others on the highest intake who were in equilibrium on 214 mg of threonine daily. When 5 subjects were on a level of 147 mg of DL-threonine daily, none of them could maintain nitrogen equilibrium. This was not surprising because 4 of the subjects, 1, 4, 8, and 9, had needed more than 103 mg of the l form (fig. 1). For subject 7 103 mg was sufficient, 73 mg purified threonine plus 30 mg supplied by the food. However, when she was given 147 mg of the DL-form no auxiliary foods were added so the intake of the L-form dropped to 73.5 mg. From the results of this study, two different amounts can be discussed as a possible minimum daily threonine require­ ment for use beyond the limits of this study. The first is the mean requirement of all the subjects, and the second one is the smallest amount sufficient for nitrogen equilibrium for all of the subjects. The mean requirement of all the subjects under the conditions of this study was 214 mg of threonine. This figure is double the 103 mg which was sufficient to keep two of the 21 subjects in nitrogen equilibrium, and perhaps some of the subjects who were not studied on 103 mg threonine also could have maintained equilibrium on that amount. How­ ever, 214 mg of threonine was insufficient for 4 of the subjects. Because an individual does not benefit from the mean, provi­ sion should be made for those who rec uire more. The second amount for consideration, therefore is the more generous intake of 305 mg of threonine daily; on this intake every subject was in the zone of nitrogen equilibrium. This intake AMINO ACID REQUIREMENTS 77 will meet the needs of many more young women than will 214 mg. This daily intake of 305 mg of threonine which prevented negative nitrogen balance in all of the subjects included 275 mg of purified, synthetic L-threonine and 30 mg of threonine supplied by the food. Little is known of the comparative physiological availability and utilization of L-threonine from synthetic and from food sources. The quantitative require­ ment may be different when threonine and the other amino acids are supplied by food than when they are supplied as chemically pure acids. Until definite information is available on the relation of the source of an amino acid to an individual’s quantitative requirement for it, the amounts of threonine from the two sources used in this study are added together. The total is rounded from 305 mg to 310 mg and suggested as the minimum daily requirement of young women for threonine. This figure of 310 mg of L-threonine is lower than the 500 mg suggested by Rose ( ’49) as the minimum daily requirement for men. Recently Rose, Lambert and Coon ( ’54) have defined the minimum requirement of an amino acid as, “ the smallest amount capable of inducing distinctly positive bal­ ance.” In the present study of young women a different criterion for minimum requirement has been used, i.e. the amount of an amino acid which will keep a subject in a zone of nitrogen equilibrium, where the difference between the intake and excretion is within ± 5 % . This study differs from those of Rose and his co-workers in the source of dietary nitrogen, the amino acid, mineral, and vitamin intake, the age as well as the sex of the subjects, and probably in many other aspects such as the experimental levels of the threonine fed, the length of period of study and the number of subjects. The following reports in the literature suggest that diets in this country are likely to be adequate in threonine. Futrell et al. ( ’52) determined the threonine content of the self-chosen diets of 4 women and found that it ranged from 1.68 to 3.44 gm per day. Mertz et al. ( ’52) have reported a range from 0.9 78 RUTH M. LEVERTON AND OTHERS to 3,8 gm threonine per day in the self-chosen diets of 18 women. Diets of common foods planned by Wharton et al. ( ’53) for use in a study of nitrogen balance of 18 college-age women supplied from 2.6 to 2.9 gm of threonine daily. Allison ( ’53) has summarized figures from Block to show “ patterns of amino acids consumed by humans” in different parts of the world; the threonine intakes ranged from 3.8 to 4.3 gm/100 gm of protein. Thus a daily protein intake of only 10 gm would supply the amount of threonine found to be required by the subjects in the present study. Even with these indications of threonine requirement and threonine intake our information is far from complete. Much more needs to be known about the interrelationships among the amino acids and between the amino acids and other dietary items, of the rates of synthesis of amino acids by the body, and of the value of the so-called dispensable amino acids in protein nutrition and nitrogen balance. The chief use of figures for the quantitative requirements for the amino acids is in helping to identify possible limiting factors in human dietaries, especially those diets in which plants are the major source of protein. Knowledge of such limiting factors and of the amino acid composition of in­ digenous foods will provide the basis for planning how best to supplement the plant proteins within the framework and economy of the existing food patterns of the population groups.

SUMMARY AND CONCLUSIONS The general plan and methods of studies to determine the quantitative requirements of young women for the amino acids threonine, valine, tryptophan, phenylalanine and leucine have been described. For the study of each amino acid there was an experimental period of 6 to 7 weeks during which a semi- purified diet was used. This included a few foods low in nitrogen, mineral and vitamin supplements, and chemically

pure amino acids in the l form (except for DL-isoleucine). The intake of the amino acid being measured was reduced AMINO ACID REQUIREMENTS 79 stepwise until negative nitrogen balance resulted. Usually the intake was then increased until nitrogen equilibrium was established. Nitrogen equilibrium was defined as the zone in which the excretion of nitrogen was within 95 to 105% of the intake, rather than a point at which excretion and intake were numerically equal. The smallest intake of an amino acid which permitted all of the subjects to remain in nitrogen equilibrium was considered the minimum requirement. After the minimum requirement of each of these amino acids had been determined separately, its adequacy was tested in the presence of minimum amounts of the other 4 amino acids. The results with respect to threonine are reported also. The first group of 6 subjects was studied on several dif­ ferent intakes of threonine. A daily intake of 214 mg of threonine was sufficient to maintain two subjects in nitrogen equilibrium and 103 mg was sufficient for two other subjects. However, in a second group of 5 subjects who were on an intake of 214 mg of threonine daily, two were in negative balance. A third group of 8 additional subjects was then used to test the adequacy of 214 mg of threonine for maintaining nitrogen equilibrium when valine, tryptophan, phenylalanine, and leu­ cine were in the ration also in the minimum amounts. Six of these subjects remained in equilibrium; the other two subjects were in negative balance until their intakes of thre­ onine were increased to 305 mg daily. The smallest intake of threonine which permitted all of the subjects to be in nitrogen equilibrium was 305 mg daily. This amount is rounded to 310 mg and suggested as a tentative minimum daily threonine requirement of normal young women similar to those studied here.

ACKNOWLEDGMENTS The authors wish to express sincere appreciation to the students who served as subjects and to the many technical assistants who helped with all phases of the work. It was 80 RUTH M. LEVERTON AND OTHERS the cooperation and devotion of all of these persons which made the studies possible. Grateful acknowledgment is made for the Vita Kaps sup­ plied by Abbott Laboratories of Chicago, the butter supplied by Roberts Dairy of Lincoln, and the L-lysine hydrochloride supplied by E. I. Du Pont de Nemours and Comnany of Wilmington, Delaware.

LITERATURE CITED

A l b a n e s e , A. A. 1950 Protein and amino acid requirements of man. Protein and Amino Acid Requirements of Mammals, Academic Press Inc., New York City. 115-151.

A l l is o n , J. B. 1953 Amino acid requirements of man. Borden’s Review of Nutrition Research, XIV: 61-79.

A sso c ia t io n of Of f ic ia l A gricultural C h e m is t s 1950 Official and Tentative Methods of Analysis. Assn. Off. Agric. Chemists, Washington, D. C., 7th ed., a, 13. b, 368.

F u t r e l l, M. F., R. N. L u t z , M. S. R e y n o l d s and C. A. B a u m a n n 1952 Studies on amino acids in self-selected diets. J . Nutrition, 46: 299-311. L everton, R. M. 1954 The amino acid requirements of man. Proc. of the Nutrition Symposium on Protein Metabolism, National Vitamin Foun­ dation, New York City. 55-75.

M a r k l e y , K. S., and R. M. H a n n 1925 Assoc. Off. Agric. Chemists J. 8: 455-467. M er t z, E. T ., E. J. B a x t e r , L . E. J a c k s o n , C. E. R oderuck and A. W eis 1952 Essential amino acids in self-selected diets of older women. J. Nutrition, 46: 313-322. M it c h e l l , H. II. 1950 Some species and age differences in amino acid requirements. Protein and Amino Acid Requirements of Mammals, Academic Press Inc., New York City. 1-32.

M it c h e l l , FI. H ., a n d R. J. B lock 1946 Some relationships between the amino acid contents of proteins and their nutritive values for the rat. J. Biol. Client., 163: 599-620.

R o se, W. C. 1949 Amino acid requirements of man. Fed. Proc., 8: 546-552. ------1952 A half'-century of amino acid investigations. Cliem. and Eng. News, 30: 2385-2388.

R ose, W. C., W. J. H a in e s , J. E. J o h n s o n and D. T. W arner 1943 Further experiments on the role of the amino acids in human nutrition. J. Biol. Chem., 148: 457-458. R o se, W. C., W. J. H a in e s , D . T. W arn er and J. F . J o h n s o n 1951 The amino acid requirements of man. II. The role of threonine and histidine. Ibid., 188: 49-58. AMINO ACID REQUIREMENTS 81

R o se , W. C., J. E. J o h n s o n a n d W. J. H a in e s 1950 The amino acid requirements of man. I. The role of valine and methionine. Ibid., 182: 511-550. R o se, W. C., G. F. L a m b e r t , M. J. C oon 1951 The amino acid requirements of man. VII. General procedures; the trytophan requirement. Ibid., 211: 815-827. W h a r t o n , M. A., T). T yr r e l l a n d M. B. P a t t c N 1953 Amino acid intake and nitrogen balance of college women. J. Am. Diet. Assn., 29: 573-577. THE QUANTITATIVE AMINO ACID REQUIREMENTS OF YOUNG WOMEN

II. VALINE 1

RUTH M. LEVERTON,2 MARY R. GRAM/ EILEEN BRODOVSKY/ MARILYN CHALOUPKA/ AMY MITCHELL 6 AND NORMA JOHNSON1 Nebraska Agricultural Experiment Station, University of Nebraska, Lincoln

ONE FIGURE

(Received for publication May 25, 1955)

The essentiality of valine for maintaining nitrogen equi­ librium has been reported hv Rose, Haines and Johnson ( ’42) and Rose, Johnson and Haines ( ’50). Based on studies of two young men Rose ( ’49) has suggested the tentative value of 0.80 gm as a minimum daily requirement for valine. He found that twice this amount, 1.6 gm of L-valine, was a satisfactory intake for 29 subjects in studies of amino acids other than valine.

1 Published with the approval of the Director as Paper no. 706 Journal Series, Nebraska Agricultural Experiment Station. 2 Present address: Division of Home Economics, Oklahoma A. & M. College, Stillwater. 3 Present address: Deoartment of Home Economics, University of California, Berkeley. 4 Present address: Department of Biochemistry, School of Medicine, Vanderbilt University, Nashville, Tennessee. 5 Present address: School of Home Economics, University of Wisconsin, Madison. "Present address: Univeisity of Nebraska Press, Lincoln. ’ Present address: Spokane, Washington. 83 84 RUTH M. LEVERTON AND OTHERS

Valine is the second amino acid for which the quantitative requirement of young women has been studied at the Uni­ versity of Nebraska.8 The purpose of the study was to determine the least amount of L-valine which would maintain nitrogen equilibrium in subjects under controlled conditions of intake. PROCEDURE The general plan of study, the description and management of -the subjects, the adjustment and experimental periods, and the nitrogen balance technique have been described in detail by Leverton and co-workers ( ’56) in the first paper of this series. Also described was the semi-purified diet which in­ cluded cornstarch, sugar, butterfat and corn oil, agar flakes, lemon juice, mineral and vitamin supplements, purified amino acids, and a few ordinary foods to increase the palatability of the diet. Adjustments were made in caloric intakes to meet individual needs and to keep the subjects from changing in weight. Seven college girls, identified here as subjects 21, 22, 23, 24, 25, 27 and 28, were on the initial study of valine require­ ment and are referred to as group A. All of the subjects were started on a daily nitrogen intake of 8.5 gm from the semi-purified diet in which glycine was the chief source of dietary nitrogen. However, subjects 21 through 24 were slow in reaching nitrogen equilibrium so their intakes were in­ creased to 9.5 gm of nitrogen with added glycine. The kind and amount of amino acids fed are shown in table 1. The auxiliary foods included 100 gm of canned tomato juice, 150 gm of canned apple juice, or applesauce or pineapple, and 50 gm of lettuce or carrots. After the requirements for the 5 amino acids — threonine, valine, tryptophan, phenylalanine and leucine — had been studied separately, 8 subjects who are identified as subjects A through H and referred to as group C were placed on the Test Mix, the composition of which is shown in table 1. The levels of

8 Supported in part by a U. S. Department of Agriculture contract sponsored by the Human Nutrition Research Branch, Agricultural Research Service. AMINO ACID REQUIREMENTS 85 these 5 amino acids in the Test Mix were the smallest amounts which had maintained nitrogen equilibrium in subjects who had been studied previously. For group C the daily nitrogen intake was 9.5 gm, and glycine and diammonium citrate sup plied equal amounts of the dietary nitrogen. The auxiliary foods included 100 gm of reconstituted frozen orange juice, 75 gm each of canned pineapple and canned peaches, and 25 gm of grape juice daily.

TABLE 1 The purified ammo acids fed in the study of valine requirement

GROUP A GROUP C AMINO ACID Amount equivalent Amount in to 20 gm egg protein Test Mix

gm/ person/ day gm/ person / day L-Arginine-HCl 1.549 1.5-9 L-Histidine- HC1 0.519 0.519 L-Lysine • HOI 1.800 1.800 L-Tyrosine 0.900 0.900 L-Phenylalanine 1.260 0.200 L-Tryptoplian 0.300 0.150 L-Cystine 0.480 0.480 L-Methionine 0.820 0.820 L-Tlireonine 0.980 0.184 1 L-Ueucine 1.840 0.600 DL-Isoleucine 3.200 3.200 L-Valine 1.460 2 0.550

Nitrogen content 1.910 1.583

1 Subjects C and F required 275 mg plus 30 mg from the auxiliary foods. 2 The amount of valine was varied during the study of requirement.

By analyzing individual foods and chemicals for nitrogen before they were fed, it was possible to keep the daily nitrogen intakes of all subjects constant within 0.2 gm throughout the study. The auxiliary foods used daily for group A and group C were assayed microbiologically with Leuconostoc citrovorum 8081 for valine and found to contain from 91 to 107 mg. The rounded figure of 100 mg is used to represent the amount supplied by the foods. Despite the difference in source, and 86 RUTH M. LEVERTON AND OTHERS therefore a possible difference in availability, this amount has been added to the intake of the purified, synthetic L-valine. T h u s, all of the levels referred to include 100 m g valine from food sources.

i.o ______

1560 650 465 375

L-VALINE INTAKE m g/day

Fig. X Valine intakes and nitrogen balances of 15 subjects for 255 subject-days. Each subject is identified by her code which ar pears beside her retention for each level on which she was studied. X = mean retention of all subjects on each level of intake. Intake includes 100 mg/day supplied by auxiliary foods. AMINO ACID REQUIREMENTS

RESULTS The mean daily nitrogen balance of each subject studied on the different levels of valine intake is charted in figure 1. The shaded areas include values from 5% above to 5% below the respective nitrogen intakes of 8.5 and 9.5 gm. As explained in the general procedure for these studies of amino acid requirements (Leverton et al., ’56), equilibrium is defined as that state in which the nitrogen excretion is within 95 to 105% of the intake. Negative nitrogen balance is not considered to exist until the total excretion is more than 105% of the intake. The mean balances of the subjects on each level of intake are joined by a solid line to emphasize the increasingly negative nitrogen balances with decreasing in­ takes of valine. The sequence in which the different levels of valine were given to all the subjects in group A was 1560, 465, 375 and then 650 mg. The individual data have been summarized in table 2 to show the mean, the standard deviation, and the range in daily nitrogen balance for each group of subjects on each level of valine. Also shown are the total number of days the subjects were studied, and the number of subjects in negative balance. On the daily intakes of 1560 and 650 mg of valine all 7 subjects were either in nitrogen equilibrium or in positive balance. When the intake was reduced from 1560 to 465 mg of valine, three subjects were in negative nitrogen balance, and when it was reduced further to 375 mg, 5 subjects were in negative balance. The variability among subjects on the same level was greater on the lower valine intakes than on the higher ones. Considering only mean daily balances, subjects 21 and 27 were in nitrogen equilibrium on the lowest intake of 375 mg of valine, but their daily balances were difficult to interpret. When subject 21 had this intake for 4 days, following 6 days on an intake of 465 mg of valine when she was in negative balance, her nitrogen balances were — 0.45, — 0.42, -j- 0.60 and — 0.53 gm. Because she did not feel well and was threatened with nausea her intake was increased to 650 mg of valine for 88 RUTH M. LEVERTON AND OTHERS

6 days. On this intake her symptoms disappeared and her mean daily balance was — 0.2 gm. Following 11 days on an intake of 465 mg of valine when her mean daily balance was 0.03 gm of nitrogen, subject 27 was given 375 mg daily for 5 days. Her nitrogen balance on successive days was + 0.11, — 0.11 (the third day she vomited the evening meal and no figure is available), — 0.69 and + 0.01 gm. Because she was nauseated, this level was discontinued and her -intake increased to 650 mg of valine for 7 days which resulted in a mean daily nitrogen balance of 0.76 gm.

T A B L E 2

Nitrogen balance of subjects on different levels of natine intake

NUMBER TOTAL NITROGEN BALANCE NUMBER OF LEVEL OF NUMBER ------SUBJECTS OF SUBJECTS O F Range IN L-V ALINE 1 ON EACH SUBJECT- Mean S.D.2 ------— - n e gative LEVEL DAYS Low High BALANCE 3

m g / d a y g m / d a y g m / d a y g m / d a y Group A 1560 7 55 0.27 0.29 0.00 0.80 0 650 7 69 0.11 0.31 — 0.20 0.76 0

465 7 48 — 0.39 0.34 — 0.95 0.03 3

375 7 31 — 0.55 0.35 — 1.16 — 0.17 5

Group C Test Mix 650 8 52 0.00 0.19 — 0.40 0.25 0

Groups A and C

650 15 121 0.05 0.25 — 0.40 0.76 0

1 Includes 100 mg supplied by the auxiliary foods. 2 Standard deviation. 3 Nitrogen excretion more than 105% of the intake.

T A B L E 3 Test of significance of differences between nitrogen balances

LEVELS OF MEAN N DEGREES OF L-VALINE INTAKE DIFFERENCE FREEDOM t

m g / d a y g m N /d a y 375 vs 465 0.16 14 12 0.889 465 vs 650 0.44 22 20 3.385 650 vs 1560 0.22 22 20 1.833

’ Significant at the 1% level of probability. AMINO ACID REQUIREMENTS 89

N t - CO 5"3 r'S w tes oo © 00 CM 04 CM CO a> t>- CO CM g S f © CO* CO CO CM cm’ t -i oq co LO LO TfÌ rH cd A < ¡25 \ LO lo TfH co ■nn co LO O w "a a

a Z s bi C l r-i CO CO CM © r-H 03 co cq o t - o 04 b - Ì S 3 ai ai o ’ O r-i CO b» CO rH r-1* o cm’ CM* oq o F? & a r—i T—1 CM CM CM 1—1 i—I 03

co lOCOOOCQOOO •

CC to CV IO O io H Si ° ^ ^3 3

00 CO t > t |H c m CD IO O Q N l> 05 CO CM* © CO* r H O O CM COOOOrHr-ir-ÌoOO ■S « M CO (M W (M W (MWNCOCOniMM

s> ns o o O LO LO LO LO o o o o o c o o \ LO LO LO CO co CO CO LO LO LO LO LO LO LO LO c* CO CO CO TjH CO CO co CO CO CO CO CO £ r-0 « Nitrogen excretion within 95 to 195% of the intake.

ti T3 ° Pi n3 ^ CD pH r r-H CD ■4-a ni O rj J3 ' oa ri O- iQ P< ,Q ¡H d d 3 ri

The 8 subjects in group C maintained nitrogen equilibrium on the Test Mix which supplied 650 mg of valine daily and minimum amounts of the other 4 amino acids for which the requirements had been studied. Results of the “ t” test gre given in table 3 and show that the difference between the mean nitrogen balances on 465 and 650 mg of valine is significant at the 1% level of probability. The smallest amount of valine which maintained nitrogen equilibrium in each subject in group A is shown in table 4 together with data on the creatinine coefficient, caloric intake, and age of each subject. Similar data are given for the subjects in group C on their intake of 650 mg. To consider the metabolic size of the subjects, the valine intake is expressed as mg/kg3/4 actual weight per day as well as the total milli­ grams per day. There were no apparent relationships between the amount of valine needed for equilibrium and the creatinine coefficient, caloric intake, or age of the subject. All of the subjects in group A experienced some nausea and loss of appetite when their intake of valine was reduced to 375 mg daily. Sometimes during the few days they were on this level, subjects 22, 27 and 28 vomited after the evening meal.

DISCUSSION The young women in this study varied greatly in the total amount of valine needed to maintain nitrogen equilibrium on a semi-purified diet in which the food supplied 100 mg daily. Individual needs ranged from a low of 20.4 mg/kg3/4/ day for subject 24 who weighed 65.5 kg, and was the largest subject in the group, to a high of 36.7 mg/kg3/4/day for subject 28 who weighed only 46.3 kg and was the smallest in the group. All 15 of the young women who were studied on a daily intake of 650 mg of valine maintained nitrogen equilibrium. Four of the 7 subjects who were studied on the lower intake of 465 mg were in nitrogen equilibrium. On the lowest intake of 375 mg of valine 5 of the 7 subjects were in negative AMINO ACID REQUIREMENTS 91 nitrogen balance, and although the other two subjects ap­ peared to be in equilibrium, their daily balances were er­ ratic and they had symptoms suggestive of a deficiency state. Whether or not the 8 subjects in group C could have main­ tained equilibrium on an intake lower than 650 mg of valine was not determined. It may be noted, however, that the nitrogen balances of 7 of these subjects in group C were numerically similar to the balances of subjects 21, 25 and 28 in group A, who had been unable to maintain equilibrium on 465 mg of valine. The other subject in group C, subject F, was no lower in the equilibrium zone than subject 23 had been on 465 mg of valine. Subject 23 lost more nitrogen than any other subject when the valine intake was reduced to 375 mg. The level of valine intake which permitted all of the subjects to be in the zone of nitrogen equilibrium was 550 mg of the chemically pure l form plus 100 mg from the auxiliary foods. The total of 650 mg is suggested as a minimum daily require­ ment for valine by young women similar to those studied here. This suggested requirement of 650 mg of valine for young- women approximates the 800 mg proposed by Rose ( ’49) as the minimum daily requirement for men. Yaline is not likely to be a limiting factor in the ordinary diet of mixed foods. The valine intake of 4 women on self- chosen diets was found by Futrell et al. ( ’52) to range from 2.85 to 5.39 gm per day. Mertz et al. ( ’52) have reported daily intakes from 0.8 to 4.9 gm of valine for 18 women on self- chosen diets. Diets of common foods which were used in studies with controlled intakes were reported by Wharton et al. ( ’53) to have supplied 3.1 to 3.4 gm per day. Allison ( ’53) has summarized Block’s calculations of “ the patterns of amino acids” consumed by people in different parts of the world. For valine the figures range from 3.8 to 4.3 gm per 100 gm of protein. Thus the protein intake would have to drop below 25 gm before the valine intake would be less than the minimum requirement suggested by the present study. 92 BTJTH M. LEVERTON AND OTHERS

SUMMARY The valine requirement of 7 young women has been studied by using a semi-purified diet, reducing the intake of l-valine stepwise from 1560 to 465 to 375 mg and then increasing it to 650 mg- daily, and measuling the nitrogen balance. All subjects were in nitrogen equilibrium, which is defined as the excretion being within 95 to 105% of the intake, or were storing nitrogen on a daily intake of 650 mg of valine. ‘ When the intake was reduced to 465 mg daily only 4 of the 7 subjects were in equilibrium, and when it was further reduced to 375 mg the subjects were either in negative balance or showed symptoms which may have been a result of an amino acid deficiency. In a later study, 8 other subjects maintained nitrogen equilibrium on an intake of 650 mg of valine when threonine, tryptophan, phenylalanine and leucine were also present in the semi-purified diet in similarly established minimum amounts. The lowest daily intake of valine which maintained all of the subjects in the zone of nitrogen equilibrium was 650 mg. This amount is suggested as the minimum daily requirement of valine for young women similar to those studied here.

LITERATURE CITED

A l l is o n , J. B. 1953 Amino acid requirements of man. Borden’s Review of Nutrition Research, XIV: 61-79.

F u t r e l l , M. F., R. N. L u t z , M. S. R e y n o l d s a n d C. A. B a u m a n n 1952 Studies on amino acids in self-selected diets. J. Nutrition, 46: 299-311.

L e ver to n , R, M., M. R. G r a m , M. C h a l o u p k a , E. B r o d o vsk y a n d A. M it c h e l l 1956 The quantitative amino acid requirements of young women. I. Threonine. Ibid., 58: 59-82.

M er tz, E. T., E. J. B a x t e r , L. E. J a c k s o n , C. E. R oderuck a n d A. W eis 1952 Essential amino acids in self-selected diets of older women. Ibid., 46: 313-322. Rose, W. C. 1949 Amino acid requirements of man. Fed. Proe., 8: 546-552. R ose, W. C., W. J. Haines and J. E. J ohnson 1942 The role of the amino acids in human nutrition. J. Biol. Chem., 146: 683-684. AMINO ACID REQUIREMENTS 93

R o se , W. C., J. E. J ohnson a n d W. J. H a in e s 1950 The amino acid require­ ments of man. I. The role of valine and methionine. Ibid., 182: 541-556. W h a r t o n , M. A., D. T yr r e ll a n d M. B. P a t t o n 1953 Amino acid intake and nitrogen balance of college women. J. Am. Diet. Assn., 29: 573-577. TRANSFER OF PHOSPHATE IN THE DIGESTIVE TRACT III. DAIRY CATTLE 1

ARTHUR H. SMITH,2 MAX KLEIBER, ARTHUR L. BLACK AND GLEN P. LOPGREEN Department of Animal Husbandry, University of California, Davis

(Received for publication July 25, 1955) The transfer of intravenously injected radiophosphate into the tissues and contents of the digestive tract has been de­ scribed previously for swine (Smith, Kleiber, Black and Lu- ick, ’55a) and for sheep (Smith, Kleiber, Black and Baxter, ’55b). The structure of the digestive tract of these two species is quite different, swine having a simple stomach and a tract adapted to a concentrated ration, whereas sheep have a complex (ruminant) stomach and the digestive tract is adapted to a bulky ration. Marked differences were observed between these two species in the transfer of circulating uhos- phate into the digestive tract which indicated functional dis­ similarities in their digestive processes. The digestive tracts of the sheep and cow are quite similar, both being ruminants; however, the relative sizes of various segments differ somewhat, and the epithelial lining shows cer­ tain structural differences (Sission and Grossman, ’53). The quantity of volatile fatty acids, per unit of body weight, in the digesta of the cow is about twice as great as in the sheep (Elsden et ah, ’45-’46). Thus, if phosphate is a major buf­ fering agent for these products of ruminant digestion, quan­ titative differences in its secretion into the digestive tract might be expected between sheep and cattle. Some of the data reported here have been previously cited by Lofgreen et al. ( ’52) in comparing the effect of single and multiple tracer injections on the secretion of labeled phos­ phate into the digestive tract of the calf. 1 This investigation was su'i ported by the U. S. Atomic Energy Commission. 2 Post-doctoral fellow of the U. S. Atomic Energy Commission. Present address: Department of Poultry Husbandry, University of California, Davis. 95 96 SMITH, KLEIBER, BLACK AND L0EGREEN

METHOD Animals for this experiment were young calves, about two months old, and mature dairy cows, at least three years old. These animals are described and the radiophosphate dosage is given in table 1. Each animal was injected through the jugular vein with a buffered isotonic solution of radiophos­ phate, after the method described by Ralston et al. ( ’49) Following injection, the animals were kept in special q.uarters, the cows in a stanchion and the calves in a special pen, and fed as usual until time of sacrifice, which varied from one- half hour to 72 hours after injection. Slaughtering followed the usual commercial procedure, the animals being stunned and exsanguinated. Samples of the tissues and contents of the digestive tract were collected and processed as previ­ ously reported for sheep (Smith, Kleiber, Black and Baxter, ’55b). Determinations of dry matter, phosphorus, and radio­ activity were made as previously reported for swine tissues (Smith et al., ’51). The viscera of Jersey calves, sacrificed 24 and 48 hours after injection, showed some pathological changes. Calf C4 had an abcess of the lung, and C8 had a serous cyst on the liver and some fibrotic change of the kidney. This pathology probably had no general effect on the results reported herein since the behavior and appearance of the animals was nor­ mal, and their body weight was greater than that of the other three Jersey calves of the same age, indicating normal alimentation.

RESULTS AND DISCUSSION The results of the chemical analyses (percentage of dry matter and phosphorus concentration in the dry matter) of the gastrointestinal tissues and contents for the calves and mature cows are summarized in table 2 as the means and their standard errors. The data on the mature cows include re­ sults on 4 additional dairy cows similarly sacrificed and ana­ lyzed for experiments other than those reported herein. No distinction was made between dry and lactating cows in sum­ marizing these analyses. Description o f experimental animals NT TNAL HOP ATE OSPH PH L A STIN TE IN O o n i O o > M X M Q O O o o c t— 0 0 5 0 M . M K H (—! X U CQ CO jf r M > i 1 - r O w w « ►"3 t-M hH o *"3 W *"0 5 !” “3 >“ « « -n '- PQ CO rH 'TH Iß CO m 0 0 CO 5 0 CO* o CO -Hi o i c o o t ti T o o c 0 0 5 0 ID t CO o o CO m o c »O CD CQ o c CO r— 1C f— CO 00* CO ’ o c M ¿4 -H be tUD n i 5 0 ’ o c TH 00* CO 1—1 o m 0 0 CO 1—1 CO O Ir­ ’ o c d i 0 0 IQ o c m o c m i— bJD CD 5 6 66 6 7 r—i CQ CQ* CO i—1 5 0 CO n i r i e fí T - t T—i CO* n i CO CQ* o c t-H n i CQ ! 1 t— fr­ CQ 05* C5 i c CO t— d i J—! N DIY CATTLE DAIRY IN —l o CQ oT a o ’ o n i l C ’ o c CO CQ l C l C OO t— CQ - 1 CQ l C l C n i Ir­ l C f t r 4 8 CO o c m o c CO n i CO - fc t—| *CG •S rS H CD tB s CD Q) o> 02 CD

*H = Holstein; J = Jersey; and H-G = Holstein-Guernsey cross.

s Dry cow; all oilier mature animals are lactating.

“ Age in days, with the exception of the last 6 columns, in which the age is given in years. 98 SMITH, KLEIBEK, BLACK AND LOFGBEEH

The water content of the gastrointestinal tissues and con­ tents is almost identical for the calves and cows. Cattle differ from sheep mainly in the greater water content of the digesta in the lower portion of the digestive tract. The phosphorus concentrations of gastrointestinal tissues and contents are different for calves and cows, the younger animals generally having a higher concentration of tissue phosphorus (except

TABLE 2

Dry matter and phosphorus concentration in the gastrointestinal contents and tiss es of dairy cattle and calves (mean and its standard deviation)

CALF (2 MO.) COW

Tissue Contents Tissue Contents Dry matter, % Rumen 18.0 ± 0.8 14.8 ± 0.7 21.7 ± 0.6 12.4 ± 2.2 Omasum 14.9 ± 1.1 21.2 ± 1.3 14.0 ± 2.2 22.9 ± 3.1 Abomasum 14.5 ± 1.2 21.1 ± 4.0 15.0 ± 0.2 15.0 ± 0.2 Small intestine 17.2 ± 0.3 9.9 ± 1.3 17.3 ± 1.5 10.9 ± 1.5 Cecum 17.7 ± 0.5 9.4 ± 1.1 20.5 ± 3.1 11.0 ± 0.7 Colon 19.5 ± 1.3 15.6 ± 1.2 15.2 ± 0.9 14.2 ± 1.6 Rectum 18.1 ± 0.9 25.8 ± 3.6 17.2 ± 1.0 18.6 ± 1.0

Phosphorus concontratirm, mg P/gm dry matter Rumen 0.6 ± 0.5 11.6 ± 1.2 23.6 ± 9.2 11.2 ± 1.5 Omasum 10.0 ± 0.3 6.2 ± 0.7 10.1 ± 1.8 14.3 ± 1.7 Abomasum 11.0 ± 0.6 6.5 ± 0.5 11.4 ± 0.8 11.4 ± 0.7 Small intestine 13.1 ± 0.4 9.1 ± 1.0 9.? ± 1.8 9.6 ± 1.0 Cecum 9.8 ± 0.8 5.9 ± 0.7 5.9 ±: 1.4 8.4 ± 1.3 Colon 9.1 ± 0.5 6.4 ± 0.8 6.0 ± 1.3 8.7 ± 1.4 Rectum 8.3 ± 0.5 4.2 ± 0.4 6.6 ± 1.1 9.6 ± 1.6

for the ruminal epithelium) and a lower phosphorus conccn- traction in the digesta (except for the ruminal contents) than found in the older animals. These differences in composi­ tion of digesta may result from a difference in composition of the rations fed, the calves receiving a liquid diet. The results of the chemical and isotopic (radioactivity) analyses and the tracer dose per kilogram of body weight have been used to calculate the “ standard specific activities” 3

s Specific activity = yc P32/m g phosphorus. /xa P32/m g phosphorus Standard specific activity = ------me P32 injected/kg body wt. INTESTINAL PHOSPHAT: Y CATTLE 99 of the various samples of gastrointestinal tissues and con­ tents, which are presented in table 3. The specific activities of the digesta in the various gastrointestinal segments at dif­ ferent times after tracer administration indicate a secretion of phosphorus which can he calculated. Calculation of the secretion of endogenous phosphorus into the ruminant stomach of the dairy cow is hampered by the lack of precise knowledge concerning the quantity of stomach contents. This situation has existed for some time; Eitzman and Benedict ( ’38) note: “ Further study of the amount and character of fill in the large domestic ruminants . . . is imperatively needed.” Veterinary anatomy texts (viz., Sisson and Grossman, ’53) generally state, without obvious basis, that the rumen and reticulum comprise 85% of the stomach capacity of the ox, the remainder being more or less evenly divided between the omasum and abomasum. Data contained in several recent reports 4 indicate that the

4 In order to determine the weight of fill in the bovine stomach, relative to body size, several recent reports citing such data were consulted. Eitzman and Benedict ( ’38) have reviewed earlier estimates of the fill of cattle, noting that they range from 10 to 40% of the live weight. Oil the basis of their own experiments they conclude that the ‘ 1 true fill in lactating coivs . . . would be in the neighborhood of 20% of the live weight.” The pertinent data from some other reports are summarized below:

W A S H B U R N ELSDEN ET AL. COMAR E T AL, BLAMIRE A ND BRO D Y MEAN VALUES ( 1 9 4 5 - 4 6 ) ( 1 9 4 8 ) ( 1 9 5 2 ) USED HEREIN ( 1 9 3 7 ) Oxen calf bull 7 7 COWS calves Herefords assorted 1 2 calf heifer bovines

Body wt., kg 364 182 39 204 466 55 Pill, % b.w. 20.7 30.2 8 .6 19.2 22 8.6 Stomach (% of fill) 84.5 82.0 88.5 91.2 81.0 82.0 82 85 Rumen and reticulum (% of stomach) 85.0 81.9 87.4 93.0 93.5 85.3 8 8 87 90 Omasum (% of stomach) 13.2 14.6 5.1 5.4 2 .0 1 0 .2 10 1 0 4 Abomasum (% of stomaeh) 1 .8 3.5 7.5 1 .6 4.5 4.5 o 3 6 100 SMITH, KLEIBER, BLACK AND LOFGREEN'

distribution of the actual quantity of digesta in the stomach may be quite different from the capacity, particularly with re­ gard to the omasum and abomasum. The following values in terms of percentage of body weight were used in calculating the endogenous secretion of phosphorus: rumen and retic­ ulum, calf 6.6%, cow 15.7%; omasum, calf 0.3%, cow 1.8% and abomasum, calf 0.4%, cow 0.5%. The ruminal secretion oi endogenous phosphorus, .SE, can be calculated from the quantities of phosphorus in the rumen and in the diet, and from the specific activities of the plasma inorganic phosphorus and the rumen phosphorus:

0

O o SB is the rate of secretion of phosphorus into the rumen (grams per h our); B is the quantity of phosphorus in the rumen contents (in grams) ; p is the standard specific activity of the rumen contents (at time,

t = P(t>) ; D is the dietary intake of phosphorus (grams per hour) ; and 7r is the standard specific activity of the plasma inorganic phos­ phorus (representing the standard specific activity of the secreted phosphorus). The quantity of phosphorus in the ruminal contents was calculated from their chemical analysis (table 2) and their estimated weight. Accordingly, the rumen of the calf would contain 6.25 gm of phosphorus, and of the cow, 102 gm of phosphorus. The dietary intake, D, was estimated from Albritton’s data ( ’53) as 0.19 gm of phosphorus per hour for the calves and 1.5 gm of phosphorus per hour for the cows. The specific activities of plasma inorganic and rumen phos­ phorus at specified times are given in table 3 (the mean values were used for 24- and 48-hour calyes). These specific ac­ tivities over various periods of time following tracer ad­ ministration were calculated as previously described for sheep Standard specific activities of tissues and contents of the gastrointestinal tract of dairy calves and cows & ¿4 be NETNL T P INTESTINAL V W X. > X M M M {> h Q LO O o rQ < s ¿1 CO -l O w P W o rH 0 0 o c O q t LO © q c - b © H< o c ’ o o - b 0 0 o c o 1 + g H< ’ o ' O > t 0 0 o q c LO o c LO q c VO i o - b & bJD h O 5 0 o o J H J H © © H r O CO CO 5 0 O V © © -rH 5 0 CO J H d d d d *¿5 s,- J H 5 0 0 5 O o o o c d d i + i + CO J H I H J H d d d d 5 0 CD O 5 0 CO CO 5 0 H 3 0 HH d d rH O l d d O J H - b CD d d O O L d d T—i o c h H

CO o i d d J H q J o H o c o 5 0 q c q b- b < H d ’ o o c q v O O < H ’ o d © o co c j H o c o o d d ’ o d J H J H J H CD < H i + i + d d O CO CO 5 0 < H d d q o q rH v O ’ o d d d VO LO LO 0 C rH J H o v 5 0 d d ^ . ■ © q v d o o i S h - b

o c o t rH 0 0 O d d 3 0 CO d * o © 5 0 j—! ~ b O © CO o c o o d d © © 5 0 LO rH 5 0 * o d d d © CO J H o i + i + © 5 0 o l H r O 5 0 d d J H H r d d O O 5 CO 0 5 0 L LO > H d d -J r 5 0 d d rH 3 0 h d ’ o O 3 5 0 0 O t d d H 5 0 o - b o rH H r -t T 0 0 . oq © O . t> 5 0 - b CO d © rH CO 5 0 CO 5 0 q v 1 © © d d o o o c d ’ o o c o t o c q v ~ b 5 0 d d d d CO J H o c 5 0 i + i + d d 5 0 CO d d VO H< CD - b o c o o d d o c o v o v - i - H t CO - t d d o c o v q v o d d D J H -( T CD - b d d — i—! r—f — rH t—i a O o c HO CO CD 1 q o J H o v o v d ’ o : r— J H 05 0 i H ’ o — rH - J b H o c * o d © o 5 0 0 0 o d d d q o q v o c o v 1 + 1 + H r J H d d O H r ~ b CO - b d d 0 5 VO vq VO 5 0 d d o c © d d o c » b o c q v 5 0 CD d d 05 0 * H © o c

d d © . O c © * +52 rA o d rH H1 rH 1rH rH O o - b 5 0 5 0 O 5 - r i - r rH ~ b © H J H 5 0 5 0 J H 5 0 CO d d d ’ o rH < - H b ^ H d d o o 5 0 4 H rH - J H b 5 0 D H rH CD d ’ o d d 5 0 J H i + i + rH f t * r H - b d d O 5 0 o CO o d d rH J H CO CD d d o c - q b v o v d d 5 0 J CD H O d d D O CD d d 5 0 rH 05 - b 5 0 H rH rH 5 0 O cq’ o 5 CO 0 ~ b © H r 5 C 5 0 O o CO H r ; d J H d d rH J 0 H 0 5 0 d d o o o CO O rH CO 0 0 H r 5 0 5 0 d d d d 5 0 CO — b 0 0 i + i + d d rH 5 0 rH 5 0 0 0 ~ b d d 0 5 CO CO 5 0 d d rH 5 0 j h CD d ’ b H r d d q o q v . 5 0 - b d © HrH ; P h 1 0 1 . The mean values for the three non-fasted calves only, i.e., C4, C6, and 07, are: Abomasum contents, 0.033 ± 0.001; and small intestine contents, 0,32 ±: 0.03. 1 0 2 SMITH, KLEIBER, BLACK AND LOFGREEN

(Smith et al., ’55b).5 The results of these calculations are summarized in table 4 as the mean secretion rates, and their standard errors, for the two age groups. The secretion rates calculated for the dairy cattle are much more variable than those obtained for calves or previously for sheep. Much of this variation may arise from the inho­ mogeneity of the group, one of the cows (XVI) being much larger than the others and two (XI and XV) being dyy. The mean ruminal secretion of the 5 mature cows is 2.75 gm per hour, or 66 gm per day. Assuming that bovine saliva, con­ taining 1 gm of phosphorus per litter,6 is the only source of endogenous rumen phosphorus, this would indicate a saliva­ tion rate of 66 1 per day. This rate of saliva formation is close to the directly estimated flow, 56 1 per day (Dukes, ’47). Thus, any extra-salivary source of endogenous phosphorus in the rumen of the cow would be small, particularly as compared with that previously indicated for sheep. Much more endogenous phosphorus, relative to body size, is secreted into the ruminant stomach of the cow than into 3 In addition to the observed values for plasma inorganic phosphorus specific activity, that at “ zero time’ ’ was also used in these calculations. The “ zero time’ ’ specific activity was calculated assuming: (a) that the tracer was distributed only through the inorganic phosphorus of plasma and interstitial fluids; (b) that these extra-cellular fluids comprise 300 ml per kilogram of body weight; and (c) that their inorganic phosphorus concentration was the same as that measured in the plasma — 7.9 mg % for calves and 6.0 nig % for cows. Thus, the tracer would be initially distributed through 1305 mg of phosphorus in calves (body weight = 55 kg) and through 84 mg of phosphorus in mature cows (body weight = 466 kg). The standard specific activity at “ zero time” following administration of “ x ” me of labeled phosphate would be: fio, P"2/m g P Standard specific activity = ----- * ‘ me P°- iinjected/kg body wt. 1000 /130- = 420, calves x / 55 1000 “x” /8400 „ : — x” 466 ":k5' ‘ No data are available concerning the concentration of salivary phosphorus in the dairy cow. However, if dairy cows have the same ratio of concentration between saliva and plasma inorganic phosphorus as reported for sheep, 15, (McDougnll, ’48), one would expect bovine saliv to contain 90 mg % phosphorus. A slightly higher concentration, 116 mg % of phosphorus has been reported for the female buffalo (Sharma, ’36). Thus a phosphorus concentration of 100 mg % (or 1 mg per liter saliva) ■would seem to be a reasonable estimate for dairy cow saliva. INTESTINAL PHOSPHATE IN DAIRY CATTLE 103

% % CM t - * > O CO TfH co o o GO o m rO O ü H o o o o GO CM O rH fc» Ci CO rH T—1 CO CO 00 S S 00 C i Ph’S +1 +1 +1 +1 r-î CD* +1 +1 CO o o o CO oo -H iO h o

<£>

oe co rH 0 0 C i ‘° . rH rH c q s* o O O o Sr- 1

£ O HH IO rH oo b - b - H «JD +1 +i +1 +1 9 +]

O 0 H CD rH CO O LO ■*

J OO rH 3 8 t - p : Eh

O Ä4

«JDS O ts £ 104 SMITH, KLEIBER, BLACK AND LOEGREEN that of the mature sheep. Most of this difference results from secretion into the rumen which, relative to size, is about 50% larger in the dairy cow. Similarly, the quantity of phosphorus in the rumen, per kilogram of body weight, is about 30% greater. Otherwise, phosphorus metabolism in the stomach of sheep and cattle is similar, the secretion of endogenous phosphorus per gram of dietary or rumen con­ tents phosphorus and the turnover time of ruminal con­ tents phosphorus being roughly equivalent [compare table 4 and Smith et al. ( ’55b), table 3]. The Secretion of endogenous phosphorus into the rumen of the calf amounts, relative to body size, to only 8% of that observed in the cow. The quantity of phosphorus in the calf ruminal contents, per kilogram of body weight, is only half as great as that in the cow. The turnover time of ruminal phosphorus in the calf, however, is about the same as that calculated for the cow. The secretion of phosphorus into the omasum and aboma­ sum was also calculated: , t Oo(tl + (D + Sr) ( J od t-J p d t)

o o

A ai(o + (D + Sr + S0) (£ adt — ^ odt 0 0 SA: adt Where: f S0 is the rate of secretion of phosphorus into the omasum (grams per hour); SA is the rate of secretion of phosphorus into the abomasum (grams per hour) ; 0 is the quantity of phosphorus in the omasal contents (in grams) ; A is the quantity of phosphorus in the abomasal contents (in gram s); o is the standard specific activity of the omasal contents (at time, t = °) ; a is the standard specific activity of the abomasal contents (at time, t = au>) ; and other symbols have the same significance as previously indicated. INTESTINAL PHOSPHATE IN DAIRY CATTLE 105

The quantity of phosphorus in the contents of these organs was calculated from their chemical analyses (table 2) and their estimated weight. Accordingly, the oinasal contents of the calf would contain 0.22 gm of phosphorus, and of the cow 28 gm of phosphorus; the abomasal contents of the calf, 0.30 gm of phosphorus, and of the cow, 4gm of phosphorus. The specific activities of the contents were taken from table 3 and integrated over various periods of time as previously indicated. The results of these calculations are summarized in table 4. The apparent secretion rate of phosphorus into the omasum and abomasum of the mature dairy cow is roughly equivalent, per kilogram of body weight, to the values previously cal­ culated for mature sheep. Omasal secretion of phosphorus in the calf, relative to body size, was only a third as great as that calculated for the cow. One striking difference be­ tween the calf and mature cow was the indication of a phos­ phorus absorption in the abomasum of the calf, which may be more apparent than real. If ingested milk were to pass directly to the abomasum via the esophageal groove, which appears likely (Dukes, ’47), and sulcus omasi, a lower spe­ cific activity could occur in the abomasum, as compared to anterior organs, without the absorption of endogenous phos­ phorus. The indication of a small secretion of endogenous phosphorus into the abomasi of the calves sacrificed two and 5 hours after injection, and which were last fed at least an hour prior to injection, supports this explanation. The ap­ pearance of an endogenous phosphorus secretion into the omasi of all calves indicates that milk entering this organ via the esophageal groove is not retained but is passed on via the sulcus omasi, the omasal contents being largely de­ rived from ruminal contents. The specific activity of the small intestinal contents of both cows and calves (see table 3) is usually greater than that of the abomasal contents (the only observed exception is cow XVI where the specific activities of the two segments were about the same), indicating an intestinal secretion of phos- 106 SMITH, KLEIBER, BLACK AND LOFGREEN pliorus. Thus, the cow appears to be intermediate between swine, where the small intestine is the main site of endogenous phosphorus secretion, and sheep, where no significant in­ testinal secretion was indicated. In calves, the specific activity of the content of the lower intestinal tract (i.e., colon and rectum) is much less than that of the contents of the small intestine or other anterior region up to 24 hours after the injection. This indicates that in the calf, endogenous phosphorus is preferentially ab­ sorbed, as compared to food phosphorus, from the contents of the lower intestinal tract. This selective absorption of endogenous phosphorus was also observed in swine. In the mature dairy cow, however, there is no evidence of a selec­ tive absorption of endogenous phosphorus from the contents of the lower intestine, the specific activity of the digesta throughout the tract being essentially the same 24 hours after the tracer administration. The secretion of other mineral elements into the digestive tract of the mature bovine appears to be somewhat different from the secretion of phosphorus, which is largely into the rumen. Neither cobalt (Comar and Davis, ’47) nor copper (Comar, Davis and Singer, ’48) is secreted to any large extent into the rumen, but both are mainly secreted into the abomasum and small intestine. The secretion of cobalt seems to be different for sheep and cattle, since Keener et al. ( ’51) were unable to detect any significant radioactivity in the abo- masal contents of sheep up to 48 hours after the intravenous administration of radiocobalt. Intravenously administered radiocalcium (Hansard et al., ’52) appears in all contents of the digestive tract; however, the main secretion, according to observations made 10 minutes after the injection, is in the abomasum and small intestine. The uptake of labeled plasma inorganic phosphorus by the gastrointestinal tissues, indicating the exchange of phos­ phorus between these two components, appeared to be great­ est in the intestinal tissues of calves and in the omasal tissue of the mature cow. This tissue-blood phosphorus exchange INTESTINAL PHOSPHATE IN DAIRY CATTLE 107 was evaluated by measuring the rate at which the difference in specific activity between tissue phosphorus and plasma inorganic phosphorus decreased. For these calculations, the specific activity of the tissue was related to that of the plasma inorganic phosphorus, and the difference in specific ictivity between tissue phosphorus and plasma inorganic phosphorus was represented as 1-(relative specific activity) - - b e “ “

Where * specific activity of tissue phosphorus relative specific activity = ------specific activity of plasma inorganic phosphorus

h is the difference in specific activity at time zero, 'which is unity for these calculations; k is the rate constant for the difference in specific activities, and is negative since the difference decreases with increasing time; and t is the time (hours) elapsed since administration of the tracer.

TABLE 5 Bates of equilibration of specific activities of plasma inorganic phosphorus and tissue phosphorus; 1 — Bel. S.A. — e~kt

k X 100 1 VALUES FOR VARIOUS PARTS OF THE ALIMENTARY TRACT ANIMAL USED Rumen Omasum Abomasum Cecum Colon Rectum

Calf — 4.0 — 3.7 — 1.8 — 4.9 — 4.3 — 3.3 — 2.4

Cow — 1.9 — 2.1

1 Since t is measured in hours, k X 100 is the percentage change in the specific activity difference per hour. Assuming that only a simple exchange exists between tissue and circulating phosphate, this would represent the percentage of tissue phosphorus replaced by plasma inorganic phosphate per hour.

The rates of equilibration of specific activities between the gastrointestinal tissues and plasma inorganic phosphorus of cows and calves are presented in table 5. The calculations were not made for the stomach tissues of the mature cow, since data were available for only two animals and these were quite different. The phosphorus exchange between tissue and circulating phosphate, indicated by these calculations, ap­ pears to be unequal among the gastrointestinal tissues of the calf. A rather large rate of phosphorus exchange is 108 SMITH, KLEIBER, BLACK AND LOB GREEN evident for ruminal epithelium and small intestinal tissue and decreases lower down. Very little difference is evident in the phosphorus exchange between the various intestinal tissues in the mature cow. The rates of phosphorus-ex­ change in tissues of the mature cow are somewhat less than those for corresponding tissues in the calf. Thus, the phosphorus metabolism of bovine intestinal tissues seems to decrease with decreasing age, as was previously observed in'most of the gastrointestinal tissues of sheep. The rate of phosphorus exchange between gastrointestinal tissues and plasma appears to be much less in cattle than previously observed in sheep. In calves, this rate of phos­ phorus exchange is roughly half that observed in one-month- old lambs (except for abomasal tissue, which is about the same for both) which were on a similar nutritional regimen. The rate of phosphorus exchange in the intestinal tissue of the mature cows appears to be less than half that previously observed in 10-month-old sheep. In order to determine the variability between individuals in the transfer cf phosphate into the contents and tissues of the digestive tract, 4 calves were sacrificed 24 hours after injection of the tracer. These animals, C4 to C7, were of different breeds, varied widely in body size, and differed in times of fasting (see table 1). An analysis of variance was made on the chemical composition and specific activity of these calves (table 6). The “ F values” (Snedecor, ’46) indicate that the chemical composition (dry matter and phos­ phorus concentration) is characteristic for the tissue and contents, and highly so in most cases. However, in spite of their wide variation in body size, breed, and nutritional status, no significant differences in chemical composition were evi­ dent between these animals. The standard specific activities of the samples, 24 hours after injection, are highly characteristic for the various tis­ sues and contents. It is interesting to note, however, that the relative variation between tissues decreases and that be­ tween the contents increases when the specific activity is cal­ INTESTINAL PHOSPHATE IN DAIRY CATTLE 109 culated in terms of the plasma inorganic phosphorus (i.e., as the “ relative specific activity” ). This would suggest a greater dependence of the gastrointestinal secretions on the inorganic fraction of the plasma as a source of phosphorus than is the case for the tissues, and also some independence between the inorganic and some other fraction (i.e., organic) as sources of tissue phosphorus.

TABLE 6

Inalysis of variance 1 for chemical composition and specific activity of 4 calves sacrificed 84 hours post-injection

MEAN SQUARES Chemical composition SOURCES OF Degrees Standard Relative VARIATION of sp. activity sp. activity Dry matter Phosphorus freedom cont. tissue tissue cont. tissue cont. tissue cont.

Mean square be­ tween animals 3 .0150 .0370 .0331 .0570 5.83 61.27 1.50 7.76

Mean square be­ tween samples 6 .0333 .0476 .0618 .0819 32.55 192.18 20.73 27.11

Error 18 .0057 .0067 .0012 .0109 7.96 49.67 1.24 3.99

E between samples 5.85 = 7.112 5.06 2 7.50 2 4.09 !' 3.87 s 16.70 2 6.8 2

E between animals 2.63 5.53 2 2.95 s 5.23 2 0.73 1.23 1.21 1.95

^nedecor ( ’46). 2 Significant at the 1% level of random probability. 3 Significant at the 5% level o f random probability.

Differences between animals are also evident in the specific activity of the various 24-hour samples, highly significant for the contents (either as standard or relative specific activity) and only bordering1 on significance for the tissues. An examination of the data in table 3 indicates that the withholding of feed (C5) results in an increased specific activity of the abomasal and small intestinal contents, which could result from a continued secretion of endogenous phos­ phorus that is not diluted with food phosphorus. Ritzman 1 1 0 SMITH, KLEIBER, BLACK AND LOFGREEN and Benedict ( ’38) have reported that fasting of an adult ruminant does not reduce the quantity of the stomach contents but merely their dry matter. A similar situation is evident in these calves; the dry matter of the abomasal contents of the lasted calf, C5, is 11.0% (mean for 9 calves — 21.1 ± 4.1%) and of the small intestinal contents is 5.8% (mean for 9 calves = 9.9 ± 1.3%).

SUMMARY AND CONCLUSIONS The appearance of intravenously injected radiophospliate in the contents of the gastrointestinal tract of cows and calves indicates a large secretion of endogenous phosphorus. In the mature cow, as in mature sheep, the main site of secretion is the rumen, although most of the phosphorus in the bovine ru­ men seems to enter in the saliva. In the calves, however, the ruminal secretion is smaller and a large part of the endo­ genous phosphorus appears to he secreted into the small in­ testine, though not as much as that previously observed in swine. The rate of equilibration of tissue phosphorus with circu­ lating inorganic phosphorus was calculated for some organs. A slight aging effect on tissue phosphorus metabolism was observed with a lower uptake rate in older animals. The phos­ phorus exchange of intestinal tissues is generally less in the cow than in sheep. Four calves, which varied widely in body size and time of fasting, were sacrificed 24-hours after injection of radiophos­ phate. An analysis of variance revealed no significant differ­ ence in chemical composition between corresponding tissues of these animals; however, significant differences in phos phorus secretion rates were observed.

ACKNOWLEDGMENT The authors gratefully acknowledge the assistance of T. N. Chernikoff. INTESTINAL PHOSPHATE IN DAIRY CATTLE 1 1 1

LITERATURE CITED

A l b r it t o n , 3. C. 1953 Standard values in nutrition and metabolism. Tecli. Rapt. 52-301, Wright air development center, Wright-Patt arson Air Force Base, Ohio, pp 380. B l a m ir e , R. V. 1952 The capacity of the bovine stomachs. The Yet. Rec., 64: 492-494. Co m a r , C. L., a n d G. K. D a v is 1947 Cobalt Metabolism Studies. III. Excre­ tion and tissue distribution of radioactive cobalt administered to cattle. Arch. Biochem., 12: 25! -266. C o m a r , 0. L., G. K . D a v is a n d L. S in g er 1948 The fate of radioactive copper. J. Biol. Chem., 174: 905-914. Dukes, H. H. 1947 The Physiology of Domestic Animals. Comstock Publ. Co., Ithaca, N. Y., 6th ed., pp. 817. E l s d e n , S. R., M. W. S. H it c h c o c k , R. A. M a r s h a l l a n d A. T. P h il l ip s o n 1945-46 Volatile acids in the digesta of ruminants and other animals. ,T. Exp. Biol., 88: 358-68. H a n s a r d , S. L., C. L. Co m a r a n d M. P . P lu m t.ee 1952 Absorption and tissue distribution of radiocalcium in cattle. J. Animal Sci., 11: 524-35. K e e n e r , H. A., R. R. B a l d w in a n d G. P. P er c iv a l 1951 Cobalt metabolism studies with sheep. Ibid., 10: 428-33. L o fg r een , G. P., M. K leiber a n d A. H. S m it h 1952 The excretion of in­ jected PK into the gastrointestinal tract of the young calf. J. Nutrition, 47: 561-570. M cD o u o a l l , E. I. 1948 Studies on ruminant saliva. I. The composition and

output of sheep’s saliva. Biochem. J., 4 3 : 99-108. R a l s t o n , N. P., M. K leiber a n d A. H. Smith 3949 Venous catherization of Dairy Cows. J. Dairy Sci., S3: 889-92. R it z m a n , E. G ., a n d P. G. B e n e d ic t 3938 Nutritional Physiology of the Adult Ruminant. Carnegie Inst, of Washington, Publ. No. 494. pp. 200. S it a r m a , G. K. 1936 Secretion and composition of parotid saliva in buffalos. Indian J. Vet. Sci. and An. Ilusb., 6: 266-69. S is s o n , S., a n d J. D. Gr o s s m a n 1953 The anatomy of the domestic animals. 4th ed. W. B. Saunders Co., Philadelphia, pp. 972. S m i t h , A. II., M a x K l eiber , A. L. B l a c k a n d C. F. B a x t e r 1955b Transfer of phosphate in the digestive tract. II. Sheep. J. Nutrition, 57: 507-528. S m i t h , A . H ., M a x K l e ibe r , A. L . B l a c k , M . E d ic k , R . R . R o b in s o n a n d H . H e it m a n , J r . 1951 Distribution o f intravenously injected radio­ active phosphorus (P E) among swine tissues. J. Animal Sci., 10: 8 9 3 -9 0 1 . S m i t h , A. H., M a x K l e ibe r , A. L. B l a c k a n d J. R. L u ic k 1955a Transfer of phosphate in the digestive tract. I. Swine. J. Nutrition, 57: 497-506. S nedecor, G. W. 1946 Statistical methods. Collegiate Press, Inc., of Iowa State College, Ames, Iowa. pp. 485. W a s h b u r n , L. E., a n d S. B rody 1937 Growth and development. X L II. Methane, hydrogen, and carbon dioxide production in the digestive tract of ruminants in relation to the respiratory exchange. Univ. Missouri Res. Bull. No. 263, pp. 40. STUDIES ON COMPARATIVE ABSORPTION AND DIGESTIBILITY OP ACETOGLYCERIDES

ANTHONY M. AMBROSE AND DOROTHY J. ROBBINS ’harmacology Section, Western Utilisation Research Branch, Agricultural Research Service, United States Department of Agriculture, Albany, California

(Received for publication August 22, 1955)

The substitution of acetyl groups for part of the fatty acids of ordinary fats and oils, such as lard or edible vege­ table oils, has resulted in the development of new types of fats of unusual properties and of potential value as edible fats (Feuge, ’55; Baur, ’54). These modified fats, which have marked resistance to changes in consistency and to heat damage and are plastic over a wide temperature range and stable against oxidation and rancidity, have been referred to as “ acetoglycerides” or “ acetin” fats (Feuge, ’55; Baur, ’54). By proper selection of suitable fats and oils as starting materials and appropriate degrees of acetylation, the aceto­ glycerides can be tailored to fit a variety of needs (Feuge, ’55). The “ acetoglycerides” differ from natural fats and oils, composed of mixed triglycerides of long-chain fatty acids, in that they are predominantly diacetotriglycerides of high molecular weight fatty acids, such as stearic acid or oleic acid. Other combinations are possible. However, as used here the term “ acetoglycerides” refers to both acetostearins and acetooleins. To distinguish between the two, the acetylated monoglycerides derived from highly hydrogenated lard or vegetable oils are referred to as acetostearins, and those derived from unsaturated fats as acetooleins. Uses for the acetoglycerides appear to be “ unlimited.” Aside from their use in cosmetics and pharmaceuticals, some of the more important suggested uses to the food industry

113 114 A. M. AMBROSE AND D. J. ROBBINS are as coating materials for edible foods, such as cheese, nut meats, rasins, in chocolate coatings for ice cream bars, etc.; in the manufacture of candy either as a slab dressing or as a component; in the preparation of plastic shortenings, des­ sert and filling fats; and in the preparation of margarine- type of spreads where monoglycerides are now permitted. Although the “ acetoglycerides” are not naturally-occurring substances, they are undoubtedly readily hydrolyzed in the alimentary tract to yield acetic acid, glycerol, fatty acid, and monoglycerides — all occurring physiologically in the normal course of the digestion of fats. Since the acetoglycerides are new products, in the sense that they are being considered for the first time as intentional additives in foods, it remains to be proved that these new fats do not interfere with normal physiological functions in ways not previously revealed in studies with acetic acid, glycerol, fatty acids and monoglycerides. The present studies were therefore designed to compare the absorption and di­ gestibility of acetostearin, acetoolein, mixtures of acetostearin and acetoolein and hydrogenated vegetable oil.1

EXPERT MENTAL The source, description and composition of the various acetoglycerides (acetostearins and acetooleins) used in the present studies are shown in table 1. In studies with rats to be reported below, the various acetoglycerides were incorpo­ rated into the basal diet on a weight basis. The basal diet had the following percentage composition: casein 10, corn meal 73, linseed oil meal 10, alfalfa meal 2, bone ash 1.5, sodium chloride 0.5 and cod liver oil (U.S.P.) 3. To 80 gm of the basal diet, 20 gm of either of the acetoglycerides or of Crisco was added and fed ad libitum. The fats were warmed to 60°C. before they were added and thoroughly mixed with the basal diet.

1 Crisco. The use of this product docs not imply recommendation. Any other similar, commercially available product could have been employed. METABOLISM Oi . OERIDES 115

o o © t o t>- fa d 03 t o l > 03 O a t o CO c o ° 1

M fc f a g O A g Os CO 03 O rH fa w r t o 03 CO CO •<5 <* fa # t o «3 H <5 o ’ o © o O fc & 3 d

SO ■« S 0 OJ 03 rH rH t o j H W b - 03 © rH t o < n o # fa *e o C o rH o ’ fa o ^

«H MQ o o to <

0 3 00 t o t o 0 3 t-^ 0 3 rH* oo’ < o rH r H rH rH iH

W w o o

P3 rH rH t - o 0 3 <1 O ^ 0 3 O 0 3 00 QO fa ¡3 CO rH 0 3 CO c o H < < «

S g w* O O ^ W rH rH 00 © C5 5 « O oo’ o ’ © P £ fa rH w

e ndfa o c$ i "2 S 'S QQ

c3 rd “r-y -fa fa» c3 c3 c8 cS d 02 __ . d nd j-j I §s ^ til 03 tyn -fa O ■2 © O V. 1 ° J3 fa ' P , - 2 r*% ^ ►» ? -fa -fa rH 0) 03 « “ © ¿3 03 J d O d « Q> o 03 O i>~, l a & 13 S ,o 1 1 t-> o 3 a d Pk & P< d GO o3 H CO rH rH © rH -fa d . 0 3 CO 0 3 03 c3 fa O rH « S to «j o O O 0 0 M3 <1 <1 ◄ <1 <1 116 A. M. AMBROSE AND D. J. ROBBINS

Absorption Studies on absorption were made over a 4-hour period in mature male albino rats weighing 200 to 300 gm each. Our procedure varied from that used by others in that the aceto- glycerides were fed as an integral part of the diet rather than by stomach tube. The diet containing 20% of one of the acetoglycerides or Crisco was fed to each rat previously fasted for 48 hours. Ten rats were used for each group. However, only those rats which consumed the greater portion of the offered diet within 30 minutes were used. The amount of diet consumed during this period was determined and the amount of fat calculated. Pour hours later the rats were killed, the gastro­ intestinal tract, between the esophagus and caecum, was ligated and removed immediately. The contents of the ligated portion of the gut were washed out and extracted with diethyl ether as outlined by Deuel, Hallman and Leonard ( ’40). The ether extract was dehydrated with anhydrous sodium sulfate and the amount of ether soluble residue (fat) was determined. The absorption of the various acetoglycerides was determined for each rat separately. A similar experiment for endogenous fat was conducted on control rats fasted for 48 hours but fed no fat or acetoglycerides. A total of 70 rats was used in these studies. The percentage absorption of the respective acetoglycerides was calculated as follows:

Ingested fa t — (fat recovered — endogenous fat in fasting rats) X 100 Ingested fat The data are summarized in table 2.

Digestibility Studies on the digestibility of the various acetoglycerides and the coefficient of digestibility were determined according to the procedure described by Augur, Rollman and Deuel ( ’47). Experiments were conducted with both male and fe­ male adult rats. Five rats of each sex were used. All rats METABOLISM OF ACETOGLYCEBIDES 117 were housed in individual screen-bottom metabolism cages. Food and water were provided ad libitum. For determina­ tion of metabolic fat, one group of 10 rats was fed the control diet which contained no added fat, other than 3% of cod liver oil. All other rats in groups oi 10 were placed on the experimental diets containing 20% of one of the acetoglycer- ides, mixtures of acetoglycerides, or Crisco. A total of 90 rats was used in these experiments. Following a preliminary adjustment period of 7 to 10 days on the respective diets, digestibility (of the various acetoglycerides) was determined

T A B L E 2 Absorption of various acetoglycerides (AG) and Crisco from the gastrointestinal tract of fasting rats

Absorption time 4 hours

SUPPLEMENT N O . OF FAT FAT ABSORBED F A T FE D F E D 1 TESTS RECOVERED CORRECTED mg mg mg % ± S.E.2 Control 0 7 81 AG-3 696 10 248 529 76.1 ± 1.6 AG-21 741 9 249 573 77.3 ± 1.2 AG-194 694 9 318 457 65.9 ± 3.0 AG-26 717 10 350 448 62.5 ± 2.3 AG-31 693 7 319 455 65.7 ± 3.1 Crisco 700 9 308 473 67.6 ± 1.1

1 See table 1 for description and characteristics of ‘ ‘ Supplements Fed. ’ ’ 2 SE = Standard error of the mean.

over a 7-day period. During the 7-day digestibility period accurate records of food consumed and feces excreted were kept for each rat. The feces from each rat were dried, weighed and ground to a coarse powder. The dried feces were an­ alyzed for neutral fat and free fatty acids by Soxhlet ex­ traction with diethyl ether. The fatty acids present as soaps were determined on the ether extracted residues after acidi­ fication with 50% sulfuric acid and re-extraction with diethyl ether. All diets were assayed for total ether extractives at the same time the experiments were run. The experiments and data are summarized in table 3. The coefficient of digesti- Summary o f digestibility of acetoglycerides (A G ) and Crisco when fed to male and female rats as part o f tie diet 118 Values in the table are the averages of 5 rats of each sex per group fc J eS He Û « O < o w o 3 w2 *o w3 5 w « K W Q . Hi g i * £ C » E- -5 «2 S« 2 H. £ É«£ s o s *w 3 => ■HS -i 3 n S \ es î>> c M p + » + Q CO Z o © ” ©cp O ¿1 01«

SEX ta 05 o H O W 3 ^rs t < rq 3 a O C 3 02 ai pi cS o m o 2 P. h «3 o S ® a> . ABOE N D J. ROBBINS . J D. AND AMBROSE . M A. e, ûQ ra +1 + 1 fiMS C* g glO io <=&>•©OSg cq 5-°. CCO GO Ç=5gco oo t- o > J“J©> a m COrH o o *«1 °. rHO 00 Oo o o o (MO CO o o oq o S P S COCQ rH o o fl © CQ .37 o © id

h .31 LCp-f o © o toco O00 H*00 o o 02 02 OO CQbOo COb- CO 02 02 CO02 CQ 1-1 r—1HH OO COCQ o o o ioo 02*COCO coco b- CQCQ l> co +1+1 t>* o < à CO S co rH rHrH00 r-i OCQ HHCQCQ b- IQ 02 o CO 02 02 02 o (M 02 ©CI +1 hh CQ O CQ LO O KO© 00 ©rH © d 02 COoo ©' d CQKOrH CQrH OSOS © © © bObO ©’ d bOrtf ©© iHKO ©© 00 HH©LO CQ COrH© H CQ© 02 +i+i bd 00 © © CQCOCQ r-iCQ LO <1 ci S iH02 tH rH00 CQ rH HH©00 ©1 co © 02 02 CO © rH t 02 +1 H CQ © CQ HH >o bOCO O© 00rH rHb- HH 00 r-i02* o © d © oqco oqic bdco* rHb- 00 ©Tfi co o rH02 © d © COco© © b-© +i+i CQCQ © © © CQ*ko 02 < ©© CQ© d rH HJCO à © rHCO ©-q rH HH SP coCQ ©TfH HHHH L~ 00 oo © 00 © o © +1 CQ © CQ HH*© r bfSb- HHCQ* coco © © 00 co*d oor HH* CQ bdco’ L0HH rH© ©© rH*COO b» ©bq b-© CQco ko ©HH HH© H*©’ il il il ©© rH CQoo© © ©© < 6 © ÎH ^ HHb- rnb- ©t-i HH© CQb-*rH ©‘ ©HH CQ HHHHCQ b- 00 00 bQ © oo 00 © II CQ © CQ LO CQ — h bded Hq rHrH CQCQ © © HHCQ© LO b-LO L0© L0© *<+ os(3 HÎ'0? P S KO© COGO*rH CQb- io COCQ bd co oq© CQb-C0* rH oq © +!+! H<00CQ rHb- © hh bd d 02 hi 00 rH +i CQ © CQ HH rH h r-î© ©os HH© HHCQ © © CQ*02 HH* © d © oo© © © © © CQrH*cq HHb-LO b- ©© iq <1+ O ° o CQ© CQcdt-irH HJ©© oo oo*© oo GOLC ©HH H1© rH*rH*©© o S hh rH^ COOOCQ b-co LO ©b- t>i +i+i HH ^ ' d' cq hh cq hh cd b- © © © CQ © CQ HH O +1 " COHH© CQ© © © © © HH© © © OQ© rH© © © r-î© ©© LOd *—j 00 ©©* rHiq 6 © © rH ©00 ©© CQ ©*d rH HH© ÏHLC HJ©HH oq CQ© r-i © ©bQ r-irH rHb-LC © HH©© CQ ©O b-LC +!+1 œ O O © © HH HH © bd CQ © CQ HJ il

1 For description and characteristics of ‘ ‘ Supplements Fed ’ ’ see table 2 Corrected for metabolic fat (77 mg/gm dried feces). 2 S.K = Standard error of the mean. METABOLISM OF ACETOGLYCERIDES 119 bility was calculated for each rat, but only the average results for each sex and group, together with the standard error of the means, are reported. All values for digestibility have been corrected for metabolic fat from rats on the control diet in which no fat was added, other than cod liver oil which was present in the diet. A correction factor for metabolic fat for male and female rats of 79 and 75mg/gm of dried feces, respectively, was obtained. The average, 77 mg/gm of dried feces, Avas used for correcting for metabolic fat. The coeffi­ cient of digestibility was calculated as follows:

Ingested fat— [excreted fat — endogenous fat (wt. dried feces X 77)] X 100 Ingested fat

RESULTS AND DISCUSSION

A b s o r p tio n A summary of the data on the absorption of the various acetoglycerides with the number of rats used in each group is given in table 2. Absorption of acetooleins (Ag-3 and Ag-21) was definitely better than the absorption of aceto- stearins or Crisco. On the other hand the absorption of the three aeetostearins, the two fully acetylated (Ag-194 and Ag-26) and the one partially acetylated (Ag-31) were of about the same order as for a partially hydrogenated fat, for ex­ ample, Crisco. These observations are in keeping with those •in studies reported by others on the absorption of oils com­ posed predominantly of unsaturated fatty acids as compared with fats consisting predominantly of saturated fatty acids (Steenbock, Irwin and Weber, ’36; Deuel, Hallman and Leon­ ard, ’40; Crockett and Deuel, ’47). No difference in absorption was observed between the acetostearin melting at 553C. as compared with the aeetostearins melting at 30.7°C. and 32.5°C., respectively. However, Crockett and Deuel ( ’47) did observe a difference in absorption between hydrogenated lard (M.P. 55°C.) and other fats with lower melting points. 120 A. M. AMBROSE AND D. J. ROBBINS

Digestibility Table 3 summarizes data on the comparative digestibility of the various acetoglycerides when fed to rats as an integral part of the diet. Data are also presented on the digestibility of mixtures of acetoglycerides, and for comparative purposes the digestibility of Crisco under our experimental conditions. Only average results are piesented When diets containing 20% of acetoglycerides were fed to rats, the acetooleins had the highest coefficient of digesti­ bility, while the acetostearins had much lower values. How­ ever, these values are substantially in agreement with those obtained in comparable studies by others with cottonseed oil and hydrogenated fats (Crockett and Deuel, ’47; Hoagland and Snider, ’40; Augur, Rollman and Deuel, ’47; Cheng, Morehouse and Deuel, ’49). When the diet containing 20% of one of the acetostearins (Ag-194) was modified to contain 15% of acetostearin and 5% of acetoolein (Ag-3) digestibility was slightly improved. However, when the diet was modi­ fied further (10% of acetostearin and 10% of acetoolein) digestibility was considerably improved. The digestibility of this mixture was 93.7% as compared to 95.9% for Crisco, which compares favorably with the digestibility of other partially hydrogenated fats and oils. It must be recalled that all of the acetostearins studied and reported here were pre­ pared from oils which, for all practical purposes, were com­ pletely hydrogenated as judged by iodine values of one or less, while commercially available hydrogenated fats and oils are only partially hydrogenated to iodine values ranging from 60 to 75. In the digestibility of various natural and synthetic fats, a number of factors have been considered: for example, melting point of the fat; composition of the fat, whether a simple or mixed triglyceride; presence of emulsifying agents, such as lecithin, monoglycerides, etc., and the presence of calcium and magnesium salts in the diet. The acetoglycerides reported here, with the exception of the acetooleins and M ETABOLISM OE ACETOGLYCERIDES 121 mixtures of acetoolein and acetostearin, had melting points of 30.7, 32.5, and 57 °C. for Ag-194, 26 and 31, respectively. The aeetooleins and mixtures of acetooleins and acetostearin were all liquid at 25°C. With the exception of Ag-31, all were within the critical temperature of 50°C. for maximum digestibility. Lyman ( ’17) and Mattil and Higgins ( ’45), as well as others, are of the opinion that the melting point of a fat is not as important a consideration as the composition of the fat (stearate content). These authors believe that stear­ ates are indigestible and that they are better utilized when fed as a mixed glyceride rather than when fed as mixtures of simple triglycerides (tristearin mixed with triolein). In our studies with the three acetostearins, the melting point did not appear to be as important a consideration as the stearate content of the diet. All three acetostearins had approximately the same digestibility coefficients (80 to 85%). The digestibility of the acetostearins was indeed definitely better than the digestibility of monostearin (47.4%) (Cheng et al., ’49). The acetostearins, perhaps, may be considered as mixed glycerides rather than as simple glycerides, since they are glycerol esters of acetic acid and stearic acid. There­ fore, it might be expected that the acetostearins would be better utilized or digested than the monostearins. A mixture in equal amounts of acetostearin (melting point of 30.7°C., digestibility coefficient of 80%, and containing approximately 64% stearic acid) and acetoolein (digesti­ bility coefficient of 99.2%) was found to be approximately 93.7% digestible. This would suggest better utilization of acetostearins when mixed with a highly digestible fat, such as acetoolein. Comparing the digestibility of this mixture (93.7%) with that obtained for Crisco (95.9%), a partially hydrogenated fat, suggests that suitable mixtures of aceto­ stearins and acetooleins would be utilized as well as or better than Crisco. Augur et al. ( ’47) have reported data on the improvement in the digestibility of various hydrogenated fats with emulsi­ 122 A. M. AMBROSE AND D. J. ROBBINS fying agents such as lecithin. Inasmuch as improvement in the digestibility of the acetostearins was accomplished by mixing with acetoolein, the use of emulsifying agents was not tried. However, there is no reason to believe that emulsi­ fying agents would not also improve digestibility of the acetostearins, since Augur et al. ( ’47) found a definite im­ provement in digestibility of high-melting hydrogenated oils with lecithin. 'Cheng et al. ( ’49) have shown that the digestibility of low- melting fats was not influenced by the presence of calcium and magnesium in the diet, but that the digestibility of high- melting fats was decreased. In our studies, we have found no great differences in the digestibility of the acetostearins on a diet containing 0.7% of calcium and 0.08% of magnesium derived from natural constituents of the diet. X-ray examina­ tions of the long bones of the lower extremities of rats on diets containing 20% of acetoglycerides for approximately 400 days showed no difference in bone calcification as com­ pared to rats on the basic diet alone, which would indicate that the amount of calcium and magnesium in the diet was adequate for calcification. Our x-ray observations are in agreement with those in studies reported hy Harris and Sherman ( ’54) on the effect on the length of the tibiae of rats when mono-, di-, or triglycerides, and a mixture of mono-, di- and triglycerides were fed in concentrations of approximately 25% of the diet. Table 3 includes data on the amount of fatty acids excreted as soap. It will be noted that the amount of soap varied with the acetoglyceride fed. With the acetooleins the amount of fatty acids excreted as soap was the lowest, while with the acetostearins it was the highest. When acetostearin and aceto­ olein were fed together in equal amounts, the amount of fatty acids excreted as soap decreased. This would suggest that the acetostearins in a suitable mixture with other fats and oils, such as a margarine-like product, containing approxi­ mately 80% of acetoolein and 20% of highly hydrogenatec METABOLISM OF ACETOGLYCERIDES 123 cottonseed oil, as proposed by Gros and Feuge ( ’54), would be completely digestible. Our observations on the digestibility of acetoglycerides are consistent with the observations of others on the digestibility of oils predominantly unsaturated as compared with hydro­ genated fats. It would appear, therefore, that as the per­ centage of stearic acid of a fat (or oil) is increased by hydro­ genation, there is a concomitant decrease in digestibility and a concomitant increase in the amount of fatty acids excreted as soap.

SUMMARY AND CONCLUSIONS Our results on the absorption and digestibility of aceto­ glycerides (acetostearin, acetoolein) justify the following con­ clusions : 1. The acetooleins appear to be better absorbed than the acetostearins over a 4-hour test period in rats. However, the acetostearins are absorbed at about the same rate as Crisco under the conditions of our experimental procedures. 2. The various acetoglycerides appear to be utilized by the rat in much the same way as natural fats. The digestibility coefficients obtained were, approximately 99% for the aceto­ oleins; 80 and 80.8% for the acetostearins melting at 30.7°C. and 57 °C., respectively, and 85% for the acetostearin melting at 32.5°C.; for the mixtures of 15% of acetostearin and 5% of acetoolein, and for 10% of acetostearin and 10% of aceto­ olein, the digestibilities were 85.1 and 93.7%, respectively. 3. Data obtained on the digestibility of the acetoglycerides indicate that the acetostearins are not as readily digestible as the acetooleins. This, in all probability, is due to the large amount of stearic acid formed during digestion, which com­ bines with calcium to form calcium stearate which is not readily utilized. 4. The digestibility of acetostearin is markedly improved when mixed with acetoolein, as indicated by a decrease in the amount of fatty acid excreted as soap. This would suggest that the acetostearins in a suitable mixture with other edible 124 A. M. AMBROSE AND D. J. ROBBINS

fats and oils, such as a margarine-like product containing ap­ proximately 80% of acetoolein and 20% of acetostearin would be completely digestible.

ACKNOWLEDGMENTS The authors are grateful to the Southern Utilization Re­ search Branch for making available, through Distillation Products Industries, adequate supplies of the various-aceto- glycerides studied and for the data contained in table 1.

LITERATURE CITED

A ugur, V., H. S. R ollman and H. J. Deuel, Jr. 1947 The effect of crude lecithin on the coefficient of digestibility and rate of absorption of fat. J. Nutrition, 33: 177. Baur, J. 1954 Acetin fats. I. Products made from mixed acetin fats. J. Am. Oil Chem. Soc., 31: 147. Cheng, A. L. S., M. G. Morehouse and H. J. Deuel, Jr. 1949 The effect of the level of dietary calcium and magnesium on the digestibility of fatty acids, simple triglycerides and some natural and hydrogenated fats. J. Nutrition, 37: 237. Crockett, M. E., and H. J. Deuel, Jr. 1947 A comparison of the coefficient of digestibility and rate of absorption of several natural and artificial fats as influenced by melting point. Ibid., S3: 187. D eu el, H . J ., J r ., L . H a l l m a n and A. L eonard 1940 The comparative rate of absorption of some natural fats. Ibid., 2 0 : 215. F euge, R. 0. 1955 Acetoglycerides — New fat products of potential value to the food industry. Food Tech., 9: 314. Gros, A. T., and R. O. F euge 1954 Consistency of fats plasticized with aceto- glyceridcs. J. Am. Oil Chem. Soc., 31: 377. H a r r is, R. S., and H. S h e r m a n 1954 Comparison of the nutritive values of mono-, di-, and triglycerides by a modified pair-feeding technique. Food Research, 19: 257. Hoagland, R., and G. G. Snider 1940 Nutritive properties of certain animal and vegetable fats. U.S.D.A., Bull. 725: 1-12. L y m a n , J. F. 1917 Metabolism of fats. Utilization of palmitic acid, glyceryl palmitate and ethyl palmitate by the dog. J. Biol. Chem., 32: 7. M a t t il , K. F., and J. W. H iggin s 1945 The relationship of glyceride structure to fat digestibility. Synthetic glycerides of stearic and oleic acids. J. Nutrition, 29: 255. S te e n b o c k , H., M. H. I r w in and J. W eber 1936 The comparative rate of absorption of different fats. Ibid., 12: 103. MULTIPLE CONGENITAL ABNORMALITIES IN THE RAT RESULTING PROM RIBOFLAVIN DEFICIENCY INDUCED BY THE ANTIMETABOLITE GALACTOFLAVIN

MARJORIE M. NELSON, CATHEBINE D. C. BAIBD, HOWABD Y. W IG H T AND HERBERT M. E V A N S 1 Institute of Experimental Biology, University of California, Berkeley

(Received for publication August 22, 1955)

The pioneering studies of Warkany and coworkers, sum­ marized in 1945, demonstrated that chronic riboflavin de­ ficiency in the rat resulted in multiple skeletal abnormalities in the offspring. The experimental procedure used was that of individually controlled riboflavin depletion of rats main­ tained on either vegetable protein or purified casein- containing diets, low in the vitamin. Approximately one-third of the young showed a definite pattern of skeletal abnormal­ ities, which included shortening of the long bones and man­ dible, fusion of the ribs and toes, and cleft palate. Malforma­ tions of the soft tissues were rarely observed. These studies have been confirmed by Giroud and Boisselot ( ’47) and by other investigators. Grainger et al. ( ’54), using similar ex­ perimental procedures, have recently observed hydrocephalus and eye defects in addition to skeletal anomalies. The production of multiple congenital anomalies in high incidence when the folic acid antimetabolite, x-methyl-PGA,

Presented in part at the 120th meeting of the American Association for the Advancement of Science, December, 1953 (Nelson, ’53, ’55), and the 19th annual meeting of the American Institute of Nutrition, April, 1955 (Baird et ah, ’55). This research was aided by grants from the College of Agriculture of the University of California the National Vitamin Foundation, and the Roche An­ niversary Foundation. 125 126 NELSON, BAIBD, WEIGHT AND EVANS was used to accentuate the vitamin deficiency (Nelson, Asling and Evans, ’52; Nelson, Wright, Asling and Evans, ’55) has prompted the following study of the effects of the riboflavin antimetabolite, galactoflavin, on fetal development in the rat.

EXPERIMENTAL Stock female rats of the Long-Evans strain, 60 to 65 days of age, were placed for the entire gestation period on the riboflavin-deficient diets containing varying levels of galacto­ flavin. The levels of galactoflavin tested were: 5, 10, 25, 50 to 54, 100 to 108, 150 and 200 to 216 mg per kilogram of diet. Ten to 12 rats were used for each level tested and practically all levels were repeated. To determine the effect of transitory riboflavin deficiency, other rats were placed for the first 7 days of gestation on the basal riboflavin- deficient diet and then given the deficient diet containing 200 to 216 mg of galactoflavin per kilogram of diet for 2, 4, or 6 days. After the transitory period the animals re­ ceived the basal diet supplemented with a high level of ribo­ flavin, 40 mg per kilogram of diet, until autopsy. All rats were autopsied on the 21st day of gestation, one day before parturition; the fetuses were removed by Cesarean section and examined macroscopically for abnormalities. The basal riboflavin-deficient diet was composed of 18% alcohol-extracted casein, 70% sucrose, 8% hydrogenated vegetable oil,2 and 4% salts no. 4.3 Crystalline vitamins per kilogram of diet were: 300 |ig of D-biotin, 5 mg of 2-methyl- 1,4-naphthoquinone, 5 mg of thiamine HC1, 5 mg of pyri- doxine HC1, 5.5 mg of pteroylglutamie acid, 10 mg of p- aminobenzoic acid, 20 mg of niacin, 50 mg of calcium D-pantothenate, 400 mg of inositol and 1 gm of choline chloride. All rats received weekly a fat-soluble vitamin mix­ ture containing 800 U.S.P. units of vitamin A, 115 chick units of vitamin D, 6 mg of synthetic dbalpha-tocopherol, and

2 Crisco or Primex. 3 Salts no. 4 of Hegsted et al. ( ’41). MULTIPLE CONGENITAL ABNORMALITIES 127

650 mg of corn oil.4 For approximately half of the rats, all crystalline vitamins except choline were doubled and 50 |ig of vitamin B12 and 5 gm of L-cystine added per kilogram of diet. This increased supplementation did not affect the experi mental results so the data for both diets were combined.

T A B L E 1

Effect of riboflavin deficiency induced by galactoflavin throughout gestation on reproduction and fetal development in the rat

YOUNG N O . O F WT. CHANGE GALACTO­ RESORP­ R A T S D U R IN G LITTERS A v . n o . FLAVIN TIONS T o ta l A b ­ BRED GESTATION p e r n o. n o r m a l litte r m g / k g d iet g m % % % Riboflavin-deficient diet None 24 + 69 4 96 211 9.2 0 5-25 35 + 39 3 97 272 8.0 7 50-54 48 + 27 35 65 246 7.9 49

100-108 4 7 + 12 51 49 144 6.3 64 150 24 — 14 88 12 22 7.3 59 200-216 34 — 21 97 3 8 8.0 1

Riboflavin-supplemented diet

216 35 - f 114 0 100 299 9.5 0

1 These young were eaten by the mother at parturition so were not examined for abnormalities.

RESULTS

Effect of riboflavin deficiency induced by galactoftavm ingestion throughout gestation Little effect on fetal development was observed when stock female rats were placed on the basal riboflavin-deficient diets for the entire gestation period, provided that the antimetab­ olite was not added to the diet (table 1). Ninety-six percent of the rats had living young at autopsy and no abnormal young were found. The pregnant rats did not gain as much during gestation as vitamin-supplemented controls, i.e. 69 gm in comparison with 114 gm average, so that a mild vitamin

Mazóla. 128 NELSON, BAIED, WRIGHT AND EVANS

deficiency Avas produced by the diet in this three-week period. The addition of 5 to 25 mg of galactoflavin per kilogram of diet accentuated the deficiency as the mothers gained less weight during gestation (39 gm) and a few young (7%) showed macroscopic abnormalities. As the galactoflavin level was raised and the vitamin deficiency progressively accen­ tuated, the maternal weight gain Avas decreased and the in­ cidence of abnormal young and of fetal death increased cor­ respondingly. With the 50 to 54 mg level of galactoflavin, fetal death and resorption occurred in one-third of the ani­ mals and half of the living young were abnormal. Raising the level to 100 to 108 mg of galactoflavin resulted in fetal death and resorption in half of the animals and two-thirds of the young were abnormal. In the rats that carried young to term the average number of living young per litter was only 6.3 in contrast to control values of more than 9. Higher levels of galactoflavin, 150 and 200 to 216 mg, resulted in fetal death and resorption in practically all animals. Ribo­ flavin supplementation, 20 mg per kilogram of diet, reversed all effects of the highest level of galactoflavin used, as shown by the normal maternal Aveight gain and reproductive per­ formance of the Adtamin-supplemented controls (table 1). In the 251 abnormal young from rats receiving Amrious levels of galactoflavin, the predominant anomalies Averc the skeletal defects previously reported by other investigators. Eighty-six percent of the abnormal young exhibited one or more of the following anomalies:. tail defects, syndactylism, clubfoot, short mandible or cleft palate (table 3). In addition, other types of anomalies, those of the soft tissues, Avere pres­ ent in 41% of the abnormal young. These anomalies included defects of the interventricular septum of the heart and of the aortic arch pattern, hydronephrosis, hydroureter, unde­ scended testes, abnormal position of the ovaries, umbilical or diaphragmatic herniations, hydrocephalus, “ open” eyes, microphthalmia, epidermal defects and subcutaneous edema. Both skeletal and soft tissue anomalies Avere observed at all levels of galactoflavin used. The higher levels of galactoflavin MULTIPLE CONGENITAL ABNORMALITIES 129

increased the incidence of abnormal young but did not affect the pattern of anomalies produced.

Effect of transitory riboflavin deficiency induced by galactoflavin As noted previously in table 1, the basal riboflavin-deficient diets which did not contain the antimetabolite had little effect on fetal development. Table 2 shows that the addition of high levels of galactoflavin to the deficient diet for two days from days 7 to 9,5 followed by vitamin supplementation for the remainder of gestation, likewise had little effect on

TABLE 2 Effect of transitory riboflavin deficiency induced by galactoflavin on reproduction and on fetal development in the rat

YOUNG TRANSITORY N O . O F WT. CHANGE RESORP­ DEFICIENCY R A T S D U R IN G LITTERS Av. no. TIONS Total Ab­ P E R IO D 1 BRED GESTATION per no. litter normal days of gestation grn % % %

7-9 23 + 110 0 100 212 9.2 l 7-11 24 + 89 33 67 132 8.3 32 7-13 24 + 74 50 50 65 5.4 68

1 The riboflavin-deficient diet contained 200 to 216 mg of galactoflavin per kilo­ gram during this period. fetal development. All rats had living young at autopsy and only three abnormal young (1%) were found. Four days of galactoflavin ingestion from days 7 to 11 with vitamin supplementation beginning on day 11 resulted, however, in fetal death and resorption in one-third of the animals and one-third of the young were abnormal. The defects in these abnormal young were limited to the cardiovascular system (table 3) and in practically all cases consisted of a persistent interventricular foramen. With 6 days of galactoflavin inges-

5 The day of finding sperm is considered to be day zero. Thus, the period of transitory galactoflavin ingestion from days 7 to 9 is started at 10 a . m ., 7 days

fter finding sperm and ends 48 hours later, at 10 a . m ., 9 days after finding sperm. 130 NELSON, BAIIïD, WEIGHT AND EVANS

vO o o

£

Ci o

* s to o

o # M o

00to o

o> o £ m

o

o> CO o £ CO 0 3

Cl o £ CO

00 o £ Ti<

rH

tOo to rHo Od I I 0 3! tO o o r 5 o o 0 3 03

o MULTIPLE CONGENITAL ABNORMALITIES 131

tion from days 7 to 13 and vitamin supplementation begin­ ning on day 13, fetal death and resorption occurred in half of the animals and two-thirds of the young were abnormal. The average number of lving young per litter decreased from control values of 9 or more to 5.4. The abnormal young from this group exhibited multiple types of defects, 66% showing skeletal anomalies and 59% anomalies of the soft tissues (table 3). The incidence of cardiovascular anomalies was 48%, considerably higher than that observed when galacto- flavin was given continuously throughout gestation. In all of the abnormal young obtained from rats subjected to transitory galactoflavin ingestion, defects of the cardiovas­ cular system predominated (72%) while skeletal defects were found in only 34% of the offspring. The cardiovascular anomalies observed in the fetuses from rats receiving galactoflavin during these 4- to 6-day periods as well as throughout gestation included defects of the heart and anomalies of the aortic arch pattern. In 117 fetuses with such anomalies, 113 exhibited interventricular septal defects and 18 showed defects of the aortic arch pattern. The aortic arch anomalies included double or right aortic arch, absence of the arch of the aorta, absence or hypoplasia of the ductus arteriosus, and distally arising right subclavian artery.

DISCUSSION This study demonstrates the multiple teratogenic effects of riboflavin deficiency when induced by the antimetabolite galactoflavin. The skeletal anomalies produced were similar to those observed by previous investigators, all of whom utilized a chronic vitamin deficiency. Cerebral or eye defects were observed only rarely, less than 1% incidence, by Wark- any ( ’45) whereas Grainger et al. ( ’54) reported consider­ ably higher incidence, 24% and 15% respectively. In the latter study the incidence of these defects was markedly de­ creased by addition of vitamin B,, to the riboflavin-deficient diet. In the present study the incidence of cerebral and eye defects was low, although higher than that observed by 132 NELSON, BAIKD, WEIGHT AND EVANS

Warkany, and supplementation with additional vitamin B i2 did not change the incidence. Inasmuch as the stock rats used in this short-term study were not depleted of vitamin B12, additional B12 would not he expected to have an effect. Cardiovascular anomalies have not been previously re­ ported for riboflavin-deficient offspring but have been ob­ served in studies on chronic vitamin A deficiency (Wilson and Warkany, ’49) and in both continuous and transitory folic acid deficiency induced during pregnancy by the antimetab­ olite, x-methyl-PGA (Baird et al., ’54). Other types of soft tissue anomalies (those of the urogenital system, the dia­ phragm, and the epidermis) have likewise not been previously reported for riboflavin deficiency. Tlie present study emphasizes the relationship of timing and duration or severity of the deficiency induced during preg­ nancy to the anomalies produced. The continuous deficiency resulting from the use of the antimetabolite throughout gesta­ tion resulted in young with predominantly skeletal defects. Both skeletal and cardiovascular anomalies were prevalent when the antimetabolite was given for the 6 days from days 7 to 13 during gestation whereas the ingestion of galacto- flavin for the 4 days from days 7 to 11 resulted in defects of only the cardiovascular system. In the latter group the prevention of skeletal and other types of defects was pre­ sumably due to the vitamin supplementation instituted on day 11. Apparently the cardiovascular tissues were more sensitive to the induced riboflavin deficiency than any other embryonic tissues during days 7 to 11. The authors are unaware of any other teratogenic agent or experimental procedure in mammals that has resulted in anomalies of only the cardiovascular system. The prevention of skeletal defects in this group is in agreement with the results obtained by Warkany et al. ( ’42) who were able to prevent such defects due to chronic riboflavin deficiency by supplementation on the 12th day 6 of gestation but not on the 13th day. “ In citing the studies of Warkany and coworkers, the timing has been changed to correspond with the timing used in this paper in which the day of finding sperm is considered to he day zero. MULTIPLE CONGENITAL ABNORMALITIES 133

The two-day period of transitory galactoflavin ingestion from days 7 to 9 was apparently too short a period to produce a riboflavin deficiency sufficiently severe to affect fetal de­ velopment. It is possible, however, that embryonic develop­ ment during these days may not be sensitive to riboflavin deficiency. The normality of development in 99% of the fetuses from this group may be contrasted with the high incidence of fetal death and resorption (62%) and of ab­ normal young (51%) observed with transitory folic acid deficiency induced by low levels of x-methyl-PCA given dur­ ing the same period, days 7 to 9 (Nelson et al., ’55).

SUMMARY Little effect on fetal development was observed when stock female rats of the Long-Evans strain were given riboflavin- deficient diets for the period of gestation. However, when the vitamin deficiency was accentuated by the addition of the antimetabolite, galactoflavin, to the deficient diet for the entire gestation period or for only 4 to 6 days early in gestation, a high incidence of fetal death or congenital ab­ normalities resulted. The abnormalities observed included those of the skeleton, the cardiovascular and urogenital sys­ tems, the cerebrum and the eyes; herniations of the diaphragm and body walls were also observed. Skeletal defects pre­ dominated when the antimetabolite was given throughout gestation. Both skeletal and cardiovascular anomalies were prevalent when the antimetabolite was given from days 7 to 13 followed by vitamin supplementation for the remainder of gestation. Cardiovascular defects were the only anomalies 'esulting from the transitory deficiency induced by galacto- avin ingestion from days 7 to 11. When diets containing he highest level of the antimetabolite were supplemented with riboflavin throughout gestation, fetal development was normal. ACKNOWLEDGMENTS The authors wish to express their appreciation to Dr. Karl A. Folkers of Merck and Company, Inc., Rahway, New Jersey, 134 NELSON, BAIRD, WRIGHT AND EVANS for generous supplies of the antimetabolite, galactoflavin, to Dr. E. L. Severinghaus of Hoffmann-LaRoche, Nutley, New Jersey for crystalline D-biotin, calcium D-pantothenate and alpha-tocopherol, and to Dr. T. H. Jukes, Lederle Labora­ tories, Pearl River, New York for synthetic pteroylglutamic acid. LITERATURE CITED

B aird, C. D. C., M. M. N e l s o n , I. W. M o n ie a n d H. M. E v a n s 1954 Con­ genital cardiovascular anomalies induced by pteroylg.utamic acid deficiency during gestation in the rat. Circulation Research, 3 : 544. B aird, C. D. C., M. M . N e l s o n , I . W. M o n ie, H. V. W rig h t a n d H. M . E v a n s 1955 Congenital cardiovascular anomalies produced with the ribo­ flavin antimetabolite, galactoflavin. Fed. Proc., 14: 428. G iroud, A., and J. B o isse l o t 1947 Repercussions de l ’avitaminose B, sur l’embryon du rat. Arch, franc. Ped., 4: 1. G raing er, R. B., B. L. O’D ell a n d A. G. H ogan 1954 Congenital malforma­ tions as related to deficiencies of riboflavin and vitamin B12, source of protein, calcium to phosphorus ratio and skeletal phosphorus metabolism. J. Nutrition, 54: 33. H egsted, D. M., R. C. M il l s , C. A. E l v e h je m an d E. B. H ar t 1941 Choline in the nutrition of the chick. J. Biol. Chem., 138: 459. N e lso n , M. M. 1953 Mammalian fetal development and antimetabolites. 120th A.A.A.S. Meeting, Med. Sci. Section. ------1955 Mammalian fetal development and antimetabolites. A N TI­ METABOLITES AND CANCER, p. 107, published by A.A.A.S. N e l s o n , M. M., C. W. A s l in g a n d H. M. E v a n s 1952 Production of multiple congenital abnormalities in young by maternal pteroylglutamic acid deficiency during gestation. J. Nutrition, 48: 61. N e l s o n , M. M., H. V. AVr ig h t , C. W. A s lin g an d H. M. E v a n s 1955 Multiple congenital abnormalities resulting from transitory deficiency of pteroylglutamic acid during gestation in the rat. Ibid., 56: 349. W a r k a n y , J. 1945 Manifestations of prenatal nutritional deficiency. V IT A ­ MINS AND HORMONES, 3 : 73. W a r r a n t , J., R. C. N elso n a n d E. S chraffenberger 1942 Congenital mal­ formations induced in rats by maternal nutritional deficiency. IT. Use of varied diets and of different strains of rats. Am. J. Dis. Children, 64: 860. W il s o n , J. G., a n d J. W a r k a n y 1949 Aortic arch and cardiac anomalies in the offspring of vitamin A deficient rats. Am. J. Anat., 85: 113. STUDIES OF REPRODUCTION IN RATS USING LARGE DOSES OF VITAMIN B12 AND HIGHLY PURIFIED SOYBEAN PROTEINS

LUTHER R. RICHARDSON AND RUTH BROCK Department of Biochemistry and Nutrition, Texas Agricultural Experiment Station, Texas Agricultural and and Mechanical College System, College Station

(Received for publication August 15, 1955)

The essential nature of dietary vitamin B12 for female rats to wean the maximum number of young and for the young to attain normal weaning weight has been demonstrated by several investigators (Schultze, ’49; O ’Dell, Whiteley and Hogan, ’51; Richardson and Russell, ’51; Dryden, Hartman and Cary, ’51, ’52). Meyer, Thompson and Elvehjem ( ’51) determined the effect of various supplements to a corn-soybean soil meal diet on lactation and on the vitamin B12 content of milk of rats. The B12 activity of the milk varied with that of the diet, but lactation performance did not necessarily parallel the vitamin B12 content of the diet. Only fish solubles allowed normal reproduction and lactation. Although the fish solubles in­ creased the B12 content of the milk above that of the basal group, the milk from the females receiving fish solubles still did not contain more than one-fifth as much B12 as that from females receiving 10 ng of crystalline vitamin B12 per 100 gm of diet. It was suggested that the difference in response to fish solubles and vitamin B]2 was due either to an unidentified factor in the fish solubles or to the toxicity of the high level of vitamin B12 in milk of the females which received vitamin B12. Since vitamin B12 is administered to practically every 135 136 LUTHER R. RICHARDSON AND RUTH BROCK pregnant women, sometimes in relatively large amounts, it is desirable to know whether the administration of large amounts of the vitamin to the mother has a detrimental effect on the survival and well-being of the offspring. Schultze ( ’50, ’53) and Schultze, Liener and Glass ( ’52) reported that female rats receiving a diet containing purified soybean protein supplemented with vitamin B12 failed to rear a normal number of young to weaning age. Piccioni, 'Rabbi and Moruzzi ( ’51) reported that crude casein contained a factor (or factors) which was indispensable for normal growth and reproduction in rats. When washed casein was substituted for crude casein, the mortality rates in the first and second generation were 70 and 100% respectively. With crude casein the survival usually was close to 100%. The factor was water soluble and heat stable, and was not identified with any of the recognized B vitamins. Sherman, Schilt and Schaefer ( ’55) reported that the addition of natural ingredients such as distillers solubles, liver, liver meal and fish meal to a vegetable protein diet produced a growth response in the off­ spring above that obtained with vitamin B1=. Recently Gander and Schultze ( ’55) questioned whether the results of the Italian investigators, as well as some of their own, should be interpreted as conclusive evidence of the existence of an unidentified “ animal protein factor” required for the survival of young rats. In another report Schultze ( ’55) described the results of a study with a diet containing 16 amino acids as the only source of protein nitrogen. Parent females which had received the amino acid diet from weaning produced 11 litters of 76 young. Seventy-three, or 96%, of the young were weaned, but their rate of growth was subnormal. Further­ more, these observations do not support the claim that an unrecognized “ animal protein factor” is required for opti­ mum lactation in the rat. An investigation has been in progress in this laboratory for several years on the nutritive requirement for reproduc­ tion and lactation in rats. In experiments conducted at different times with the same diet the numbers of young that DIET AND REPRODUCTION IN RATS 137 were weaned varied; frequently it was difficult to interpret the data with regard to the requirement of some of the more obscure nutrients for reproduction in the rat. During the past three years, uniform results have been obtained in studies which were conducted on (1) the effect of large doses of vitamin B12 in the diet of the mother on the survival and well-being of the offspring, and (2) on the effect of different protein concentrates including highly purified soybean pro­ teins on reproduction and lactation in rats.

EXPERIMENTAL Female rats which had received the soybean oil meal diet (diet 2) without added vitamin B12 from weaning were given the experimental diets when they weighed 150 to 160 gm. Food and water were supplied ad libitum. The composition of the three basal diets used in these studies is given in table 1. The temperature of the animal room was automatically regulated at about 75° F. during the cool months, and re­ frigerated air conditioners kept the temperature below 85° F. during the hot periods. Each female was given an opportunity to produce 4 litters and all litters were reduced to 8 young at birth. After the females in the parent generation had received the experimental diet for 4 weeks, a mature male from the stock colony was placed in a cage with 5 females. The females in succeeding generations received the experimental diet throughout their life span. They were weighed weekly and a pregnant female, as indicated by a large increase in weight, was transferred to an individual maternity cage of the Wah- mann type. A false bottom 6 inches wide and turned up three inches at the outer edge was placed in the back of the maternity cage before the litter was born, and remained there until the young were 8 to 10 days old. The false bottom extended the entire length of the cage and was covered with shredded paper in advance of partuition so that the female might make a nest for the young. When this procedure was fol­ lowed, consistent results were obtained. 138 LUTHER R. RICHARDSON AND RUTH BROCK

TABLE 1 Composition of diets

DIETS CONSTITUENT 1 2 3 gm gm gm Protein concentrate 25.0 Soybean oil meal 70.0 21.4 Dried skim milk 22.0 Dried yeast 8.0 Alfalfa leaf meal 5.0 Cerelose 62.7 23.0 Ground milo 35.6 Wood pulp 3.0 Mazola 5.5 3.0 5.5 DL-metMonine 0.3 Mineral mixture 1 5.0 4.0 Steamed bone meal 2.0 NaCl 0.5 A.D.K.2 + + + Alpha tocopherol, mg 2.5 1.0 Vitamin mixture 3 A B

1 Richardson and Hogan ( ’46). 2A.D.K. mixture supplied the following: vitamin A 3000 I.U., vitamin I) 400 I.U. and menadione 0.5 gm per 100 gm. 8 Vitamin mixture A supplied the following: thiamine • HC1, riboflavin and pyridoxine ■ HC1 1 mg each, calcium pantothenate 4 mg, niacin 5 mg, choline 100 mg, inositol 10 mg, para aminobenzoic acid 50 mg, folic acid 0.1 mg and biotin 0.02 mg per 100 gm of diet. In vitamin mixture B PABA, folic acid and biotin were omitted. 4 The following salts were added: ferric citrate 40 mg, CuS04 ■ 5 H20 20 mg and MnS04 • H20 100 mg per 100 gm.

Effect of nesting material at birth on survival of young Since a large percentage of the young survived when nesting material was placed in the maternity cage before partuition, a test was carried out to determine whether the addition of the nesting material before or immediately after partuition had any effect on the survival and weaning weight of the young. Two groups of 10 females each were given diet 1 supple­ mented with 25 ng of crystalline vitamin B12 per kilogram of diet. Shredded paper was provided for one group of DIET AND REPRODUCTION IN RATS 139 females before birth of a litter and for the other immediately after birth. These data are summarized in table 2. When the paper was added before birth, 100% of the litters and 95% of the young were weaned. When it was added after birth, only 78% of the litters and 70% of the young were weaned. These data emphasize that it is essential to maintain uniform conditions in management of the colony in investiga­ tions o‘n the nutritional requirement for reproduction and lactation.

Vitamin B 12 and reproduction

Diets containg soybean protein 1, vitamin-free casein and crude soybean oil meal were used in studies with different amounts of crystalline vitamin B12 in the diets. These data are summarized in table 3. Two series of experiments were conducted with soybean protein and different amounts of vitamin B 2. In one series (A) three groups of females received 0, 10 and 25 Mg of crystalline vitamin Bi2 per kilo­ gram of diet respectively. In a 4th group each female received one dose of 100 Mg of vitamin B12 per week by subcutaneous injection. In series B, 4 groups of females received 0, 25, 500 and 1000 Mg, respectively, of crystalline vitamin B12 per kilogram of diet. In series C, three groups of females received the basal diet containing vitamin-free casein and supplemented with 0, 25 and 500 Mg of vitamin B12. And in series D, 4 groups of females received the basal diet containing crude soybean oil meal (diet 2) supplemented with 0, 10, 100 and 500 Mg of vitamin B12 respectively. Two groups of females (series E) received a diet com­ posed of natural foodstuffs (diet 3) with and without vitamin B12. When vitamin B12 was added, the average weaning weight, the average number of young weaned per litter and the percentage of litters anu of young weaned were greater (with the exception of diet 3) than without vitamin B12. Regardless

Drackett 220 or C-l, obtained from The Draekett Company, Cincinnati. Ohio. TABLE 2 Influence of nesting material at birth on the survival of the young

AVERAGE NUMBER OF YOUNG LITTERS S H R E D D E D YOUNG PER LITTER PAPER ADDED 1 Obs. Weaned Av. wt. Obs. Weaned Born Obs.2 Weaned no. % gm no. % Before birth 279 95 34.4 39 100 8 . 3 7.2 6.8 After birth 254 70 33.1 41 78 7.5 6.4 5.6

1 Ten females per group. 2 Average number observed after the litters containing more than 8 young had been reduced to 8.

TABLE 3 Vitamin B^ and reproduction in rats

AVERAGE NUMBER OF YOUNG VITAMIN NO. OF YOUNG PER LITTER Bia FEMALES Obs. Weaned Ay. wt. Born Weaned Born Obs.2 Weaned no. % gm no. % Soybean protein (diet 1) Series A None 10 258 43 31.7 41 49 8.1 6.9 5.5 10 9 259 95 37.6 37 95 8.2 7.0 7.0 25 10 243 92 39.3 34 94 8.8 7.1 7.0 100 1 10 256 92 35.1 39 92 8.0 6.6 6.5

Series B None 10 272 69 28.8 51 67 6.9 5.6 5.5 25 10 274 87 35.6 40 92 7.8 6 . 9 6.5 500 10 230 96 35.5 30 100 9.1 7 . 6 7.3 1000 8 86 99 35.0 12 100 8.1 7.2 7.1

Series C Vitamin-free casein (diet 1) None 21 449 61 29.2 71 68 6.9 6.2 5.9 25 10 302 95 39.0 40 100 9.0 7.6 7.2 500 10 291 93 37.9 40 100 8.8 7.3 6.8

Series D Soybean oil meal (diet 2) None 30 706 38 25.0 114 51 6.9 6.2 4.6 10 38 492 96 36.1 70 96 8.9 7.1 7.0 100 10 164 99 35.9 21 100 10.0 7.8 7.7 500 10 161 93 34.9 21 95 9.5 7.6 7.4

Series E Natural foodstuffs (diet 3) None 10 285 95 33.2 38 100 8.7 7.5 7.1 25 10 315 91 34.2 40 98 9.9 7.9 7.3

1 Injected per rat per week. 2 Average number observed after the litters containing more than 8 young had been reduced to 8. 140 DIET AND REPRODUCTION IN RATS 141 of whether the amount of vitamin B12 supplied was 10 or 1000 |jg, a high percentage of the young and litters was weaned. The young at 21 days were slightly heavier than those which received a practical diet. The number of young weaned without vitamin B12 varied in different experiments. The females which received soybean protein weaned only 43% of the young in series A and 69% in series B. Those in series C, which received vitamin-free casein without vitamin B12, weaned 61% of the young, and those which received soybean oil meal (series D) without vitamin B12 weaned 38%. It is obvious that females either had various amounts of the vitamin stored in their bodies or their diets contained different amounts of the vitamin. These data show that the mother can receive a dose of vitamin B12 at least 100 times that required for satisfactory reproduction without any ill effect upon the offspring. These observations are in agreement with those of Daniel, Gardiner and Ottey ( ’53) who found no evidence that milk from females re­ ceiving 200 |jg of vitamin B12 per kilogram of diet was toxic to the young.

Unidentified factors for reproduction and lactation In the study described in the previous section with large doses of vitamin B12, there was no evidence that the diet was deficient in any factor required for reproduction and lactation in rats. However, in view of the reports in the literature that unrecognized vitamins are required for re­ production in rats, attempts were made to remove possible unidentified factors from two commercial soybean proteins. Soybean protein and sodium proteinate 2 were used for this purpose. The extraction procedure for each protein was as follows : approximately 25 pounds of protein were mixed with 18 to 22 gallons of water in a 30 gallon kettle and boiled gently with constant stirring for 30 minutes. The protein

’ Archer-Oaniels Company, Minneapolis, Minnesota. 142 LUTHER R. RICHARDSON AND RUTH BROCK was allowed to settle and the supernatant liquid was syphoned off. Water was added and the mixture was boiled a second time for 30 minutes. After the protein settled, the supernatant liquid was syphoned off and the protein was filtered in a large Buchner funnel. The water-extracted protein was heated in a steam-jacketed kettle with 8 gallons of 95% isopropyl alco­ hol and stirred constantly for 30 minutes. The hot mixture was filtered through the Buchner funnel and the residue was

TABLE 4 Purified soybean proteins and reproduction in rats

AVERAGE NUMBER OP YOUNG LITTERS GENERA- N O . O P YOUNG PER LITTER T IO N N O . FEMALES Obs. Weaned A v . w t . Born Weaned Born Obs.1 Weaned no. % gm no. % Purified soybean protein2 P 19 517 94 35.0 74 100 8.3 7.0 6.0 F, 15 400 90 35.5 58 89 8.6 7.0 6.9 f 2 15 360 96 35.8 54 100 8.4 6.7 6.4 f 3 15 391 98 34.5 60 98 8.3 6.6 6.6 F, 15 138 100 37.3 21 100 7.0 6.6 6.6

Purified sodium proteinate® P 10 217 82 36.3 32 82 8.6 6.8 6.8 F, 10 268 94 39.3 38 100 8.3 7.1 6.6 f 2 10 277 94 32.1 40 98 7.7 6.9 6.7 f 3 10 174 94 34.2 28 96 6.5 6.2 6.0

1 Average number observed after the litters containing more than 8 young had been reduced to 8. 2Drackett 220, obtained from The Drackett Company, Cincinnati, Ohio. 3 Archer-Daniels Company, Minneapolis, Minnesota. re-extracted 5 times by the same procedure with the excep­ tion that 70% isopropyl alcohol was used instead of 95%. The extracted protein was air-dried, ground in a hammer mill and stored at room temperature. The first alcohol ex­ tract had a strong yellow color while the last was almost colorless. The results with the two soybean proteins which were purified by this procedure are summarized in table 4. The composition of the diets containing the purified soybean DIET AND REPRODUCTION IN RATS 143 proteins was the same as diet 1 supplemented with 25 qg of crystalline vitamin B12 per kilogram of diet. The females in the parent, F x, F2 and F 3 generations which received the purified soybean protein weaned 90 to 98% of their young. The females in the F4 generation were allowed to produce only one to two litters each, but all of 138 young were weaned when the observations were discontinued. The parent generation females which received the diet containing purified sodium proteinate weaned 82% of their young, and those in F 1? F 2 and F 3 generations weaned 94%. It is evident from these data that the diets were adequate for reproduction and lactation. Furthermore there was no evidence that an unidentified factor or factors required for reproduction in rats was removed from the protein by thorough extraction with boiling water and boiling 70% al­ cohol. Likewise the females were not depleted of any unidenti­ fied factors by maintaining them on the same diet through 4 to 5 generations. These conclusions are in agreement with the most recent observations of Schultze ( ’§5) who reported that 96% of 76 young were weaned by mothers which received a diet containing 16 pure amino acids as the only source of protein.

SUMMARY Two studies are described concerning reproduction in rats with the use of purified diets. One is the effect of large doses of vitamin B12 on the survival of young, and the other is the effect of different protein concentrates, including highly purified soybean protein, on reproduction. When vitamin B12 was added, the number of young weaned ranged from 90 to 100% regardless of whether the amount added was 10, 25, 500 or 1000 qg per kilo of diet. There was no evidence that large doses of vitamin B12 in the diet of the mother were "toxic to the offspring. Without vitamin B12, the average weaning weight and the percentage of litters and of young weaned were less than with vitamin Bi2. 144 LUTHER R. RICHARDSON AND RUTH BROCK

Two commercial soybean proteins were exhaustively ex­ tracted with boiling water and with boiling 70% isopropyl alcohol. Five generations of females which received a syn­ thetic diet that contained these purified proteins weaned 90% or more of their young. These data show that the diet was not deficient in an un­ identified factor or factors required for reproduction and lactation in rats, and it is concluded that if unidentified fac­ tors are required for reproduction in rats, they are held very tenaciously to the protein concentrates.

ACKNOWLEDGMENTS Acknowledgment is made to Merck and Co., Inc. for many vitamins, to Lederle Laboratories for folic acid and to Dow Chemical Company for the DL-methionine used in these studies. LITERATURE CITED

D a n ie l , L. J., M. G ard in er and L. J. O t t e y 1953 Effect of vitamin B,, in the diet of the .rat on the vitamin B13 contents of milk and livers of young. J. Nutrition, 50: 275. D ry d e n , L. P., A. M. H a r t m a n an d C. A. Ca r y 1951 The relation of vitamin B12 deficiency to fertility of the female and birth weight of the young in rats fed purified casein rations. Ibid., 45: 377. ------1952 The effect of vitamin B12 deficiency upon the survival of young born to rats fed purified casein rations. Ibid., 46: 281. G an der, J. E., and M. 0 . S c h u l tze 1955 Concerning the alleged occurrence of an “ animal protein factor” required for the survival of young rats. I. Studies with unpurified rations. Ibid., 55: 543. M e y e r , M. L., H. T. T h o m p s o n a n d C. A. E l v e h je m 1951 The effect of vitamin B12 on reproduction and lactation in rats receiving pork or beef diets. Ibid., 45: 551. O’D e l l , B . L., J. R. W h it e l e y and A . G. H ogan 1951 Vitamin B ,,. a factor in prevention of hydrocephalus in infant rats. Proc. Soc. Exp. Biol. Med., 76: 349. P ic c io n i, M., A. R abbi and G. M oruzzi 1951 Animal protein factor for the rat present in crude casein and its relationship with vitamin B,2. Science, 113: 179. R ic h a r d s o n , L. R., and A. G. H ogan 1946 Diet of mother and hydrocephalus in infant rats. J. Nutrition, 32: 459. R ic h a r d s o n , L. R., and E. L. R u s s e l l 1951 I olie acid, crystalline vitamin B12 and penicillin in reproduction in rats. EM. Proc., 10: 391. DIET AND REPRODUCTION IN RATS 145

S c h u l t ze , M. 0. 1949 Nutritional value of plant materials. II. Prevention of acute uremia of the newborn rat by vitamin B,:. Proc. Soc. Exp. BioL Med., 72: 613. ------1950 Nutritional value of plant materials. V. Reproduction of rats fed purified rations containing soybean protein. J. Nutrition, 42: 587. ------1953 Nutritional value of plant materials. IX . The effect of various supplements on lactation failure of rats fed diets containing purified soybean proteins. Ibid.. 49: 245. ------1955 Concerning the alleged occurrence of an “ animal protein far tor ’ ’ required for the survival of young rats. II. Reproduction of rats fed protein-free amino acid rations. Ibid., 55: 559. S c h u l t ze , M. O ., I. E . L ien ee an d R . L . G l a s s 1952 Nutritional value of plant materials. V II. Dietary factors affecting pre.-weaning and post-weaning mortality of rats. Ibid., 46: 171. S h e r m a n , W. C., H. L. S c h il t an d H. C. S ch aefer 1955 The use of depleted rats for investigations of vitamin B,2 and unidentified factors. Ibid., 55: 255.