I. Biological Synthesis of Creatine. II. Corneal Vascularity and Ariboflavinosis. III. Root "Bleeding" Citation Dubnoff, Jacob William (1944) I. Biological Synthesis of Creatine. II. Corneal Vascularity and Ariboflavinosis. III. Root "Bleeding". Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/t2ay-dg34. https://resolver.caltech.edu/CaltechTHESIS:04012025-230418750 Abstract No abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Biology) Degree Grantor: California Institute of Technology Division: Biology Major Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Borsook, Henry Thesis Committee: Unknown, Unknown Defense Date: 1 January 1944 Additional Information: Pages 92-94 not available. Record Number: CaltechTHESIS:04012025-230418750 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:04012025-230418750 DOI: 10.7907/t2ay-dg34 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17127 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 03 Apr 2025 18:52 Last Modified: 03 Apr 2025 18:57 Thesis Files PDF - Final Version See Usage Policy. 45MB Repository Staff Only: item control page

I Biological Synthesis of Creatine II Corneal Vascularity and Ariboflavinosis III Root "Bleeding" by Jae ob Dubnoff Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1944 TABLE OF CONTENTS page I The Biological Formation of Creatine The Formation of Creatine from Glycooyamine • • • • • 1 Creatine Formation in Liver and in Kidney • • • • • • 17 A Micromethod for the Determination of Glycocyamine. 22 Formation of Glycocyamine in Animal Tissues • • • • • 30 The Formation of Glycocyamine in Man and Its Urinary. 45 Excretion II A Micromethod for the Determination of Arginine • • • 51 The conversion of Citrulline to Arginine in Kidney. 57 The Relation of Corneal Vasculari_ty to Ariboflavinosis. 79 III Factors in Plant Bleeding • • • • • • • • • • • • • • • • 109 Summary Part I Part II . Part I I I . . . . . . . . . . . . . . . . iv . . . . . . . 91 . • . . . . . . . . . . . • . . . . . . . 12 7 The candidate wishes to express his sincerest appreciation for the opportunity of working with Dr. H. Borsook in the work herein described and for the invaluable experience it has given him. In the following section, a series of reprints is presented of research carried out with Dr. Borsook on the biological synthesis of creatine and its precursors. SUMMA.RY It has been established that liver slices can convert glycocyamine to oreatine. Of the large number of compounds teste~ only methionine can methylate glycocyamine to form creatine. The following animals were able to methylate glycocyamine: ·oat, dog, guinea pig, frog, pigeon, rabbit, and rat. In order to study the formation of glycocya.m.ine a more rapid and accurate method was. developed for this compound. It was shown -that glycocyamine is formed by cell free extracts of' kidney by the transfer of the .a midino group of arginine to glycine. Beef, cat, dog, guinea pig, pigeon, rabbit, rat, and sheep kidney can carry out this reaction. It was shown that a similar mechanism exists· in man, for there is a rapid rise in the amount of glycocyamine of the urine after the ingestion of arginine and glycine. A method was developed which permitted the simultaneous determina- tion of' both arginine and glycocyamine. With this method it was shown that the arginine for the above reaction was f0rmed by the transfer of the amino group of glutamio acid to citrulline. Glycocyamine NH CH2 COOH }¼N'cfNH Methionine CONVERSION OF GLYCOCYAMINE TO CREATINE + s CH2 CH2 CHNH?. COOH CH~ Creatine CH2 COOH NH 'c1/ ~N H?N 1 Reprinted from THE JounNAL OF BIOLOGICAL CHEMIS:rnY Vol. 132, No. 2, February, 1940 THE FQRMATION ·OF CREATINE FROM GLYCOCY-· AMINE ·IN THE. LIVER 'iiBY HENRY BORSOOK AND JACOB w. DUBNOFF : I (From the William G. Kerckhojf. Laborato•ries of the Biological ScierJ,ces,, California Institute of, T ~chnology, Pasadena) . , . ·- (Received for publication,· November 15, 1939) The study of the precurs~rs , of creatine in 3tnimals has been beset by two difficulties :principally. One has been _the lack of really adequate biological material; the other, the lack of a specific, and at the same time sensitive analytical 'method. -~ Experiments hitlierto have consisted in 'a tte~pts to change the urinary excre:tjon of creatine and creatini~e, <>r the creatine content of the tjs.:. sues of intact animals or of isolated perfused organs. The normal, i.e. uncontrolled, fluctuatiollS in tissue composition and urin/4ry excretion are relatively large compared with the changes induced experimentally; it is often imp,ossible to qjstinguir,h when e·xpe~irnent~r ~ffeit& ·are q1:iser':e1·, whether these ' have arise~ •fi:oll¼ cha~ges in . tp.e proces~~s of excretion or synthesis; there may,_be va,ria9ons ,i~J }ie. w~t~r cont,~nt s°f the tissues, thereby affec~i'ii.g their per.ceJ?,We co~posi'tion1,ru\it these ha".'e stood in the wa~b 0f firm conclµs~ns bemg draw.1¼.dT , . ,;. 0 Some o'f.the.se difficulties have been avoided in perfusion ex~~ •v " .,~~r- , uo~, > vi ments on, i~,;,.at,i:id organs (lJ }nese experiments are extrein~_:r laboriou~, . ~e--consuming, . ap.d costly. The experimentally induced change in creatine q~ut~nt is _at the most- about 30~1?~~ cent~a small deviation from,<~1¾e normal. The nor~al base-)ine is not· co,n,st~nt but varies· witn ,t'he age and' ~eight of the animal. Since the sarne specimen ~arinot be used fa/ the experiment and controi,, a· lai-ge number of .ani~~ls must be used first to establish the p,p.rmal base-lin~:i~ncompassing the variations of the norm'al, and t hen an equali_y large n1:1,mber for each single experi:rp.ent in order to obtain -a bqgy of data s,ufficiently Ja!ge for ·a statistical con<;3l:µsion which .'In:ay be s~gn~ficant_ . 1 . , ,. 0 559, , ·-~ ' . .L,-:-,...,-,. 2 560 Creatine Formation in Liver In 1935 we published some observations of a slight increase in in "apparent:' creatine when rat liver slices were incubated with a protein hydrolysate (2). Granted that such liver slices could synthesize creatine, most of the difficulties in working with whole animals or whole organs are avoided. With tissue slices the one specimen of tissue provides material for controls and for testing a variety of experimental variants simultaneously. This was the reason that the observation of even a slight synthesis of creatine by tissue slices seemed promising. ' We were faced here, however, with the second difficulty mentioned above, uncertainty whether the material which is aug•mented by the liver slices is really creatine. A number of substances give a positive test with the Jaffe reagent. The amount otthe material in question formed by the liver slices was too small tc/be identified by isolation. Accordingly further study of this problem was postponed until a . hi.ore specific analytical method which could be adapted to 'ubmicro scale was available. Such a method was provided by Dubos aJd Miller (3) . They discovered and succeeded in culturing a soil bacterium which specifically destroys creatine and creatinin;.~- We again took up the problem, employing tissue slices and this new adjuvant to our former submicromethod for creatine (4). We.havenfound that liver slices of the cat, abbit, and rat ar~ able convert glycocyamine to creatine. In the experimentai con~htions vie have observed, the increase is 5 to ·20 times the·amount originally .present in the slices. The difficultie.s and uncertainties which exist in conclusions resting on statisticai analysis offsmall <'11.fferences have therefore been overcome. The liver is ·well stiit~~- to this study, q,ecause its normal creatine content and th~refore the control or, blank value is very low. We have found further that ·when methionine is present in the Ringer's ~olution, the amount of creatine formed from the added glycocyamine was on the average about 50 per cent greater than when methionine was not added. We have tested thirty-two other substances including amino acids, methylated amines and a purine, and betaine. All of these were negative in this respect. It seems reasonable to conclude therefore that rat liver slices transfer the methyl group of methionine to glycocyamine, thus converting the latter substance· to creatine. to 3 Hr.,B:drs0ok,a;nd J. W. Dubnoff 561 Technique The tissue slice technique used is only slightly modified from that described by Warburg (5) . The details of the reaction vessels and the manner in which they are mounted are described in a previous communication (6). The blank controls and each experimental variant were carried through in triplicate. We have had twenty to thirty reaction vessels, eacli containing three slices from the same liver, running simultaneously. ,It was necessary to make the sampling of the slices as uniform as possible, because the capacity to synthesize creatine may vary significantly between the lobes. The procedure we finally adopted was to arrange · the reaction vessels in three 1:ows (when each experimental variant is carried through in triplicate), one vessel in each row for each variant. Consecutive slices were transferred_ to a Petri dish containing Ringer's solution, one slice for each vessel in the row. After 3 minutes soaking in this vessel, with gentle agitation, the slices were transferred to _the reaction vessels. The same was done for the other two rbws. The whole procedure was then repeated according to the nu{riber of slices wanted in each vessel. The reason for the 3 minutes soa~ing in the Ringer's solution before transfer to the reaction vessels is that more creatine was formed than when 'the slices were transferred directly: "!7ith orily momentary rinsing iri Ri~ger's solution. •• At the end of the reactioh period the contents of the vessels, in"cluding the slices, were t'ransferred to test-tt{bes, and the bottles and transfer pipette washed with two 1 ml. portions ·of 0.02 N HCl. The t est-tubes··wei·e placed in a boiling water b'a th for 10 ininutes witn ' occasional' stirring or shaking. After this time the 1e~coole8.!'iind the volumes made up to 6 ml. with test-tubes ~e1 • .f • ' . 0.02 N HCl;,· ' Tliese ""test-tubes are marked at 6 ml. Usually only 0.1 to 0.2 ml. was needed to bring the· volumes to the mark. The suspensions were then filtered. The coagulated slices with the coagulated protein in each vessel were transferred to · small glass dishes and the weight determined after drying at 100°: The analytical proct dure was ·essentially that described previously '(4). Some minor details were varied from time to time in addition to employment of the bacterial digestion procedure of Dubos and Miller. Table I is a protocol oi an experiment in 4 562 Creatine Formation in Liver • which the complete analytical procedure was used. Instead. of triplicate reaction vessels, six were used for each experimental variant. ,Tl\e boi\ed ·e:xitra9ts in pairs of vessels were joined and filtered. To 8 ml. of this filtrate were added 2 ml. of 0.25 M phosphate buffer at pH 7.0. This solution was divided into two portions. In one-half the creatine was determined directly. The creatine in the other half was destroyed by a suspension of NC soil° organisms prepared according to the prescription of Dubos and Miller. The difference between the color developed by the,.Jaffe reagent with and without this digestion gave the true creatine (and creatinine) in the original solution. The bacterial digestion was carried out as follo.ws: 5 ml. of the ,. ,J solution neutralized with phosphate were transfe);red to a 250 ·µil. Erlenmeyer flask. To this was added 1 ml. . {,a suspension Qf the NC soil organisms. The n,ecks of the flasks w.ere covered with squares of Parafilm and set away in an air ,b ath at 38° f.~r ½ to ¾ of ·an hour. At the end of this time l · ml. of 0.5. }'f HCl was added to each flask. The contents were then centrifuged. 6 mL of the clear supernatant, ;~lution were taken for analysis. The length of time the bacteria w'eJ:e' allowed to react on the experimenti:i,l solutions was based on a P11.~o~ determination of the potency of th~ ,pacterial suspension used.) rlt was the time required by 1 ml. of the bacterial suspension t<j>,J\estroy completely the creatine in 5 ml. of a 2 mg. per cent solution. This was a larger amount of creatine than·in any of the tissue-extracts submitted to digestion. To the 5 ml,. of tissue extr-a ct' ;ontaining the phosphate buffer, but :which had not been digested by the bacteria, were ,added 1 ml. of water and 1 ml. of 0.5 N HCl. 6 ml. were then taken for analysi~. From this point on the _analytical procedm:(i} ~p,s identical fop the solutions which had been digested with: P,a~tyria and those which had not. Both sets were carried through ~o tl\e completion of the analysis simultaneously. The. 6 ml. aliquots were transferred to thick walled Pyrex testtubes with internal dimensions of 125 X 12 mm. The tubes were cov~.red with parchment paper caps and .autoclaved for 20 minutes at, l2,5°. After they were cooled, a small amount · of Lloyd's reagent was added to each. We have found that the amount of the Lloyd's reagent may vary from 10 to 60 mg. withol},t affecting the final result. TJie test-tubes were now shaken for 7 minutes ~ 5 H. Borsook and J. W. D1abnoff 563 on a shaker of the type devised by Fisher aqd Wilhelmi (7). The tubes were then centrifuged, the supernatant solution discarded with the last drop at the rim of the test-tube ~aken up with filter paper, and the Lloyd's reagent resuspended irr 2 ml. of 0.01 N HCl. The tubes were centrifuged again, the supernatant solution again discarded, and thf adherent moisture on the walls of the t est-tube carefully taken up with filter paper. 3 ml. of a sodium picrate solution were added to each tube. This solution consists of 10 parts of saturated picric acid (purified) and 1 part of 10 per cent N aOH, these being mixed immediately before use. The test-tubes w@re again shaken for 7 minutes and centrifuged. The color was measured on a spectrophotometer with light of approximately 0.525 µ wave-length. With concentrations ranging from 0 to 2 mg. per ·cent there is a strictly linear relation between concentration of creatine and the intensity of color measured in this manner. ~ The following controls were taken through the above identical procedure inch1ding digestion by the bacteria: Ringer's solution alone, Ringer's solution· containing the ·same concentration of glycocyamine used in. t he ~xperiment, and Ringer's solution containing glycocyamine and methionine. Each of these was carried out in triplicate. In a(idition 1 ml. of the bacterial suspension alone and bacterial suspension plus 5 ml. of 2 mg. per cent creatine were incubated with the experimental solutions and carried through the same analytical procedure, performed in duplicate. Finally a set of five standard creatine solutions with concentrations ranging from 0.1 to 2 mg. per cent was treated in exactly the same manner as the expei:ime:r:itar solut ions except that they were not submitted to bacterial digestion·. The amounts of creatine in the experimental solutions and controls were determined by interpolation from the straight line given by the readings of these standard solutions. The above controls and the standards were carried through afresh in every experiment with the experimental solutions. It is a testimony to the convenience of this analytical method that we have frequently carried through more than 60 individual arralyses from the bacterial digestion to the final spectrophotometer reading in less than 5 hours. We found in experiments with rat and rabbit liver' that essen- " " Incubated 6 hrs. 6 " " 6 " " 6 " " 6 " " 6 " " 6'" " 6 " " 6 " " 6 " " 6. " " " " ... .,.. 1· :~~ ~ . •• < ..... ~~-.,·· ~ ~,;_ - ~ -· c: :,".;-. ~;'.· ~.. Analyzed immediately (1) Treatment (2) (3) a >, "g >, " ·s" "' "' (4) ~ "' (5) Q >, "' .!'i' :a., 0 ..0 'o ., ·a" "' " .!!l .," I -2~>, 4 . 4 4 4 4 4 ·4 4 , 4 4 4 4 4 4 m l. ~ 0.25 0.25 0.25 0.25 0 .25 0.25 0:25 0.25 }_ ~.;;· mg. ·~: ' 0 mg. ;.! ' ,. r ~ 40.8 41.5 '~ .: 59 .8 34 .1 40.2 38.0 35.4 45 .0 35.5 1.45 32.5 1.45 37.8 1.45 35.5 mg,_ - (8) < 0 >, "" Soii ~ - cent --,:'"I cent , 0.21 0.12 0.26 0.14 0.22 0.12 1.20 0.60 0.38 1.54 0.94 0.66 1.31 0.71 0.49 1.54 0.90 0.70 1.54 0.90 0.64 l.'53 . 0~'89', o,6Z ,. 0.61 ,.. ;-, ., it; 0.60 0.17 0.12 0.14 cent d J, (9) ~-£ o; i'l" ·..=""' , g 0 cent mg . per . "~ "'i, .s El 0 ..... a)ciS x-= • El _., El 0 0 .§ . " · >, i:-~ .G- :. ~g :... " '8 t..:;' g ~~ ~-- i'l" .·,:] :~ "Cl , "'" s.s .!:: .b'. os - - cent mg. per mg. per mg~r ·cent 0 ::l " ~ .,l5. "'i, "'~ "Cl True creatine 0.07 0.07 0.07-· 0.10 -:- 0.,07 0.14 • 0.07 0.03 0.18 0.08 0.03 0.08 0.03 0.14 1. 9 p.70 0.50 0.45 0.42 2.5 0.72 0.82 0.77 0.74 2.4 0.72 0.59 0.54 0.51 3.7 0.72 0.82 0.78 0.75 2.9 0.70 0.84 0.79 0.76 0.82 0.77 0.74 ~-2~- 0.71 0.61 .,; -t --- i: · 0.60 ' I ""' (0.15) (0.13) (0.15) 2.1 2.8 2.5 3.9 3.4 3.6 ' - -- ie'•.]""' -~~ "gj~-~ : Soii 13 , o; ii:: < :>i ~~~'§!: , < (10) ~ ' (14) (H J-, ~o:w (13) • , __- •-·· _. _ ]~ '1:1-~ C ~ ""' 2] ..o .... x-= ~8 ;::., ~rs El" ·aca a,·- Ei ~ -- (,) ;a . . El _g 0 --- - - - - - f--< (7) 0 .," 0" mg. per mg . per mg. per - f--< (6) i 3" 8 0 "" 8-- "'g ~ -~ 'fl':f .r - 8 _g ~-r ..~~ "Cl Apparent creatine "'El .8., - -- - - - - - - - - - I ~ C . I ?~" ::l :~ ·a;"' Composition of m~dium I TABLE I Formation of Creatine by Rat L iver Slices in 6 Hours _at 37.5° from Glycocyamine ~ ..... i:::l >-l (i) r ..... <1 i:::l 0 c+ ..... i:::l .. . ~ s >-l 0 l'zj (i) c+ ~ (i) 0 >-l ,.pa. C.,7 0) CJ) 4 4 4 4 0.25 0.25 0.25 0.25 1.45 1.45 1.45 The factor co~verting mg. per cent to Q(creatine) is 4/ 100 X 1.5 X 171 X 1/ 6 X 1/(dry weight of tissue): 4/ 100, since there were-only 4 ml. in each reaction vessel for the amount of tissuiiindicated; 1.5 beb:ause the; e 4 ml. were diluted to 6 ml. in the tranifer from the reaction vessels to the test-tubes in which they were b'oiled; 171, because 1 mg. of creatine would occupy 171 c.mm. at s.T.P. if i-t were a gas; 1/ 6, because the experiment was run for ·6 hours. Q(creatine) is therefore the amount- of creatine formed, expressed as if it -were a gas in c.mm. at s.T.P., per mg. of tissue (dry weight) per hour. Incubated 6 hrs. " 6 " 6 " " 6 " 1 ml. bacterial suspension incubated 45 min. at 37° 0.1 mg. creatine incubated with 1 ml. bacterial suspension 45 min. at 37° 01 ~ 01 ;:l:i 0 Ci' :::I ~ t1 ~ ~ [ ~ ~ O• ~ en. p i-.-l w 0 p::1 -4 8 566 Creatine Formation in Liver tially the same result was obtained whether bacteria were used or, not. Accordingly we later dispensed with the use of the bacteria, except as indicated. Results Table I is a condensed protocol of a typical experiment. The figures in Column 6 are the spectrophotometer readings converted to mg. per cent by interpolation from the standard curve. The figures in Column 7 are obtained by subtracting from those in Column 6 the values of the glycocyamine or glycocyamine plus the methionine· blank, and the amourit present in the tissue at zero time. The glycocyamine blank value is quite large. It arises from the conversion of glycocyamine to glycocyamidine during the autoclaving. It wa:3 essential that the glycocyamine blanks be treated exactly the same as the experimental solutions through all the operations frorri: immersion in the water bath at 37.5° for the same length of time to the fi:hal development of the color. The figures in Column 8 are obtained by subtracting fro:rri those in Column 7 the amount of chromogenic material formed by: the tissue in the Ringer's solution without glycocyamine. Q(creatine) (Columns 9 and 14) is the amount of creatine formed, expressed as if it were a gas in c.mm. at s.T.P., per mg. of tissue (dry weight) per hour. The figures in Column 11 are the differences between those in Columns .6 and 10. From the figures for the blanks in Columns 6 and 10, it is seen that the bacteria digested none of the glycocyamine. The figures in ,Col~mns 9 and 14 are not significantly different. They are a little higher in Column 14 than in Column 9 because ·in Column 7 a correction should }iave been applied to the glycocy'amine blank value for the glycocyamine converted to creatine, from 10 to 16 per cent. If this had been done, the figures in Column 8 would have been increased by 0.06, which would have made them the same, within experimental error, as those in Column 13. In a previous determination we have found that there Wl:1,8 practically no creatinine in the tissue extracts. Table I shows that the liver slices convert glycocyamine to cre,atine. This, as far as we know, is the first time the biological conversion of glycocyamine to creati1~e has peen demonstrated by an unequivocally specific analytical method for creatine, and in which 9 H. Borsook and J. W. Dubnoff 567 the tissue used both in the controls and in the experiment came from the identical organ specimen. The increases in the experimental vessels were from 10 to 15 times the amount present in the tissue at the beginning of the experiment. The figures in Columns 9 and 14 show that with added methionine there was 40 per cent more creatine formed than when methionine was not added. Most of the chromogenic material in the liver slices at the beginning and end of the experiment was not true creatine. '.This .is in accord with the findings of Baker and Miller (8). There is, however, a slight increase in true creatine in the liver slices suspended in Ringer's solution containing no glycocyamine. We have found this repeatedly. Most of this non-creatine chromogenic material found in the tissue blanks is also not glycocyamine. This was ascertained by a direct determination for glycocyamine. The values for Q(creatine) are much smaller than those found with liver slices for the formation of urea, amino ac_ids, or hippuric acid (6). Nevertheless, this rate, small though it is, is more than sufficient to account for the total creatine plus creatinine excretion in the rat. Thus, '. an adult rat wit~ a liver weighing 12 gm. might excrete 9 mg. of creatine plus c1'.eatinine in 24 hours (9). A Q(creatine) of 0.02 would correspond in such an animal to the formation of 7 mg. of creatine in 24 hours. The following compounds and combinations of compounds were tried instead of glycocyamine to determine whether they could serve as precursors of creatine : arginine, arginine plus glycine, arginirre plus glyc0lic acid, choline, glycine, glycine plus urea, glycolic acid, and methionine. Each of these has been tested on both rat and rabbit liver slices several times. The results have been consistently negative. Table II is a summary of most of our experiments with glyc~cy,. amine and methionine. In the course of these experiments we have used glycocyamine from two different commercial souilces, and two specimens of dl-metfiionine, one _prepared in this laborl:l,tory and one obtained com~~rcially. A different animal was used for each pair of figures. The data in Table II show the increase in creatine formation invariably obtained when methionine was added to the glycocyamine. Approximately the same relative increase occurred regardless of age, sex, and nutritional condition of the animal, and 10 • 568 Oreatine Formation in Liver II TABLE Formation of Creatine by Rat Liver from Glycocyamine with and without Added Metb,ionine at 37.5° From glyco- From glycocyamine cyamine and alone methionine Duration of incubation Age and sex Nutritional condition Normal nutrition. Q(creatine) Q(creatine) X 100 X 100 ltrs. 3.2 6.1 4.0 4.0 5.7 1. 9 4.2 6.6 6.5 4.6 8.5 6.1 7.3 7.7 3.1 2.3 1.5 1.5 4.1 4.4 2.7* 2.8* 2.5* 2.4 3.1 2.5 2.2 2.7 6.2 9.0 5.5 6.5 8.1 3.4 7.4 7.9 9.0 6.6 11.3 8.5 10.6 11.1 5.8 4 .5 2.3 · 2.4 6.4 6.9 1 Adult. o" 2 " o' " . o' ,31 J 3 • o' 3 o' 3 3 mos. 3 ·3 " o' 3 3 " 9 3 3 " 9 4 Adult. o" 4 ,.", o' 4 r~~~ o' 4.8* 3.6* 3.3 4 ~] U3~ '! -4 r.Hl4 4 4 4 ~ 6 6 6 6 6 6 6 6 r " 1 mo. 1 " o' o' o' o' '" 'fl o' ) 1 f:~ ( 9 9 ",.· " " " " " " " " " " " " " " " " " " " " " " " Fasted 66 hrs. " 66 " ~" 'Adult. o" ",i " " " " " " t Normal nutriti,0;1 " " " o' o' ·o" ' " " " " " o' " o' OS. o' It'. 9 •OJ;C 9 I r' " " " " " " ",~ " Each of the above figures is the averlj,g,~~of a triplicate determiµa,tion. Each reaction vessel contained 20 to 40 , , gif(dry weight) of liver, 4 ¥11. of Ringer's solution containing 0.25 mg. of glycocyamine, and in the methionine series in addition 1.49 mg. of the amino acid. ' * Values obtained with the employment of bacteria in the analysis. regardless also of the duration of the experiment. The over-all average value for Q(creatine) X 100 from glycocyamine algne is 4.1, and from glycocyamine plus methionine 6.6. ,a ,,..~ - • 11 H. Borsook and J . W. Dubnoff 569 We have no explanation as yet for the large variations in the rate of creatine formation in different experiments. It will be noted that most of the results in Table II wei:e obtained without the employment of bacteria in the analysis. Three e~periments with rat liver and three with rabbit liver were carried 0ut in which bacteria were used. The values for Q(creatine) were, as in Table I, essentially the same with and without the use of bacteria. We feel therefore that the results obtained without the use of bacteria with this tissue and under these experimental TABLE III Rate f,or Formation of (Jreatine from Glycocyamine with and without Added Methionine TiJie ,.:~ Without methionine With methionine Amount of creatine formed per Q(creatine) X 100 100 mg. (dry weight) of tissue ;f,J Amount of creatine formed per Q(creatine) ,x itoo 100 mg. (dry weight) of tissue hrs. mg. 1 2 4 6 0.018 0.070 0.095 0.144 mg. 2.8 6.0 4.6 4.5 I 0.036\ 1 0.10;,, ,.(f 0 .150 1;. 0.203 Each of these figures is the average of a triplicate determination. T~~ composition of the Ringer's solution and amount.,of tissue were the same as in the experime;ts of Tables I and II. Two 'lobes of "liv;er were used'. Consecutive slicJs' wi:re placed alternately in the vessels with and witho~t methionine. The comparison at each hour, therefore, is of the activity of immediately adjacent1 sections of liver. Bacteria were not employed in the analyses here. ' _,, , ,. conditions are as reliable indices of true creatine as those obtained with .bacteria. Table III contains the results in more detail of an experiment in which l the rate of creatine formation from glycocyamine, with and without methionine, was studied. The figures show -that the methylation of glycocyamine proceeds unchecked for at least 6 hours. The· maximum at 2 ho~rs in each series is accidental. It did not occur in other similar experiments._ The absolute amount of glycocy!tmine methylated was not increased by a 1O-fold increase in the initial concentration of gly- 12 570 Creatine Formation in Liver cocyamine. We have not yet explored thoroughly the effect of changing the concentration of methionine. In one experiment 0.0025 M methionine was as effectiv:et. as 0.01 M. The fact that addition of so much methionine inoreases the rate of methylation only 40 or 50 per cent leads us to suspect that the methylating agent in the liver itself may not be methionine. Another piece of evidence which points in this direction is that the ratio of the rates of creatine formation with and without methionine is nearly the same throughout the whole period of from 1 to 6 hours (Table III) . In other words, the effects of the methionine and of the methylating agent in the tissues were additive. It is possible that the methylating agent in the tissues is derived from methionine. The following compounds were tested with rat liver as possible methylating agents of glycocyamine. All gave negative results: acetylcholine, d-alanine, d-arginine, Z-asparagine, Z-aspartic acid, betaine, caffeine, choline, Z-cysteine, Z-cystine, ethanolamine, d-glutamic acid, d-glutamine, glycine, glycolic acid, guanidine, Z-histidine, dZ-isoleucine, Z-leucine, d-lysine, mono-, di-, tri-, and tetramethylamine, d-ornithine, Z-hydroxproline, dl-phenylalanine, Z-proline, dl-serine, d-threonine, Z-tryptophane, and Z-tyrosine. The final concentration of glycocyamine in the Ringer's solution in these experiments was always approximately 0.0005 M, and 0.0025 M . of the compound whose methylating possibilities were being tested. The significance of the positive effect invariably obtained witl:i methionine is heightened obviously by the fact that all of the above compounds were negative. Some experiments have been made with the kidneys of the cat, rabbit, and rat. Slices of the cortex with and without methionine either failed to methylate glycocyamine or the slight positive results ·were within the experimental error. These experiments are part of a survey not yet completed of the organs of a number of animals. The details of these experiments will be published later. Minced liver of the rat or rabbit failed to give any measurable increase in creatine in 6 hours at 37.5° on incubation with glycocyamine, with or without methionine. Similarly negative results ·were obtained· with slices of heart 13 H. Bo.rs0;ci>ktandrJ AW.. Dubnoff 571 and sartorius. B.utathe~pell ' tircmu,r is mot ,preserved in slices of these muscular struotures'as ifls·in.,slices of liver. Until conditions are discovered in which positive results are obtained with rriinced liver, we feel 'that no significance can be attached to negative results with sections whose cell structure has been broken or to extracts of other organs. DISCUSSION Beard, Boggess, and Pizzolato (10) proposed that glycine and urea condense to form glycocyamine, which is then methylated by· more glycine or glutamic acid. We have observed neither this condensation nor the proposed methylating reaction ;\Yith rat or rabbit liver slices. The conclusions of Beard et al- ta-re based largely on experiments on the rat. The negative •1'8sults in our experiments with glycine and urea w.e believe are>isi~nificant in view of the positive results with glycoeyamine aQd.i methionine. Experiments on the synthesis of urea, amino acids, and hippuric acid have· shown that results obtained witlt tissue slices afford reliable qualitative information, at least, regarding the p©tentialities of the intact tissue in situ. Bodansky (11) fed glycocyamine to normal rats and at 3, 6, 12, and 24 hours after feeding measured the glycocyamine and creatine concentrations in the liver, muscle, heart, and kidney. Bodansky's interpretation of the data obtained in these experiments was that significant increases in creatine content occurred only in the kidney, and that the increases in the liver were not sufficiently clear cut to be significant. Bodansky concluded that his findings suggest "that methylation of the guanidoacetic acid may have occurred in the kidneys ... In view of the occurrence of guanidoacetic acid in large amounts in the liver and the failure to show an increase in creatine, it is surmised that the liver plays an insignificant role, if any, in creatine production." This surmise is in direct conflict with our observations. Bodansky's experimental observations, however, and ours are not in conflict. ffhus the creatine coniienbo'f·, the liver in two controls· in Bodansky's experiments was ]fu{hahd 20.4 mg. per 100 gm. of tissue, and-in three experimentsooth,,glycine, 18.6, 16.8, and 19.0 mg. After glycocyamine feedin"g1tp.-e1figures are 21.7 mg. in 3 hours • 14 • 572 Creatine Formation in Liver and 23.3 in 6 hours, and in a second experiment 30.5 in 3 hours and 22.3 in 6 hours. The relative increases over the control values in the liver with glyc6cyamine were of t he same order of magnitude as those found in kidney. The increases in true creatine in the liver were probably relatively much larger, since most of the chromogenic material in the liver with the Jaffe reagent is not creatine (Table I , and also Baker and Miller). There is another physiological factor which must be taken into account in interpreting the data obtained by Bodansky. 'T is factor is that the kidney is better able to store creatine than the liver. The analyses of Baker and Miller and our own show that the true creatine content of the kidney is 4 or 5 times that of the liver. Bodansky .found that when creatine was fed the creatine content of the liver was twice the control value at the 3rd hour but· had declined to the control value by the 6th hour, whereas in tlie kidney the concentration was 70 per cent above the control value at the end of the 6th hour. Our observations show conclusivelJ that glycocyamine can be methylated by rat liver. The rate of methylation by kidney slices is much slower than in liver, if it is not absolutely negative. All these observations ,are brought into accord by the hypothesis that in the experiments of Bodansky the creatine synthesized from glycocyamine in liver was quickly removed by the blood and stored for a relatively long period in the kidney. We have no reliable data .of -our own at present on the possible conversion of glycocyamine to creatine in other organs or in the muscles. . Fisher and Wilhelmi found that when isolated male rabbit heart was perfused there was an increase in creatine when arginine was added to the perfusate. No increase in creatine was observed under these conditions in the hearts of prepubertal _animals. Davenport, Fisher, and Wilhelmi (12), extending these observations, found that glycolic acid was essential for the methylation of glycocyamine. They suggested the following mechanism of creatine formation in the rabbit heart. Arginine is broken down to glycocyamine and glycolic acid; the glycolic acid then methylates the glycocyamine to form creatine. In rat and rabbit liver slices the results with arginine, with and without glycolic acid or glycine, and with the two acids alone did 15 H ' iBorsook and J. W. Dubnoff 573 not yield detectable amounts of either glycocyamine or creatine. The only substance we have yet found which is capable of methyl~ting glycocyamine is methionine. We have attempted to repeat the observations of Fisher et al. with slices of rabbit heart. These experiments were unsatisfactory because of the difficulties of obtaining uniform sampling and because the experimental effects were small compared with the amount of creatine initially present. For the reason stated above we do not attach any significance to these experiments. The difference between the observations on the perfused heart and ours on the liver stand, for the time being at least, either as an unresolved discrepancy or as indicating important differences in the mechanism of creatine formation in the heart and in the liver. SUMMARY 1. Liver slices of cat, rabbit, and rat convert glycocyamine to creatine. 2. This methylation is accelerated in rat liver by methionine, (other animals are now being studied). 3. Methionine is the only substance we h_ave yet found among a large number of amino acids, methylated amines, and other compounds which is able to effect this methylation in rat liver. The authors wish to thank Mr. Y. Tajima for the assistance he gave them throughout this work, and Dr. H . W. Davenport for advice and assistance in the construction of the shalfor used here. They are indebted to Dr. R. Dubos, and wish to thank him for specimens of the bacteria used in these experiments and valuable information on the culture details. BIBLIOGRAPHY 1. Fisher, R. B., and Wilhelmi, A. E ., Biochem. J., 31, 1136 (1937). 2. Borsook, H., and Jeffreys, C. E. P ., J . Biol. Chem., 110, 495 (1935). 3. Miller, B. F., and Dubos, R., Proc. Soc. Exp. Biol. and Med ., 36, 335 (1936). Dubos, R., and Miller, B. F., J . Biol. Chem., 121,429 (1937). 4. Borsook, H., J . Biol. Chem., 110, 481 (1935). 5. Warburg, 0., The metabolism of tumours, translated by Dickens, F., London (1930). -l 16 574 Creatii;ie.@orm.,ati014_j..n0Li'¼..er 6. Borsook, H. , and Dubp.of{, J. W., J. Biol., Chem.,, 132, 307 (1940). 7. Fisher, R. B., and Wilhelmi, A. E., Biochem. J ., 32, 609 (1938). 8. Baker, Z., and Miller, B. F., J . Biol. Chem ., 130, 393 (1939). 9. Donaldson, H. H., The rat, Memoirs of The Wistar Institute of Anatomy and Biology, Philadelphia, 2nd edition (1924). 10. Beard, H. H., Boggess, T. S., and Pizzolato, P., Proc. Am. Soc. Biol. Chem., J. Biol. Chem., 119, p. ix (1937). 11. Bodansky, M., J. Biol. Chem ., 116, 641 (1936) . 12. Davenport, H. W., Fisher, R. B., and Wilhelmi, A. E., Biochem. J., 32, 262 (1938). 17 Reprinted from Tm,: JOURNAL O>' BIOLOOICAL CHEMISTHY Vol. 134, No. 2, July, 1940 CREATINE FORMATION IN LIVER AND IN KIDNEY* BY HENRY BORSOOK AND JACOB w. DUBNOFF (Frorn the Wi lliam G. Kerckhoff Laboratories of the Biological Sciences, California Institute of Technology, Pasadena) (Received for publication, April 12, 1940) J We reported recently (1) the formation of creatine from glycocyamine by rat liver slices; and that 40 to 50 per cent more creatine was formed when methionine was added with the glycocyamine to the Ringer's solution in which the slices were immersed. Among some thirty .odd amino acids, methylated ·amines, a methylated puriIJ.e,t and betaine only methionine gave this increased rate of. ,methylation. The rate of creatine formation under these conditions is sufficient, if it is of the same order of magnitude in vivo, to make good the entire loss as urinary creatinine. If the methylation of glycocyamine in the liver is important both as a mechanism and as a site of creatine formatiq,n in animals, one may expect to find it in the livers of most other animals. W'e accordingly examined the livers of a number of animals for their ability to carry out this reaction. At the same time the possibility of its occurring in their kidneys was investigated also. The findings in this suvvey; summarized in Table I, were that g1ycocyamine can be methylated by the liver of every animal examined; the stimulating 'effect of methionine observed with rat liver was found with some but not all the other animals; the kidney appears to be unimportant (except in the pigeon) for the methylation of glycocyamine without or with methionine . . The experimental and analytical procedure was the same as was described in our previous communication, except that the experi}!!,ents with frog live1; and kidQey were carried out at 25° instead " ~ 1 • Presented before the meeting of the American Society of Biological Chemists a t New Orleans, Mal·ch 13-16, 1940 (Proc. Arn. Soc. Biol. Che'm., J. Biol. Chem., 133, p. xv (1940)) . 635 18 636 Creatine Formation of 38°. All the creatine figures given represent differences in chromogenic material in the Jaffe reagent before and after digestion with tlie NC bacteria of Miller and Dubos (4). Baker and Miller (3) have reported that creatine is formed from unidentified p·re<mrsors in rat liver ahd kidney (slices 'and mash) . Our .fi,0.dings on this point are that small amounts of creatine (beyihd the experi~ental error) were for~ed from pre) ! J . ~ TABLE I Formati,on~of Creatine in Liver and Kidney Slices Temperature 38°; glucose-Ringer's solut ion. • The creatine is given as m·g. per 100 gm. of wet weight of tissue. Creatine in liver. Creatine in kidney After 4 hrs, incubation After 4 hrs. incubation '- r , -• With, ,With 6.25 With With 6.25 In. In 6.25 mg. pe~ • In6.25 mg. per ibally . , mg. ent._glrco- i.t ially . In , mg. cent glycor nt Ringer s per cyam1ne Present Ringer s per cyamine P ese solution cent and ~Q,n;ig. , solution cent and 40 mg. alone glyco- per ce·nt •alone glyco- per cent CJ;'am- dl;-m_ei~i' • c,1:am- dl-m~thi1ne on1ne 1ne omne -- -- . --·-- - - - - - - - - - 2.5 . 2.4 Cat. 12.0 12 .2 21.1 2.3 26 Dog . 2.9 2.9 3.5 26 26 Guinea pig 1. 7 6.6 3.4 12.4 4.5 8.5 Frog (25°) .. .. . 2.2 2.2 4.8 47 4<! 45 Pigeonf. ....... . . 1.0 2.0 l °I. 8 10 .7 17.9 22 .3 : Rabbit. 5.8 5.7 26.4 15 . 2; ~5 .0 i.5.4 Rat ... .. . 1.4 2.8 20 .6 r 28 .,4 1,8.,6 .l{i•.3 18.2 ' ~The i.nitial values are lower .than those g.iv.'_enl b-.y Ba-'keiilkrfa Miller (2, 3) for rat liver and kidney. In our pro,cedmz.~•t hgsliQe.s)~ W!!-Sli:~ ii). a large volume ,of Ringer' s solution for at leas.~} ~i.ll t~J>t1Wrp gi.p:¥; ave trans~ ferred to t~neaction vessels or subipittea,,\9 .analy~is ;.f.J~~p~ed creatil}e undoubtedly diffuses out bf the slices during t his wa'~ing'. -• , ! .. , \ , " _.(l:j:1,~,(.} .. ,; - ·existing ~r~3ursors in the liver slices of'.the guinea rH~; pikeort, irr&. rat. Kidney slices gave significiint incr'e'aset'in ev.ery ca~e ex.cilp!t in the pigeon. • • t· • ,t •• ' ., . ·t • • ~.,1~ ,} After the celtstructure of rat liver and kitl'ney was destroyed, no c'reatine was· formed from preexisting precurs'ots nor from glycocyam~~-without or with methio,nine. , The hoi:n.0genizing p-rocedure of Potter and E lvehjem (5) was used to disrupt. the cells. It would seem that the creatine formed by chopped rat live1:,and '. f • • ,{ ,.. > 19 H. Borsook and J. W. Dubnoff kidney fro~ preexisting precursors (Baker and Miller Om}must1 be ascribed to the intact cells in the mash rather than to a ltl»~mi:t~d enzyme. •. _:, • •. It is difficult to account for the large (absolute) am0unts of creatine formed from preexisting precursors in the kidney. slices in every case exc.ept in the pigeon, in view of the inability,1@'fi' cat, dog, rabbit, and rat kidney slices to methylate added\ giyc·o·- , cyarmi-ne;< even in the 'presence of added methionine. 'DB:e~0ne •• animal which was negative in this respect- the pigeon- wa the only animal whose kidney gave a positive effect with glyc'0c&a.: mine and a further increase with methionine. t ~-- ~ When the Ringer's solution contained glycocyamine, itbere .were large 'increases i.n creatine with the liver slices of eveFy ;afiima.1 exicept the dog. In. the interpretation of this negati,ve effect-!in tl'l.e case Qf tlie dog''the posi~tive effect observed when methionir.re ,Jas acl'cled :with h'e, glycdcyamine must be taken into account.:· One possible· ,inifel'pvetation is that the preexisting concentratibJi ... of the glycocyartii.irie-methylating agent is very low in dog;liveF-4+.. lower than in theFfi\rer-s 0f any of-the other animals investigmt~. With kidney ·sl.ie€s. the added glycocyamine was not methylat~d or "Vas1methylmte'd, only. to· an insignificant extent in every ctise except in the 1guin€a· pig and pigeon. " 11•11, • B·otlaniky (!<ii) found ,an inenJ~se in the creatine conte~t of rat kicln~~ftet gly,eGcyamine was; •fied. Our interpretation of this . increase was t hat it a1·ose from ,temporary storage in the kidney of cr:e~tihe synthesized in the liver from the glycocyamine fed (17'. r;r .his i,n.terpr~tation is base1:l on the finding that rat kidney slie@s l;lo~nob~metliylate, glycocyamine under conditions in which m~tnylatio:ti.'Cteadily occurs in liver. Bodansky concurs in this interpl(€t$ti0.11 '(personal communication)·. •Methiomine accelerated the m~thylation of glycocyamine by liver slice~ of the dog, frog, pigeon, and rat. This stimulation was not observed with the cat, guinea pig, and rabbit. Two possible r~asons for·~th,is failure are that the livers of these animals contained a plethora of. the methylating agent, or that the relatively high concentiiation of the unnatural isomer in the dl-methionine used was inhibitory (as in the rabbit liver slicesr The pigeon . was ~he only animal whose kidney slices were stimulated ol'methionine in their methylation of glycocyamii1e. 20 638 Creatine Formation' The pigeon is exceptional as far as creatine formation is concerned in that its kidney appears to share this function with its liver and resembles in this respect the livers rather than the kidneys of the other animals. The negative effect of methionine with the kidney slices of an the animals tried except the pigeon is a little unexpected. Oxidative deaminat.ion is more active in kidney than in liver slices (7); and after oxiaative deamination the !ability of the S-methyl group is· increased (8). It might have been expected therefore that the methylation of glycocyamine would be more activ~. m kidney than in liver. This is not the case. L Our finding in this survey, that methylation of glycocyamine is·ran<active function of the livers of all the animals examined; strengthens ·the conclusion that creatine is formed normally in the liver by the methylation of glycocyamine. It would be surprisiRg 'if thi_s is a fortuitous coincidence of little or' no physidlogical significance. Other· mechanisms and other sites of creatiDe formation are, of course;. not excluded by these findings.. . The significance of methionine here is less clear. Our findings suggest that it'isflthe methylating agenti of gly.obcy11mine in liver or,:its 'precursor: We are hesitant at present to;,acceP,t methionine itself as the methylating agent because its effect is small consider-, ing the concentration used. It may be that the:actual methylating agent is the product of the oxidative deamin_,atfon qf'illlethio,nineJ' This and other possibilities are now under investigation. ,,-rr Du Vigneaud and coworkers (9) .have sho that the presence of choline in the diet enables the animal to utilize homocystine or homocysteine i.n Jl)lace of methionine. They have suggested that choline enables. the body to methylate ·,the sulfhydryl group of homocysteine by furnishing .t he :required methyl group, and further that methionine may furnish the methyl groups of choline. The reality of the· latter relationship is now established by direct proof recently presented by du Vigneaud.1 Our observations suggest that the S-methyl group of methionine may participate also in the formation of the N-methyl •group of creatine, In this connection and in view of @ur findings with dog liver slices it is interesting that Stekol and Schmidt (107 found 1 Verbal communication at the meeting of the American Society of Biological Chemists at New Orleans, March 13-16, 1940. • 21 H. Borsook and J . W. Dubnoff 639 in the dog an increase in urinary creatinine after dl-methionine was fed, and none after glycine, glutamic acid, or Z-pyrrolidonecarboxylic acid. SUMMARY 1. A survey was made of the possibility of creatine formation in vitro by slices of the livers and kidneys of the cat, dog, guinea pig, frog, pigeon, rabbit, and rat. 2. Evidence of the existence of small amounts of creatine precursors in the livers of the guinea pig, pigeon, and rat, and in the kidneys of all of the above animals except the pigeon was found. 3. This finding appears to be much less significant quantitatively than the much more active glycocyamine-methylating .function of the liver slices of all animals investigated. Only guinea pig and pigeon kidney slices possessed this ability. 4. The above findings are interpreted as indicating that creatine normally is formed by the methyl ation of glycocyamine in the liver. • BIBLIOGRAPHY 1. Borsook, H ., and Dubnoff, J. W., J . Biol. Chem., 132 , 559 (1940). 2. Baker, Z., and Miller, B. F., J . Biol. Chem ., ·130, 393 (1939). 3. Baker, Z., and Miller, B. F., J. Biol. Chem., 132, 233 (1940). 4. Miller, B. F., and Dubos, R., Proc. Soc. Exp._Biol. and Med., 36, 335 (1936) . Dubos, R., and Miller, B. F., J. Biol. Chem., 121, 429 (1937). 5. P otter, V. R., and Elvehjem, C. A. , J. Biol. Chem., 114, 495 (1936) . 6. Bodansky, M., J . Biol. Chem., 116, 641 (1936). 7. Krebs, H . A., Z . physiol. Chem., 217 , 191 (1933). 8. Waelsch, H ., and Borek, E., J. Am. Chem. Soc., 61, 2252 (1939). 9. du Vigneaud, V. , Chandler, J.P ., Moyer, A. W., and Keppel , D. M., J. Biol. Chem., 131, 57 (1939) . 10. Stekol, J. A., and Schmidt, C. L. A., Univ . California Pub . Physiol., 8, 31 (1933) . 22 Reprinted from THE JounN AL OF BIOLOGICAL CHEMISTRY Vol. 138, No. 1, March, 1941 r.. r ,. A MICROMETHOD FOR THE DETERMINATION OF ~ GLYCOCYAMINE IN BIOLOGICAL. FLUIDS 'AND TISSUE EXTRACTS t I BY JACOB W. DUBNOFF AND HENRY BORSOOK I- (From the Willi_am G. Kerckhoff Laboratories of th~ Biological Sciences, California Instit,ute of T~chnology, Pasadena) • 0 (Received for pub1ication, September 18, 1940) ,, In .the two following communications (1, 2) evidence is presented that glycocyamine is a normal precursor of creatine in the animal body. These studies required a satisfactory micromethod for the determination of glycocyamine. The most reliable method described in thy literature consists in adsorp~ion on Llqyd's reagent in acid solution, elution with baryta, removal ofJ arginine from' "" ~t •J_ .. 1,J the eluate byJ.\. {F P,eated adsorption on permutit, and.,. olor-,i.metric 1 ~ tw 1 1.;.t,.• qeterminatio40 ~ l,the remaining glycocy1tmine bYi, JRflans of _,the Sakag~p)lj_ reaction. There are only two substan,ces lf,hich rare cpm op in- biological fluids and which gi v~ _an in te~e colo; ipJp.e Sakagp.ch~;- :~p.ction. These are arginine and glycocyamine. 1Jihis method , _was ~first introduced by Weber (3) and was( _modifier i;-y .t."" .ITVJv ,Jwi:/,,. BodaUSfY ~) and by J?avenport and Fisher (5). ) ..ar•· In O.tJ.&iJmRds 1ve1\:the latest version of the method, t~rhlescribe~9.'.i. J)av,eR-gQ t-and Fisher, had the following shortcqmiflgs: it was laborious and time-consuming, the adsorption of the ,glycocyamine on th!t Mw~d's reagent was incomplete, .fur,ther)&~~s of glycocY,amine cwq~J, ·ed in the repeated treatment with Prfofl.Utit (Dav; 1t-I?ort and.,,flsher report losing only 10 per cep.} l9 rt~ee adsorJ?Ripns; wit~ ~he permutit available to us we lost sp ;ijvfo~_Mt)' and . PA color developed was unstable. Furthermore, tJ.lJ,fHD,,Pe1t,J?.t of g~ cocyamine lost qn, the permutit varied accormPihtf.l:') Jte amount of arginine present, the less arginine the greater the lps <;if g!y.cocyamine. . . ,, _; AIL these disadvantages have been removed in the method .~u .P.l:lG~ scribed · below. It is t~e first method in which glycocY;~HW. 1 t. 381 23 382 Glycocyamine Determination, ~ •0 added to blood or urine can be determined quantitatively, even in concentrations as low as 0.1 mg. per cent. 2 to 5 ml. are sufficient for an analysis. An indication of the speed and convenience of the method is that twenty to forty analyses can be carried through simultaneously in about 2 hours. • Hl ReagentsC. > 0.02 M sodium-potassium phosphate solution at pH 7.0 (S¢rensen buffer); or 0.3 per cent sodium chloride solution. Permutit. ' A stoc~'solution of 0.2 per cent a-naphthol in absolute alcoliaI, diiuted with water to 0.0'4 per cent before use. 40 per cent solution of urea in water. .,, , , .• Hypobromite solution made as follows: 0.66 ml. of liquid bromine (specific gravity· :to) in 100 ml. of 5 Jper cent sodiuhi hy~roxid~. 1 Procedure Urine is diluted 5 to 10 times, according to its concentration.'~ Bl'ood may be deproteinized eithe1: by adjusting the pH to 6.0,' 1 5:..fBld dilution with water, boiling and filtration, or by precipitation ofithe proteins with sodium tungstate and sulfuric acid according ' to the pi:ocedm:e of Folin and Wu, with the mod5fication that fo/ each ml. of blood, 2 ml. of water and 1 ml. each of the 10 per cent sodium tungs tate and 0.66 N sulfuric acid al'e used. Tissue ' extracts are diluted with water until the final ·solution contains 1 gm. of tissue (fresh weight) in 40 ml.' of suspension. The pH is adiusted to 6.0, and the suspension idimersed in a boiling water bath for 10 minutes, cooled, and filtered. The analysis is carried oue von the deproteinized filtrate. Any argini,n~ which may be present is removed by sending the solution through a column of permutit contairted in tlie stem of a funnel made of two pieces of glass ~ubing. The dimensions are ai, follows: upper piece 100 mm. long, 15 mm. external diameter·; I' wer pieM.'rlO0 mm. long, 7 mm. external diameter (5 mm. inMi·nal dian\eter). The bottom end of the funnel is constricted s1igntly. .., c1 The dimensions of the lower naiTow tubing should be as, close as possible to those given above. If the tubing is nari:ower, the filtration will be slow; if it is much wider the reduced Might of the C - "· • 24 J. W. Dubnoff and H. Borsook 383 column of permutit may allow some arginine to .escape adsorption. The lower end of the funnel is stoppered loosely with a small piece of absorbent cotton just sufficient to hold back the permutit. 0.9 gm. of permutit is poured in and tapped gently tc{ remove entrapped air. It should form a column about 85 mm., tall. A testtube graduated at 10 ml. is placed under the end of the funnel. 5 ml. of the solution to be analyzed, obtained as described above, are pipetted into the funnel. After all of this solution has passed below the top of the permutit column, 5 ml. of either the 0.02 M phosphate buffer or of the 0.3 per cent sodium chloride solution are pipetted in above the permutit. Each 5 ml. portion of solution takes about 15 minutes to pass through the column. When the second 5 ml. portion has passed through, the filtrate in the graduated test-tube below is made up to the 10 ml. mark with water and then shaken. Washing the p~rmutit column with the phosphate or sodium chloride · solution removes the small amount of glycocyamine which remains in the permutit column after the 5 ml. of the solution under analysis have passed through. The dilute phosphate buffer and the sodium chloride solution are equally effective. We prefer the sodium chloride solution, because it is simpler to make up and also because a faint turbidity is sometimes obtained in the final filtrate when the phosphate solution is used. Different batches of permutit may differ in their affinity for arginine and glycocyamine. We have mentioned above that Davenport and Fisher lost only 10 per cent of the glycocyamine present in three successive treatments with fresh permutit, whereas we lost 80 per cent. It may be advisable therefore before 'a fresh batch of permutit is used in the procedure described to _ascertain the height of the column and amount of permutit necessary to remove all the arginine without loss of glycocyamine. 'J:lhe prescription we have given allows some latitude and it may cover all the variations in permutit now sold in this country. The permutit can be regenerated after it has been used by the following procedure. It is washed several times with tap water, thrown on a Buchner funnel, an equal volume of 25 per cent sodium chloride solution passed through, and then washed with distilled water until the wash water is chloride-free. The development of the color by means of the Sakaguchi reaction 25 Glycocyamine Determination. 384 and . the colorimetry may be carried out with any convenient aliquot so long as the proportions of the reagents are kept the same. The quantities of reagents given below are for 2 ml. aliquots which we have been using. Twenty to forty analyses can be carried through simultaneously. Test-tubes containing the aliquots of the solutions to be analyzed, inserted in a rack, are ' .1 I Determination of Glycocyamine in Aqueous Solutions Contain,ing Different • , Am,ounts of Interfering Substances All values are concentrations measured in mg. per cent. The concentration!', of the added substances were those in the final mixtures before they were sent through the permutit columns . The final filtrates contained only half the concentrations of glycocyamine shown. Accordingly th~ standards were diluted 1: 1 with 0.02 M phosphate buffer solution . TABLE l Arginine added Creatine added Glycocyamine added Gly cocyamine found by analysis 3.0 3.0 3.0 3.0 1.0 1.0 1.0 1.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 3.02 3.01 3.01. 3.00 1.01 1.02 1:00 1.00 0.10 0.10 0.10 0.10 0.10 0.10 0.09 0 .09 J' 35 35 ' 35 35 35 35 35 35 35 35 35 35 70 35 18.5 0 10 5 1 0 10 5 1 0 10 5 1 l 0 25 25 25) v 2.5~ 1o f.t.n! ., ' placed in an-ice water bath. •1n this bath also are the a-naphthol 1 (0.04 per cent), urea,. and hypobromite solutions. After 5 -t d io minutes 0.4 ml. of:the a-naphthol s0lution is added to each tube, ' the contents shakent 0.4 ml. ·of the chilled urea solution added, and the contents shaken again. After another 5 minute interval 0.2 ml. of the chilled hypobromite solution is added, the contents immediately mixed by vigorous shaking, and the tube replaced in the ice bath. 26 J. W. Dubnoff and H . Borsook 385 The color develops slowly, attaining its maximum intensity in about 20 minutes, after which (so long as the solution is in the ice bath) it remains constant for at least 1 hour. Before the color is measured, the solution is warmed to room temperature by shaking in water for ½ minute and the dissolved gas removed by vigorous tawing for 20 seconds. If the colored solution is left at room temperature for½ hour or more, the color slowly fades. ,.TABLE II Determination of Glycocyamine in Ringer's Solution and in Rat Hearl and Kidney Extracts to Which Different Quantities of Arginine and Glycocyatnine Were Added All values are concentrations measured in mg. per cent. The concentrations of adued glycocyamine and the experimental values gi-<ren were those in the finaF filtrates; i.e ., after they had passed through the permutit and were diluted with an equal volume of washing solution. The concentrations of added arginine were those before the solution s had passed through the permutit. Glycocyamine Arginine added Glycocyamine added Ringer's solution Found --0 35 35 0 35 , 35 35 0 0 1.50 0.75 0.75 0.375 0.075 0 , 1.53 '0.765 0.80 0.41 0.08 After subtracting blank value 0 1.53 0.765 0.80 0.41 0.08 Kidney extract Found --0.04 0.04 1.59 0.79 0.82 0.44 0.11 Heart extract After subtracting blank value Found After subtractin g blank valu e 0 0 1.55 0.75 0.78 0.40 0.07 0.08 0.08 1.56 0.85 0.84 0.45 0.15 0 0 1.48 0.77 0.76 0.37 0.07 With each group of analyses the color is developed simultaneously in four standard solutions containing 0, 0.25, 0.5, and 1.0 mg. per cent of glycocyamine. To measure the color we have used a spectrophotometer or a colorimeter. With either instrument the best light is that at approximately 0.525 µ. There is a linear relationship between the concentration of glycocyamine up to 1 mg. per cent and the intensity of the color when it is measured with light near to 0.525 µ. Tables I to IV indicate the sensitivity and reliability of the. method described above. Table I gives some typical results 27 386 Glycocyamine Determination with prepared mixtures, in aqueous solution, containing varying amounts · of glycocyamine and of two interfering • substances, arginine and creatine. Table II summarizes some tests of the method with Ringer's solution and two tissue extracts. These tests were necessary, even after the results in Table I were obtained, because the equilibrium relation between the glycocyamine adsorbed on permutit and that remaining in solution is affected by the concentration of T ABLE III Determination of Added Glycocyamine and of Added Arginine in Human Urine (Dilut ed 5-Fold with Water) All values are conce,ntrations measured in mg. per cent. , 11 ,. 1:he concentrations of a4ded glycocyamine and the expe~imen,~al, values foun.c;l were th_ose in the f\Jtrntes; i .e. , after they had passed through the per!Ilutit and were dilute~ with aIJ. equal volume of was hing solut~on. The concentrations of added arginine were those before the s_olutions had passed through the permutit. ,., Glycocyamine Glycocyamine .. added Arginine added 9 22 o, ' .. , ,. I .') -0 0 0 0 30 22 10 0 F ound ; •AfVer•subt~actin~ 'amount originally present 0.21 0 .21 1.26 0.72 0.71 . 1 0.70 0.69 0.46 -·-- 0 00 01.05 ~8 ,. I 0.51 i1S .Co.so 1 .00.49 0.48 0.25 salt. The reliability of ip.ost of the experimental results described in the next communication (1) depei;ids on the accuracy of this analytical metl;iod. The data in T~ble II show that the agreement bet~eei\ t_heoretical a11:d experiIJ:l;ental values is satisfactory. The det~r~ m;1tion of gly'cncyamine' in ufine is one of the most severe teiWlt'o' which an arral~ti'oal method for gly:cocyamine can be subj~'hJ{f Udne contdint'l·fllatively~ high concentrations of interferirig..,substances such ~ s ~urea, a&monia, creatinine, and amino acids including arginjne. The rigQr of the test ·was in':l ... ~ iR~f • 28 J. W. Dubnofflm~ H. Borsook 387 creased by adding to the ·uriri.e varying amounts of arginine and glycocyamine. , Tal;>le IIj: shows that the . same •satisfactory agreement is obtainl:ld between theoretical . and experimental values as in simple aqueous solutions. This is also the case with blood (Table IV). TABLE IV Determination of Glycocyam.ine Added to H1iman Blood All values are given in mg. per cent. Varying amounts of a 10 mg. per cent glycocyamine solution and of a 20 mg. per cent arginine solution were added to 4 ml. aliquots of whole human blood. To these mixtures water was added so that the final dilution of the blood was the same in each case; i .e., 4 ml. of blood and 8 ml. of water or of the combined glycocyamine and arginine solutions. 4 ml. of 10 per cent sodiuin tun&state followed by 4 ml. of 0.66 N .H2S04 were then added to each mixture. The an,alyses were carried out on the filtrates. The concentrations of glycocyamine and arginine in these filtrates were further reduced by one-half in the analytical procedure. Concentration of glycocyamine in Folin-Wu filtrate Concentration + 2 in Folin-Wu filtrate Glycocyamine added Arginine added Found After subtracting amount originally present 1.0 0.5 0.2 0 .1 0 0 .5 0.5 0.5 0.5 0 0 0 0 0 0 2.0 1.5 1.0 0.5 2.0 0.99 0.55 0.24 0.15 0.05 0 .55 0.55 0.54 0.54 0.05 0.95 0.50 0 .19 0.10 0 0.50 0.50 0.49 0.49 0 I SUMMARY A micromethod is described for the determination of glycocyamine iri biological fluids and tissue extracts. The advantages of this method over those previously described are that added glycocyamine is recovered quantitatively; it is faster and more convenient. 29 .J 388 Glycocyamine Determination BIBLIOGRAPHY 1. Borsd6~ H., and Dubnoff, J. W., J. Biol. Chem., 138, 389 ·(1941). 2. Borsodk~'•H., Duonoff, J. W., Lilly, J. C., and iVIar.Hott, W., J. Biol. Chem., 138, 405 (1941). . ,., 3. Weber, C. J., J. Biol. Chem., 86, 217 (1930); 88, 353 (1930). 4. Bodansky, M., J. Biol. Chem., 116, 641 (1936). 5. Davenport, H. W., and Fisher, R. B., Biochem. J ., 32, 602 (1938). The Formation of Glycocyam.ine Glycocya.mine Glycine Arginine CHNH? CH2 COOR kidney COOR + NH?, CH2 HN~ ,, NH? C NH COOR CH2 CH2 CH?. IHN ~c ,,NH2 j NH 30 Reprinted from 'f'aE J ounNAL QF' BIOLOGICAL CHEMISTRY Vol. 138, No. 1, March, 1941 THE FORMATION OF GLYCOCYAMINE IN ANIMAL TISSUES* BY HENRY BORSOOK AND JACOB W. DUBNOFF (From the William G. K.erckhoff Laboratories of the Biological Sciences, Ccilifornici Institute of Technology, Pasadenci) (Received fo_r publication, September 18, 1940) It was shown in preceding communications that glycocyamine is converted into creatine by surviving liver slices (1). Our findings indicated that the methylating agent is methionine or a derivative of methionine. Liver slices "can methylate glycocY.amine rapidly enough to permit assignment to the liver alone, if necessary, of the task of making good the loss of creatine and creatinine in the urine. This holds for the livers of all mammals studied. We found no evidence of this niethylating mechanism in any 11) other tissues, except possibly slight activity in the kidney. In the pigeon the kidney is as effective in this respect as the liver. l These experimental facts were corroborated (as far as the rat is concerned) by experiments on living animals with tracers. Bloch and Schoenhei~er, using. N 15, found that glycocyamin.e is readily converted to qj·eatine (2): Du Vigneaud and hfi{ collaborators fed rats methionine with deuterium in the methyl ,group; after only 3 days a r~latively large quantity of deuterium was found in the muscle creatine (3). Glycocyamine has had a favored position among the possible precu~soi!s of1 creatine. It is nearer to creatine structurally than •1rq . any other precursor which has· 1been proposed; and its con vertibility to creatine in the living' organism has been proved. The case against glycocyamine r~sted on two argu'ments: (1) that only a small frac.~ion, 5 t~ 15 per cent, of administered glycocyamine is methylated, and (2) that glycocyamine had not been found as a normal constituent of animal tissues. It has, however, been isolated by. Weber; from . human and dog urine (4). ,. ► * A summary of this work has appeared (Science , 91, 551 (1940)) . 389 31 390 Glycocyamine Formation The findings with surviving liver slices and the tracer studies reinstated glycocyamine, more firmly than before, as a possible normal precursor of creatine. The quantity of creatine synthesized daily is so large that, if glycocyamine is its normal precursor, an active mechanism for the formation of glycocyamine must exist. We undertook a search for this mechanism. The first necessity was an adequate analytical method. Such a method is described in the preceding communication (5). With it we have found that kidney slices rapidly form glycocyamine from arginine and glycine. All other tissues tested are negative in•this respect. This interaction of arginine and glycine is also catalyzed by thoroughly macerated cell-free kidney tissue suspended in a phosphate buffer solution . . The formation of glycocyamine from arginine and glycine is a new biochemical reaction which may be ca1led "transamidination." We propose that the enzyme be designated "glycine-transamiclinase." The discovery of this reaction provides direct proof that argini:q.e and glycine are precursors of creatine. '•. Bloch and Schoenheimer fed ammonia containing N 15 to rats and later found the isotope in the amidine nitrogen of creatine. After glycine containing N 15 was fed, the isotope was found in cre'atine in the sarcosine nitrogen. In a later communication the same authors presented more direct evidence (again obtained oy the use of N 15) which confirmed the findings we had reported that glycocyamine is formed by the transfer of the amidine group from arginine to glycine. They fed Z( + )-arginine with N1 5 in the amidine group to rats (6); afterwards the creatine in the muscles had a far higher isotopic content than after the administration of isotopic tiinmonia, urea, or any other amino acid except glycine. It was so high that they considered that the amidine group of creatine must have originated from arginine. The present communication contains the details of our experimental procedure, our findings on the effects of a fairly large number of amino acids and derivatives of arginine and of glycine, surveys of the capacity for glycine transamidination of the organs of a number of common experimental animals, and the results of some studies of the effect of concentration of reactants, pH, and time on the rate of transamidination. 32 H. Borsook and J. V\T. Dubnoff 391 , T_echnique and Results . The tissue slice' technique used and the· details of the reaction vessels are described in a previous communication (7). At the end of an experiment the c~ntents of the reaction vessels were transferred with the slices to test-tubes graduated at 20 ml.; the vessels were washed with three 2 ml. portions of 0.02 M phosphate buffer solution at pH 6.0 and the washings added to ,the main solution. The P{I was adjusted finally to 6 with a drop of 0.5 N hydrochloric acid. The test-tubes containing the slices, the main solutions, and washings were kept in a boiling water bath for 10 minutes, after which they were cooled to room temperature, and made up to the 20 ml. mark with water and mixed by shaking. These solutions were then filtered. 5 ml. of the clear protein-free filtrate were analyzed for glycocyamine by the procedure described in the preceding communication (5). The coagulated slices and protein in each test-tube were transferred to small glass weighing dishes, heated overnight at 105°, and,. after cooling in a desiccator, were weighed. Table I is the detailed protocol of a typical experiment. A significant amount of glycocyamine is formed when arginine alone is added to the Ringer's solution. Glycine without arginine also leads to a slight increase in glycocyamine. When both a~ino acids are added together, the increase in glycocyamine is more than 10 times the increase with arginine alone. The increases obtained with arginine alone and with glycine alone indicate· either the presence of these amino acids in the free state (more glycine than arginine) or their formation in small amounts by autolysis within the slices. In the experiment whose results are recorded in Table I, and in a number of others, creatine analyses were carried out with the specific bacterial (NC) enzyme of Dubos and Miller (8, 9) . No evidence of creatine formation was found. Th6 steps in the proof that the subst.ance we were measuring was glycocyamine were as follows: Autoclaving in acid solution produced a substance which was adsorbed by Lloyd's reagent and gave a positive test with the Jaffe reagent. This chromogenic material was not digested (before autoclaving) by the NC bacteria tr ·Dubos and Miller ltnder conditions in which creatine and creatinine were completely digested. A strongly positive Saka- 33 '1' 392 Glycocyamine Formation guchi test was obtained in the unautoclaved solution after all the arginine was remov,:ed by exhaustive adsorption on permutit. The depth of color which the unknown solution gave with the Jaffe reagent (after autoclaving) corresponded, assuming it to be glycocyamine, to the intensity of color it gave in the Sakaguchi reaction after removal of the arginine. A liter of solution was TABLE I Protocol of Typi cal Experiment Showing Format·ion of Glycocyamine from Arginine and Glycine by Rat Kidney Slices Ringer's solution, 38°, 4 hours. Dry weight of slices Ringer's solution 0.Q2M arginine in Ringer's solution Concentration G lycocya1nine 0.04 M glycine of glycofound per " in Ringer's cyamine in gm. fresh aliquot taken 100 tissue JM~lution for analysis (I) (2) (3) }J, (4) (5) (6) mg. ml. ml, f ) ' 'ml. mg. per cent -mg. 26.4 30.4 22.5 21.0 20 .0 22.0 23.6 21.4 4 4 3 3 3 3 2 2 1 1 { 1 1 1 1 1 1 0 .02 0.04 0 .10 0.11 0 .04 0.05 1.03 , 0.97 ,lJ 6 q 11 36 42 16 18 349 . 363 The figures in Column 6 are obtained oy multiplying those' in 8olumn 5 by 8000 and "dividing them by the dry weight of the tissue in mg. (Column 1). The figur e 8000 is obtained as follows: the solution is diluted 5-fold before analysis ; in the course of analysis it l\nderg9es a further 1: 1 dilution ; the results in Column 5 expressed as mg. per cent must be divided by 25, since there were only 4 ml. of the original reaction solution ; to express the results on the basis of 100 gm. of fresh tissue, the factor 100,000/5 X W is used, W being the dry weight in mg. given in Column 1. The factor therefore is 5 X 2 X 1/25 X 100,000/(5 X W) = 8000/W. now collected in which kidney slices had acted,"ll.pon arginine and glycine, and which contained, according to knalysis, about 50 mg. of glycocyamine. The glycocyamine was isolated by adsorption on Lloyd's reagent in acid solution, elution with baryta, removal of the arginine by repeated adsorption with permutit, and crystallization from glacial acetic, acid as glycocyamine acetate. These crystals were the characteristic needles and thin prisms (10). 34 H. Borsook and J. \V. Dubnoff 393 The free glycocyamine was regenerated from the acetate by boiling in dilute aqueous solution· and crystallized by evaporation of the water. 25 mg. of crude glycocyamine were thus obtained. It was thrice recrystallized from water, with a final yield of 11 mg. 'of the pure dry material which was analyzed. It gave the following figures. 1 Observed. Theoretical for glycocyamine. C 30.8, H 5.95, N 35.8 " 30.8, " 6.0, "35.9 I',' T ABLE II J.> ,Formation of Glycocyamine by Rat Kidney Slices from Arginine and Glycine ., . or Glycine Derivatives Glµcose-R1Jl£~;,s solution, · 3' hJ urs, 38~0 • • Concentration ~f arginine 0.005 M; glycine'or derivatives 0.CJ1f'r M~'... ,;: ~ /Ji &gfoine Glycine or glycine derivative I I Glycocyamine found per 100 gm. fresh tissue mg. 8 ')' + + + + + + + + + + + Glycine Betaine Glutathione Glycine anhydride Glycylglycine Glycolic acid + ammonia 1 Hippuric acid • Leucylglycine' ..,, ' Sarcosine " • an hydride 35 19 44 296 382 8 436 , 6 35 254 109 6 Table II summarizes the relative effectiveness of glycine and some glycine derivatives as precursors of glycocyamine. iu1l'he effect of the glycine peptides is ac·<wtinted for on the hypothesis that these are first hydrolyzed and that it is the free glycine which reacts with arginine to form glycocyamine. The argument is as follows: The rate of glycocyamine formation is proportional to the concentration of free glycine (Table VIII). When glycine 1 We are indebted to and wish to thank Dr. A. J. Haagen-Smit for these analyses. 35 394 Glycocyamine Formation peptides were the source of the glycine, the concentration of free glycine was initially 0 and only in the course of the 3 hour experim:ental period did it approximate 0.01 M, whereas when glycine itself was added the initial concentration was 0.01 M. Hence smaller amounts 0£ glycocyamine w~re formed from glutathione, leucylglycine, and hippuric acid than from the same initial concentration of glycine. Glycylglycine gave a higher v_alue than glycine, because on hydrolysis it yields 2 molecules of glycine and as a result the concentration of glycine rose well above 0.01 M before the end of the 3 hour period. It follows, if the above is the correct explanation of the effect of glycine p~ptides, that rat ·kidney co~tains a dipeptidase for leucylglycine, an enzyme which liberates glycine from glutathione, and no enzymes capable of forming free glycine at a significant rate from betaine, glycine anhydride, or hippuric acid. The results with sarcosine indicate that this substance is demethylated without deamination in rat kidney. Separate analyses showed that no creatine was formed; this proved that demethylation of the sarcosine had occurred prior to the transamidination. The kidney contains, therefore, an enzyme which demethylates sarc9sine. • These findings with sarcosine are complemented by those of Bloch and Schoenheimer (2, 11) who, using N 15 as a tracer, found . that sarcosine is converted to glycine in vivo and that in the course of the demethylation the glycine nitrogen originally attached to the carbon chain is not replaced. Their experiments therefore excluded intermediate deamination of sarcosine in the course of its conversion to glycine. Analogous to its inability to hydrolyze glycine anhydride the kidney is unable to hydrolyze sarcosine anhydride. ' The negative ·re-sult' shown in Table II with glycolic acid and ammoniaindicates lthat rat kidney is unable to form glycine from these two substances at a, significant rate. The experiments summarized in Table III revealed that rat kidney evidently is able to synthesize arginine from citrulline. Thd,other possible amidine donators which were tested, guanidine, orriithine, and urea, were negative. The positive result with citrulline was retested in a number of more adequately controlled experiments. A group of typical 36 H. Borsook and J . W. Dubnoff 395 TABLE III Formation of Glycoe,yamine by Rat Kidney Slices from Glycine and Arginine or Other Possible Donp,tors of,Amidine G~oup Qlucose-Ringer's solution, 3 hours, 38°. Concentratlon of glycine 0.01 M; of arginine or other amidine donators 0.005 M. , . ' Glycine Amidine donator Glycocyamine formed per 100 gm. fresh tissue mg. - V ' '} . l( + )-Arginine + r " ,jt fl ' l( + )-Citrulline, + ammonia + + + Guanidine d(-)-Ornithine + ammonia Urea ' !.,T ABLE 35 68 498 330 16 12 33 IV Formation of Glycocyamin~,Ji'o'fl'/,_ Glycine and Citrulline by Rat Kidney Slices and by C,§ll-Free Macerate of Rat Kidney Slices in glucose 1Ringer's solution, macerate in 0.1 M phosphate buffer, pH 7.0, 3 hours, 38°. Glycine 0.01 M; amidine donator (ammoni'a, arginine, citrulline, or ornithine) 0.005 M. ·, Glycine Amidine donator Glycocyamine formed per 100 gm. fresh tissue mg. Kidney slices ' ~ t,, Cell-free macerate + - + + + + - + + Ammonia Arginine Citrulline Arginine Citrulline " + ammonia " Ornithine + Arginine Citrulline+ ammonia 22 31 18 68 58' 357 200 227 8 30 120 59 ' 8 results is shown in Table IV. Citrulline was only slightly les effective than citrulline plus ammonia. Included in Table IV are some typical ·i'esults obtained with a 37 396 Glycocyamine Formation cell-free macerate of kidney. This suspension was less active than an equivalent amount of kidney tissue in the form of slices but qualitatively the results were in every respect the same. The suspension was made by the homogenizing procedure of Potter and Elvehjem (12). The kidney was stripped of its capsule, split down the longitudinal axis, the pelvic fat and the medulla cut away, the remainder homogenized with 4 times its weight of 0.01 M phosphate buffer at pH 7.0, and the resulting suspension passed through gauze. Such a suspension retains all its transamidinase activity for at least 2 months if kept in an ice box. The optimum pH of glycine-transamidinase is in the neighborhood of pH 7.0 (Table V). ,. TABLE V Effect of pH on Activity o{Glycine-Transamidinase 38°, 4 hours. Glycine 0.01 M; arginine 0.005 M. pH fH 6.0 6.5 7.0 7.5 8.0 Glycocyamine formed per 100 gm. fresh tissue mg. 214 259 320 278 246 The activity of glycine-transamidinase in cell-free solution is unaffected by 0.001 M potassium cyanide or by carrying out the reaction in vacuo. For example, one extract under the same conditions as those described above formed 67 mg. of glycocyamine per 100 gm. of fresh tissue; in the presence of 0.001 M potassium cyanide it formed 80 mg. and anaerobically, 72 mg. The following amino acids and amides were tested with rat kidney slices and arginine as possible precursors of the glycine radical in glycocyamine. The initial concentration in every case was 0.005 M . They were all negative: d-alanine, dl-alanine, Z-asparagine, Z-aspartic acid, Z-cysteine, Z-cystine, d-glutamic acid, d-glutamine, Z-hydroxyproline, Z-histidine, dl-isoleucine, lleucine, d-lysine, dl-methionine, dl-norleucine, Z-proline, dlphenylalanine, dZ-serine, d-threonine, Z-tryptophane, Z-tyrosine, 38 H. Borsook and J. Vv. Dubnoff 397 and d-valine. These· negative results indicate that under the conditions of these experiments none of these amino acids is· a precursor of glycine. Glycine-transamidinase activity was found in the kidney of every animal tested except the frog (Table VI) . Whenever activity was found in kidney slices, it was also found in the cellfree extract. The beef and sheep kidneys were used at least 24 hours after the animals were slaughtered; they were obtained in a butcher shop. Iri the cases of all the other animals the kidT A BLE VI Formation of Glycocyamine from Glycine and Arginine by Kidney Slices and by Cell-Free Macerate of K idney of Various Animals Slices in glucose-Ringer's solution, macerate in 0.1 M phosphate buffer, pH 7.0. Glycine 0.01 M; arginine 0.005 M. 38°, 3 hours. Glycocyamine formed per 100 gm.' fresh tissue by Animal Kidney slices Cell-free macerate mg. Beef ..... . . .. . .. . . . .. .. .... . . ... .. . Cat . . ......... . . _. . . . . .. . . ..... . . . Dog .. .. .. . . . ...... . .. .. . . . .... . .. . Frog..... ........ . . . .. . .... . . . Guinea pig ....... . . .. ... ... . . .... . Pigeon ..... . .. . ......... . . . R abbit ..... . . . . . . .... .. .. . . . . .... . Rat ..... . . . , . . ................ •... l Sheep .. . . 190 93 32 281 480 38 14 27 187 16 160 357 120 0 160 neys .yere removed immediately after the animals were killed and the extracts made soon afterwards. The negative results with extracts of frog kidney call for furthei: study. Liver slices and cell-free extracts of heart and of muscle of all the animals listed in T able VI were tested fbr glycine-transamidinase activity. Except in the case of the pigeon (see below) they were all negative. The blood, brain, intes't ine, and spleen of the rat were also examined; they were negative. Rat liver slices also gave negative results with glycine plus arginine, urea, or guanidine. Before the advent of Weber's method no glycocyamine could be ·demonstrated in animal tissues. With this method glyco- I f, 39 398 • Glycocyamine Formation cyamine was detected in urine (4), in intestine, testes, and kidney (13). ·' Using our more sensitive method, we found glycocyamine to be widely distributed in the tissues of the rat. The concentrations (mg. per 100 gm. of fresh tissue) were blood 0.5 to 1; brain, heart, liver, skeletal muscle, and spleen 3 to 6; small intestine 10; kidney (cortex) 15 t0 30 . The question a11ose whether transamidination in the kidney can be sufficiently rapid under physiologica:J conditions to ·account for the total production of creatine in the body as indicated by the TABLE VII Rate of Glycocyamine Formation by Rat Kidney Slices at 38° from Arginine Initially 0.005 M and Glycine 0.01 M Glycocyamine per 100 gm. fresh tissue Time • Found Formed Average rate of formation per hr. hrs. mg. mg. mg. 0 0.5 1.0 2.0 4.0 6.0 32 95 164 315 63 • .... . 132 • 283 • 473 709 126 132 142 505 741 118 118 daily excretion jn the urine of creatine and creatinine. A number of experiments were carried out to obtain what information we could on this point. Table VII shows that the glycine-transamidinase in rat kidney slices remained practically un1mpaired for 6 hours. This is in accord with the stability of the enzyme in cell-free extracts. In Table VIII are some figures on the effect of the concentration of the reactants, arginine and glycine, on the rate of transamidination. With equal arginine and glycine concentrations from 0.001 to 0.0001 M the rate was approximately line~r with respect to concentration. • •. t •• We can estimate what the rate of glycocyamine formation .in the kidney must be to make good the loss of tissue cr:;tine r ~r c which appears in the urine as creatinine. Two human kidneys ~ 40 H. Borsook and J . VV. Dubnoff 399 weigh approximately 300 gm. An average figure for the creatinine excreted in the urine in 24 hours is 1. 7 gm. To make good this loss the average hourly production of glycocyamine which is nec.ess~ry must be approximately 25 mg. per 100 gm. of tissue per hour. • • • ~" r< 'IT'! This rate of glycocyamine formation would have bee1t0Qlbtaii?:eq in the experiments of Table VIII with concentrations 9f glycine and arginine between 0.0005 and 0.001 M . This is probably the concentration range of these amino acids in kidney. The basis of this estimate is as follows: The arg~nine concentration in dog TA BLE VIII Variation in Rate of Glycocyamine Formation by Rat K idney Slices with Different Concentrations of Arginine and Glycine Ringer's solution, 4 hours, 38°. GlycoInitial cyamine concentration formed ·> per 100 fr esh Arginine Glycine gm. tissue Arginine Glycine Initial concentration - -- -mole - - -- - -- per l. mg. mole per l. mole perl. 0.02 0.01 0.01 0.01 0.005 0.01 0.0025 0.01 0.00125 0.01 0.000625 0.01 190 590 533 535 303 201 0.005 0.005 0.005 0.005 0.005 0.005 0.02 0.01 0.005 0.0025 0.00125 0.000625 mole per l. GlycoGlycoInitial cyamine cyamine concentration formed formed per 100 per 100 gm. fresh Arginine Glycine gm. fres h tissue tissue --- --mg. 617 533 361 252 201 103 mole per l. 0.005 0 .0025 0.001 0.0005 0.0001 mole per l. ~81 1110.· Sil 0.005 , r 36o 0.0025 310 0.001 fa4 0.0005 78 0.0001 ..20 ' blood is bet'ween 2.6 and 3.9 mg. per cent (14) ; we have found 2.0 to 5.0 mg. per cent in human blood. This range corresponds to 0.0001 to 0.00025 M . The concentrations of these amino acids in the kidney are certainly muGR higher than in blood, since the total free amino nitrogen in kidne,y and other tissues is 10 times or more that in blood (15, 16). 'The rates of glycocyamine formation recorded in Table VIII are sufficient therefore to make good the total loss of creatine from the tissues. This estimate is, of course, based on the assumption that the rate of glycocyamine formation in' _h uman kidney in vivo is of the same order as in rat •kidney slices in vitro, which seems not unreasonable. ( j 41 400 Glycocyamine Formation DISCUSSION The experimental results given above, confirmed and complemented as they now are by the findings in studies with tracers, make a strong case for the following mechanism of creatine formation in animals: (a) arginine and glycine in the kidney form glycocyamine; (b) •in the liver glycocyamine and methionine form creatine. The mechanisms for these reactions are widely distributed in mammals. The quantitative aspects of the data show that the transamidination mechanism for the formation of glycocyamine is fast enough to replace all the creatine lost from the body. Other mechanisms of creatine formation are not excluded; but the tracer studies indicate that the arginine-glycinemethionine mechanism is quantitatively the most important one in vivo. The argument contra glycocyamine, that it had never been found as a 9onstituent of animal tissues, antedated the development of suitable analytical methods. This argument is now completely dispos~d of by the detection of glycocyamine in practically all tissues. •The controversy whether arginine is a precursor of creatine (17, 18) may b~ considered as settled in favor of the affirmative. The path of its conversion to creatine, however, is different from what was gene~ally believed. It was proposed that arginine was converted ·to glycocyamine by way of deamination and {3 oxidation; thus all of the nitrogen was derived directly from argmme. This made it necessary to explain away the evidence which suggested that glycine played some part in creatine formation, unless it bf! in th() methylation of glycocyamine. Bergman:t;t and Zervas were the exception. They observed that "triacetyl anhydro arginine" and glycine ethyl ester reacted in the absence of water to give a fi:l,irly good yield of diacetyl glycocyamin~ ethyl ester (19); and the same arginine derivative with sarcosine ethyl ester gav.e diacetyl creatine ethyl ester (20). Their view was that, "der wesentliche Punkt der biologischen Kreatinbildung in einer direkten Umsetzung eines reaktionslustigen Argininabkommlings mit einem Aminoathanderivat beruht." This clear statement that the guanidine group of creatine arises in vivo by transamidination needs now to be modified in two respects: that an enzyme, glycine-transamidinase, and not a 42 H. Borsook and J. W. Dubnoff 401 _ .. " ·'-1 yNH2 reactive derivative or split-product renders the: .,.:+Q~ • group ~NH in :rurginine labile as it is in "triacetyl an.hydro arginine." The enzyme arginase exerts a similar influence. The other modification is that glycine itself and not a derivative is the acceptor of the amidine group in the biological transamidination reaction. We may now infer that one of the reasons for the essential character of arginine and methionine in the rat is that they participate in the fotrhation of creatine. These amino acids, of course; hllso serve other functions, for example the role of arginine in urea '·formation -and of methionine in other methylation reac.,. tions (3), in addition to their participation in the constitution of tissue protein. The recent findings on essential amino acids for the chick indicate that arginine and , glycine are required for creatine formation in this animal. Arnold et al. (21) reported that arginine is essential for rapid growth. This was confirmed by Klose et al. (22) who showed 'further that arginine is necessary for maintenance as well as growth. Recently Klose and Almquist reported that citrulline is as effective as arginine, whereas ornithine alone or with urea is ineffective (23) . We have found (Tables III and IV) that glycocyalffiine is formed from citrulline and glycine, the citrulline presumably being first converted to arginine, while ornithine, with or without added ammonia, is completely negative. The parallel between the amino acid requirements for creatine formation and for growth was extended further when Almquist et al. found that glycine is essential for the growth of the .chick (24) and that creatine as a substitute for glycine is even more effective than glycine itself. Glycolic acid and betaine could not replace creatine (25). We have examined the organs of the pigeon for tra,nsamidinase activity. Activity was found not only in the kidney but also in heart, liver, and skeletal muscle. The limiting amino acid appeared to be glycine; i.e., nearly as much glycocyamine was formed when glycine alone was added as from glycine and arginine together. The differences between experimental and control were, however, small in absolute terms compared with those found in the kidneys of other animals. Although these 43 " •• f 402 e Glycocyamine Formation differences were greater than could be ascribed to analytical or sampling variations, we cannot consider our findings in the pigeon as established until more determinations have been made. For this reason the detailed figures are not presented hei'e. iWe hope -that 0th.er :workers with facilities for a study such as this oru birds may unclertake the investigation. Our laborato~-y does riot at present possess such facilities. , r,i , 1 SUMMARY I 1. Beef, cat, dog, guinea pig, pigeon; rabbit, rat, 1and . sheep kidaey form glycocyamine from argirune· and glycine. This r.eaction is catalyzed by cell-free extracts of kidney as well · as by surviving kidney slices. . J 2. It is proposed that this reaction be designated ·"transamidination," and the enzyme "glycine-transamidinase." The optimum pH of this enzyme is about 7.0. It is not affected by potassium cyanide nor by anaerobiosis. ,. , , 3. Transamidination does not occur in the 'liver, heart, or sketetal muscle of the animals mentioned above; the blood brain, and spleen of the rat were tested also and found ina.ctive. It is possible that in the pigeon a low glycine-transamidinase actiw.ty resides in liver, heart, and skeletal muscle as well as ,in kidney. - 4. Glycocyamine is also formed in the kidney from glycine and citrulline. 'Glycine- plus ornithine (with or without ammonia,), urea, or guanidine is negativ.e in this respect'. 'j i5. A large number of amino acids, several amides, and anhydrides were tested as possible precursors of the gl;¼cine radical of glycocyamine. They· were all;:;-negative, as w:as also _glycolic acid plus ammonia. r::. f , 6. Glycocyamine is formed from argi:rline, and sarcosi,1;1e: .-. E vidence is presented that the sarcosine is ·first demethylated, thus being converted to glycine, indicatiytg,the presence of a .dem.e thylating enzyme in kidney. Sarcosine anhydride·'is negative. · ,i. 7. The above findings, complemenited by t he tracer studies .in the laboratories of Schoenheimer and of du Vigneaud, and in conjunction with our previous findings, prove the existence of tlie following mechanism ,of creatine formation iri animals: arginine and glycine., form glycocyamine in the kidney,; ~the glycocyamine ds methylated in the liver by methionine (or a derivati:v.e of methionine) to form creatine. 44 H. Borsook and J. vV. Dubnoff 403 8. Quantitative aspects of the dat a indicate that all of · the creatine formed in animals may normally be formed by this mechanism. 9. Evidence of the generality of transamidination is seen in t he close parallel between the above findings and those on amino acids essential for the growth of the chick. BIBLIOGRAPHY 1. Borsook, H., and Dubnoff, .J. W., J . Biol. Chem., 132, 559 (1940); 134, 635 (1940). 2. Bloch, K., and Schoenheimer, R., J. Biol. Chem., 133, 633 (1940) . 3. du Vigneaud, V., Chandler, J.P., Cohn, M. , and Brown, G. B., J. Biol. Chem., 134, 787 (1940). 4. Weber, C. J ., Proc. Soc . Exp . Biol. and Med ., 32, 172 (1934); Proc. Am. Soc. Biol. Chem., J. Biol. Chem., 109, p. xcvi (1935). 5. Dubnoff, J. W., and Borsook, H., J. Biol. Chem., 138, 38i (1941). 6. Bloch, K., and Schoenheimer, R., J . Biol. Chem., 134, 785 (1940) . 7. Borsook, H., and Dubnoff, J. W., J. Biol. Chem. , 132, 307 (1940). 8. Miller, B. F ., and Dubos, R., Proc. Soc. Exp . Biol . and Med., 36, 335 (1936). 9. Dubos, R., and Miller, B. F., J. Biol. Chem., 121, 429 (1937). 10 . J affe, M., Z. physiol. Chern., 48, 430 (1906) . 11. Bloch, K., and Schoenheimer, R., J . Biol. Chem., 136, 99 (1940). 12. Potter, V. R., and Elvehjem, C. A., J. Biol. Chem., 114, 495 (1936). 13. Bodansky, M., J. Biol. Chern., 115, 641 (1936). 14. Weber, C. J., J. Biol. Chem., BB, 353 (1930). 15. Mitchell, H . H., Nevens, W. B., and Kendall, F. E., J. Biol. Chem., 62, 417 (1922). 16. Van Slyke, D. D., and Meyer, G. M., J: Biol. Chern., 16, 197 (1913-14). 17. Hunter, A., Creatine and creatinine, Monographs on biochemistry, London and New York (1928). 18. Mitchell, H . H., and Hamilton, T. S., The biochemistry of the amino acids, American Chemical Society monograph series, New York (1929). 19. Bergmann, M., and Zervas, L., physfol. Chern., 172, 277 (1927). 20. Bergmann, M., and Zervas, L., Z. physiol. Chem., 173, 80 (1928). 21. Arnold, A., Kline, 0. L., E lve hj ern, C. A. , and Har t, E. B., J. Biol . Chern ., 116, 699 (1936). 22. Klose, A. A., Stokstad, E . L. R., and Almquist, H. J., J . Biol . Chem., 123, 691 (1938) . 23. K lose, A. A., and Almquist, H.J., J. Biol. Chem., 136, 153 (1940). 24 . Almquist, H.J. , Stokstad, E. L. R., Mecchi, E., and :Manning, P. D. V., J. Biol. Chem., 134, 213 (1940). Almquist, H. J. , Mecchi, E ., Stokstad, E. L. R., and Manning, P . D. V., J. Biol. Chem., 134, 465 (1940). 25. Almquist, H.J. , and Mecchi, E., J. Biol. Chem., 136, 355 (1940) . z. 45 Reprinted from THI, JOURNAL OP BIOLOGICAL CHEMISTRY Vol. 138, No. l, March, 1941 THE FORMATIOf. OF GLYCOCYAMINE IN MAN AND ITS URINARY EXCRETION BY HENRY BORSOO:K, JACOB W. DUBNOFF, JOHN C. LILL;¥, AND . ·t~WILLIAM MARRIOTT • : (From the William G\ Kerckhoff if,abqratories of the BiologicalSciences, California Institute-·ot Te°bhnology, Pasadena) I CRecei ved for publication, September 18, 1940) ) Glycocyamine was' first isolated from human and dog urine and identified by Weber (1- 3). He supported the view that glycocyamine is a normal precursor of creatine and that its appearance in urine (2) is "an overflow phenomenon of an intermediate metabolic product ... " He expressed no views on the mechanism of its formation. Glycocyamine is formed by the interaction of arginine and ., glycine in the kidney (and probably only in that organ) of the cat, dog, guinea pig, rabbit, and rat (4). In order to ascertain whether this reaction occurs in man we have studied the effect of arginine and glycine, ingested separately and together, on the glycocyamine content of urine and blood. The subjects were seven normal men. The methods used for the determination of glycocyamine in urine and in blood have been described (5). The first experiments consisted in comparing the urinary exfretion of glycocyarn.ine during 24, later only 12, hours in which no protein was ingested with that on a day in which gelatin was ingested. Gelatin consists of about 25 per cent glycine and 9 per · cent arginine (6). Each experiment ran for 2 days. On the evening preceding the 1st day the subject had a normal meal at 6 p.m. and 300 ml. of water in the course of the evening. During the, ~ollowing day he drank 1700 ml. of orange juice and an additional 300 ml. of water. This was the "non-protein" day in Fig. 1. Ne~t day the subject drank 875 ml. of orange juice, and 1000 ml. of water containing 275 gm. of sucrose. He ingested 405 46 406 Glycocyamine in Urine 65 gm. of gelatin between ·8 and 9 a.m. The gelatin contained 16 gm. of glycine and 5.8 ' gm. of arginine. This was the "gelatin" day. A typical result is given in Fig. l. It shows the rapid rise in the urinary~excretion of glycocyamine after the ingestion of gelaC) w i~ 20Qir--,-__,r:----t-----t----t-----it----t---+-c-i Uo ')J ·\ c;S::z::: ~~ 100111'--+----'<~'-'=c!s'-""'!-'---+-r-'!I--"<-+--~--\ c::t- :::,:z: _j2 .Q 0 )•>·cm u., ... § 20 i----.,,-P----+----1--!----+---+--l 1 i b, ~ z /GELATIN u ~ \ cj(/') 15 g--,,-+-----+----+-+---+--!----+--+----a wa::: z:::> -o .. J!) Jb- :·• ,/ ~::z::: JQ1--.'-+---!--t------\+---+---- -+----+,-,--l uO o3: ~f-- <3~ 0 :::e 10 12NOON2 4 6 8 TIME Fm. 1. Glycocyamine excretion in urine. 11he ordinates represent the glycocyamine excreted in the urine in the 2 hour period ending at the time it\dicated; the abscissae, the time at which the urine was voided. 65 gm . of gelatin were ingested between 8 and 9 a.m. The horizontal dght-hand ends of both curves represent the tot.al· ex'cretion. in the night urin e (10 p.m . to 8 a.m.) divided by 5. ti.' tin. The maximum was attainetl between 10 a~m·.··and 12 noon. By 4 p.m. this rate had subsided near to the value on the nonprotein day, and remained so auring the following 14 hours. This rapid rise and fall in: tlie. hi,te of glycocyamii1e excretion occurred both"Urhei1uthe argiciri'c:i1 1id glycine were taken in the form of gelatin and as the pure ainino acids (see FigsJ l and 2). It is improbable· that the increased urinary excretion of glyco-· 47 Borsook, Dubnoff, Lilly, and Marriott 407 cyamine observed after the ingestion of gelatin represents glycocyamine "washed out" of the tissues. If that had been the case, the excretion during the last 14 hours of the gelatin day probably would have fallen below that during the same interval on the nonprotein day; and the total amount of glycocyamine excreted over the 24 hour period would have been nearly the same on these 2 days. The 24 hour excretion of glycocyamine on the non-protein day was 65 mg. and on the gelatin day 117 mg. The lack of any relation in Fig. 1 between urine volume and the amount of glycocyamine excreted is in accord with this interpretation. Table I is a summary of the find ings in experiments similar to that in Fig. 1 in five other subjects. The data for the period from 6 a.m. to 6 p.m. only are given. TABI,E I E:ffecl of Ingestion of 65 Gm. of Gelatin on Urinary Glycocya-mine Gelatin day Non-protein day Subject No. Glycocyamine Urine volume ml. mg. •' ml. 475 1715 370 1220 470 43_ 734.3 34.0 32.9 24.4 Urine volume 1 2 3 4 5 ... ·890 1785 600 > 1520 1350 Glycocyamine mg . 87.5 114.0 60.0 88.2 81.6 The figures for glycocyamine excretion on the' non-protein days in Table I indid:te that over 24 hours they would have been of the order .of magnitude of 40 to 60 mg. Webe ••(3) reported average figures ~or 24 hours of 39.5 mg. for males ?,nd 75.1 mg. for females. Weber's values are lower, probably because the figures obtained with his analytical method represent only 75 per cent or less of the glycocyamine present. A series of experiments was carried out on t,vo subj ects in which arginine ,an.1el glycine, alone and together, were administered in the form of the pure amino acids. The two subjects remained on the identical diet for the 9 days of the experiment. Breakfast and luncheon consisted of orange juice and coffee; dinner was a normal meal in which the meat course was a weighed amount of Hamburger steak. During the day 200 ml. of water were drunk 48 408 Glycocyamine in Urine every 2 hours. Urine was collected every 2 homs from 6 a.m. to 6 p.m. On the 3rd day of this regime the urine voided every 2 hours during the day was analyzed; the values obtained gave the "normal" graph in Fig. 2. On the 5th day the subjects ingested 16 gm. of glycine between 8.30 and 9.30 a.m.; the glyc0, } . V. n J Loi· 11\ "' !5 40 0 ~ ,- "= 30 § 0 "' .::;~ I ~ ' GLYCINE + -:i ARGININE \ ' ~~ :i ~ zl ~ 20 1<"',1 , \ I :,: \ \ I •.A \ I ~"'-t-£lPI.NINE \ 1 _, , /,,1/;. ~---:: ... --.::~ ·- . .. GLYCINE NORMAL 10 12N00N 2 TIME ' r ,v >-:.::-.~ 4 Fm. 2. Glycocyamine excretion in urine . The ordinates represent the glycocyamine. · c,11eted in the urine in the 2 hour period endipg at the time indicated; th !J9~C~~~ae, the time at which the urine ,~as -void<r<l·, (!'he amino acids wei;e(·ngestecl, on each of the 4 days, in four portions between • 'H ' • J)'I•) 8.30 and 9.30 a.Il'l., 1'he blood concentrations, at successive 2 hom1 int~rvals aft'e r the begin'il.itfg 'of the experiment, on a day on which kP.Y n2 and arginine were ing'estcd together were 1, 0.6, 2, 0.4, 3, 0.7, 4, 0.8, afld! 5, 1.0 mg. per cent .;, , cyamine excretion on this day is designated in Fig. 2 as "glycine." On the 7th day between 8.30 and 9.30 a.m. the subj ects con~ sumed 16 gm. of glycine and 4.8 gm. of arginine, and on the 9th day 4.8 gm. of arginine. The glycocyamine excretion on these days shown in Fig. 2 -ig1 designated "glycine + arginine" and "arginine." Fig. 2 shows that after the ingestion of glycine alone more glycocyamine appeared in the urine. These increases were small, 49 Borsook, Dubno:ff, Lilly, and Marriott however, compared to those obtained when the two amino acids were taken together. Thus on the normal day the glycocyamine excretion in 12 hours was 29.0 mg., on the glycine day 54.3 mg., on the arginine day 55.4 mg., and on the glycine + arginine day 99.0 mg. The sum of the increases over the normal day with glycine and arginine when these were ingested separately was 51.7 mg.; when ingested together it was 70 mg. In fJ:ie second subj ect the glycocyamine excretion on the normal day was 31.2 mg., on the glycine day 30.7 mg., on the arginine day 28..4 mg., and on the glycine + arginine day "' .' 61. 8 mg. The increase in glycocyamine in the urine after the ingestion of 8f arginine and 'glycine was of the same order of magnitude as when the same amounts of these amino acids were ingested in the form of gelatin. The increase in urinary glycocyamine after the ingestion of gelatin may be ascribed to the arginine and glycine it contained. • • These findings taken in conjunction with those obtained with surviving slices of kidney and with kidney extracts (4) may be considered as, establishing the formation of glycocyamine by transamidination in man. Blood was analyzed every 2 hours on the days on which the subjects ingested the glycine and arginine together. In the subj ect of Fig. 2 the concentrations were, in mg. per cent, 0.6, 0.4, 0.7, 0.8, and 1.0. The peak of the urinary excretion of glycocyamine occurred between the second and third of these values. There was no discernible relation between the glycocyamine concentration in the blood and its increased rate of excretion in the urine. The findings were essentially the same in the other subject. We have made some preliminary observations on the excretion of glycocyamine by human subj ects with kidney disease. On an ordinary diet they excreted less glycocyamine than normal individuals and the increase after the ingestion of arginine and glycine was less also. Thus in one subject who was diagnosed as having subacute glomerulonephritis the excretion of glycocyamine in the urine in an experiment similar to that of Fig. 1 was on the non-protein day 1.5, 3.0, 2.5, 2.6, 2.1, and 1.0 mg. in successive 2 hour intervals between 6 a.m. and 6 p.m.; and on the gelatin day 1.0, 3.1, 5. 3, 2.4, 1.9, and 2.8 mg. J,i l 409 CJ• \ 50 • Glyc@'cy~rn.ine in Urine 410 stll\iMARY' 1.' When arginine and glycine are ingested together by human subjects, there is a .rapid rise in the amount of glycocyamine excreted in the uririe. This increase is greater than r'the sum of the increases which may occur when the same amounts of amino acids are taken separately. 2. A similar increase in glycocyamine excretion is observed after the ingestion of gelatin, which is rich in glycine and arginine. The order of magnitude of this increase is the same as that given by the quantity of arginine and glycine contained in the gelatin when these are administered as pure amino acids. 3. These findings indicate that in man glycocyamine is formed by transamidination. BIBLIOGRAPHY 1. ~ e1': •c . J ., Proc . Soc. Exp. Biol. and Med., 32,172 (1934) . 2. Weber, C. J., Proc. Am. Soc . Biol. Chem., J . Biol. Chem ., 109, p. xcvi (1935). ; . .; 3. Weber, C. J. , Proc . Am . Soc. Biol. Chem. , J. Biol. Chem., 114, p; 1gvii (1936). ,}){· 4. Borsook, H., and Dubnoff, J. W., J. Biol. Chem ., 138, 389 (1941). 5. Dubnoff, J . W., and Borsook, H., J. Biol. Chem., 138! 381 (1941). 6. Mitchell, H . H., and Hamilton, T. S., The biociheinistry of the amino acids, American Chemical Society mon.0graph series, New York (192~) ·ll r !J(' Hi . tl 51 ./_ I ...,.- Reprinted from THE JounNAL OF B10LOOICAL CHEl\HBTRY Vol. 141, No. 3, D ecember, 1941 'A MICROMETHOD FOR THE DETERMINATION OF ARGININE . 'rt L BY JACOB W. DUBNOFF L0 t1'"¾ _n{Rrom ,the William G. Kerckhojf Laboratories of the Biological Sciences, . . j alifornia Institute of Technology, Pasadena) H 1rt ' (Received for publica.tion, August 6, 1941) • I ~ I . l ." Micrometh0ds for the determination of arginine based on the -tJ7i f i.1~ us~ Jof tl\e , Sal~ ~l\Chi reagent have bee11 described (1-4). _;;\;'l).is{ reagent gives a strong color with glycqcyq.mine, arginine, ~nd other . mo1rnsubstituted. guanidine derivativei., •" . J. .i • I 1,~!-hWe:vious qc;ijm munication (5) a method for the determiqj tion o£6 1• ~:YJ-.u. rrhwocyamine,,,w s described based on the Sakaguchi reac,tioi1 n,., ,..-.;i! i :34l~.a+%: ql\anti~ative separation of glycocyamine from aq~inine bx- ,tles,1ive adso p tion of the arginine on permutit. I1;1 the_ m~tJ;!~/1 outlined qelow the separated arginine is eluted ) ro1tl th~ perrnutit and deter<IB\ned independently. , ..• .1 A number1 of rppmon non-chromogenic substances such , asr ammonia, urea, ht'?tidine, and ci;eatine reduce the color ob_~ained , with the Sakagu · t reagent. A change in the usual order ot, addition of the coio.r reagents similar to t,hat previously desoribed 1 Ji. s (5) decreases the interference by these compounds. ·, Reagents and Apparatus3 per cent sodium chloride. 0.3 per cent sodium chloride. J, Permutit according to Falin. Permutit can be regenerated_ after use by allowing 3 per cent sodium chloride to percolate. through and then washing with distilled water until chloride-free. This is most conveniently done in large batches on a Buchner ~~' 10 per cent urea in water. 0.2 per cent naphthol in absolute alcohol, diluted with 4·,yolumes of.the 10 per cent urea solution before ut,~,·• sHypobromite solution: .o·.Q6 ml. of liq~°id bromine are added "l ~ , • • I ~- • 711 .J } J 52 712 Micromethod for Arginine to 100 ml. of 5 per cent sodium hydroxide. Since this solution deteriorates, it should not be kept for more than 1 or 2 days, and is best kept in a refrigerator. 10 mg. per cent arginine in 0.1 N hydrochloric acid . Standard solutions are made on the day on which they are to be used, by diluting this stock solution with water. Adsorption column. The permutit is co11tained in the stem of a glass funnel whose dimensions are: u ppe1· spart 15 mm. external diameter, 100 mm : long; stem 7 mm. externaH:ii'ameter, 100 mm. long. The lower knd of the stem is slightly cci1\.stricted . . A small amount of cotton. is placed ab0ve the const1,i:ction. 0.9 gm. of permutit is poured in and tapped gently to settle the particles. • A 0.2 ml. micrp pipette whose contents are deliyered under pressure. The delivery time should be sufficiently sho~·t to insure rapid deliver 'and complete mixing (6). •'' Preparation of Solutions for Analysis____:For complete separation of glycocy~ and arginine the salt concentration of the sol'ution should noel5'erbver 0.5 per cent. If neither of these compouhds is pte~ent in 'Jounts over 2 mg. per cen't; tl5.e salt concentration 'may be1as high [is 1 per cent. Urine is usually diluted 5 to 10 times with water.'' Blood filtrates may be prepared by deproteinizing according to Folin and Wu or by heat coagul~tion at pH 6 after 1: 10 dilution with water. Tissue extracts are l:liluted to contain 1 gm. of tissue (fresh weight) in 40 ml. of suspension. The pH is adjusted to 6.0, and the suspension immersed in a boiling water bath for 10 minutes, cooled, and filtered. Analyses are carried out on the filtrates. e a. Procedure 5 ml. of the solution to be analyzed are passed through the permutit column and the small amount of glycocyam'ine 1:emaining in the column is removed with 5 ml. of 0.3 per cent sodium chloride. The combined filtrate contains all the glycocyamine. A test-tube graduated at 10 ml. is now placed under the funnel, and the arginine is eluted by passing 10 ml. of 3 per cent sodium chloride through the column. The solution in the test-tube is made up to the 10 ml. mark and shaken. A 2 ml. aliquot is taken for analysis. It is first cooled in an ice bath, then 0.5 ml. of the ice-cold naphthol-urea solution is added, 53 J. W. Dubnoff 713 and after 2 minutes 0.2 ml. of ice-cold sodium hypobromite solution added by means of the micro pipette. This pipette is placed above the solution level and kept away from the sides of the test-tube to prevent contamination by urea in the solution. The color is simultaneously developed in a series of standard solutions containing 0, 0.25, 0.5., 1.0, and 2.0 mg. per cent of arginine. 1 After 20 min tes the color development is complete and remains stable for 2 hours if the solutions are kept in an ice bath. The tubes are shaken for a few seconds to remove excess gas, warmed by immersion in water at room temperature, and the intensity of TABLE I Elution of Arginine from Permutit Column 5 ml. of a 2 mg. per cent solu tion of arginine in 0.5 per cent sodi um chloride were passed through the adsorption column followed by 5 ml. of 0.5 per cent sodium chloride. The arginine was then eluted by the salt solution as shown. Sodium chloride eluting solution Arginine recovered Amount Concentration ml. per cent per cent 10 5 10 5 10 10 10 5 10 0.5 1.0 1.0 2.0 2.0 3.0 5.0 5.0 10.0 0 2 40 60 100 100 100 88 88 the color measured in a spectrophotometer or a colorimeter, with light of approximately 0.525 µ (yellow-green). DISCUSSION As shown in...Table I, the complete removal of arginine from the l permutit column depends on the amount as well as the concentration of salt solution used. Under the conditions described, the separation of glycocyamine and arginine is complete over a wide 1 If an appropriate correction factor is applied, glycocyamine standards which are stable may be used. 54 714 1\1icromethod for Arginine range of concentrations (Table II). With concentrations of arginine of over 2 mg. per cent the color is too intense to be read and must be diluted before the color is developed . The blanks are yellow instead of colorless as· in the procedures described in the literature. In these older procedures a large excess of urea is added after a given interval to prevent destruction of the chromogenic compound by the hypobromite. During this interval other substances may compete for the hypobromite and reduce or completely inhibit the formation of color. The advantage of the procedure described TABLE II Limi ting Concentrations of Glycocyamine and of Arginine That Can Be Separated on Permutit Column All,.values a re concentrations in mg . per cent. The concentrations were those of the solutions passed through the perrnu tit column . The fin al filtrates were di lu ted to bring the concentration to a pproximately 1 mg. per cent before the color was developed. Arginine added G lycocyamine added Arginine found 0.0 0.0 0.0 0.0 1.0 2.0 5.0 10.0 20.0 2.0 2.5 5.0 10.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 l~d 2.0 5.0 9.6 18.4 above is that this interference is greatly reduced and for most purposes is negligible. Table III shows the recovery of arginine added to urine, kidney extract, and blood. Weber (2) has studied the influence of some non-chromo 6enic substances on color development. He found that 6 mg . per cent of ammonia, 12 mg. per cent of histidine dihydrochloride, 6 mg. ner cent of tyrosiBe, 6 mg. per cent of tryptophane, 40 mg. per cent of creatine, or 160 mg . .per cent of urea2 "either prevents all color 2 These a re the concentrations in the fin al dilu ted filtrates. 55 J. W. Dubhoff ,·. 715 formation or the color is so altered that even for qualitative purposes the test is worthless/' With the method described in this communicati:@n, 60 mg. per cent of ammonia, 5 mg. per cent of histidine hydrochloride, 8 mg. per cent of tyrosine, 8 mg. per cent of tryptophane, 20 mg. per cent of creatine, and 2000 mg. per cent of urea are without influence on the recovery of arginme. If tryptophane and histidine are present in excess of the III TABLE Recovery of Arginine Added to Urine, Kidney Extract, and Blood All the values are concentrations in the eluate measured in mg . per cent . The urine was diluted 1:4 with water. The kidney extract was made by gri nding up 1 part by weight of kidney with 40 parts of 0.5 per cent sodium chloride, bringing the extract to .pH 5.0, boiling 10 minutes, cooling, and filtering. The an alysis was carried out on the protein-free filtrate . In both the kidney extract and blood, arginine was added before protein precipitation. Argin ine Found 0.0 0.2 0.5 1.0 1.5 Kidney extract Urine Arginine add ed 0.13 0.13 0.345 0 .34 0.61 0.61 1.12 1.13 After subtraction of blank value Found After subtraction of blank value 0.15 0 .14 0.215 0.210 0.48 0.48 0.99 1.00 0.65 0.655 1. HJ 1.19 1.64 1.64 0.505 0.51 1.045 1.045 1.50 1.50 Folin-Wu blood filtrate Found 0 . 19 0.20 0.41 0.43 0.68 0.68 1.19 1.19 1.67 1. 71 After subtraction of blank value 0.21 0.22 0.48 0.48 0.99 0 .99 1.47 1.51 --· amount given, they are chromogenic, and, a further complication, histidine reduces the color given by the arginine present. In pregnancy urine the concentration of histldine may be as high as 160 mg. per cent (7) and the method is inapplicable as it stands. The method is satisfactory for normal urine, although the possible presence of other chromogens, such as methylguanidine, must be borne in mind. 56 • 716 Micromethod for Arginine SUMMARY A micromethod for the determination of arginine in biological fluids and tissue extracts is described. The advantages of this new method are that a complete separation of arginine from glycocyamine is effected, and that it gives satisfactory results in the presence of common biological substances which interfere in the procedure of previous methods. BIBLIOGRAPHY 1. Sakaguchi, S., .f. Biochern., Japan, 6 , 13, 133 (1925). 2. Weber, C. J., .f . Biol. Chem., 86 , 217 (1930); 88 , 356 (1930). 3. Fisher, R. B., and Wilhelmi, A. E., Biochem . .!., 32 , 606 (1938). 4. Jorpes, E., and Thoren, S., Biochem. J., 26, 1504 (1932). 5. Dubnoff, J. W., and Borsook, H., J. Biol. Chem., 138,381 (1941). 6. Borsook, H. , and Dubnoff, J. W., J. Biol. Chem., 131 ,163 (1939). 7. Racker, E ., Biochem. J., 34, 89 (~940). Lysine CH9. CHNH2 COOH ca; NH2 CH,;, CH2 Ornithine NH2 CH9. CH9. CH,;, CHNH2 COOH Proline H N C. ,~c-c~-·1 I ,~.c_ POOH · FORMATION Glutamic Acid COOH c~ c~ COOR CH9. OF + ARGININE Citrulline Arginine CH2 CH2 CH2 CHNH2 COOH NH NH CH2 CH2 CH2 CHNH2 COOH kidney HNI~c,, NH?.- 0 . 'c~ 2HN._ 57 Reprinted from THE JounNAL m· BIOLOGICAL CHEMISTRY Vol. 141, No. 3, December, 1941 , a THE CONVERSION OF CITRULLINE TO ARGININE IN KIDNEY* BY HENRY BORSOOK AND J ACOB W. DUBNOFF (From the William G. Kerckhoj]' Laboratories of the Biological Sciences, California InstituJe of Technology, Pasadena) (Received for publication, August 6, 1941) Glycocyamine is formed in the kidney by the transfer of the amidine group of arginine t0, the nitrogen atom of glycine. In the study of this reaction it was observed that glycocyamine was also formed from citrulline and ifycine. No other donor or precursor of the amidine group was fou:rld (1). . The most probable expl-;,riZ,ion of this effect of citrulline was that • was fi rst conver t ed t o arg1mne, ti'.)}. f rom wluc • h t h e am1'd·me group was 1t then transferred to glycine. This has now been proved. , This formation of arginine consists in an interaction of citrulline with either glutamic acip. or aspartic acid. Its speed is of the same order of magnitude as that of oxidative deamination in the kidney. The two dicarboxylicamino acids are equally effective in this respect. An oxidation is involved in the reaction; it is nearly completely inhibited by such oxidation inhibitors as KCN, ~203, and As 20 5 in low concentration. The KCN inhibition is padr relieved by hydrogen acceptors; that of As203 and As20 5 is not. There are i~ general three possible types of mechanis,m for this reaction, the essential differences in them consisting in the point at which the oxidation, i.e. _the dehydrogenation, occurs: (a) at the dica'rboxylicamino re6id e%rg it reacts with the citrulline, (b) at an intermediate comp-6\rff2P1 _oil i'sting of the citrulline and the dicarboxylicamin9 a2i\¥,l 0" {~~·5Jfter the cleavage of this hypothetical intermediate compound, 5.i''h, aerivative of the citrulline whicli is the immediate precursor of the arginine. f)·t * Aided by a grant from Mr. I. Zellerbach. A summary of this work has appeared (Proc. Am. Soc. Biol. Chem ., J. Biol. Chern., 140, p. xviii (1941)). 717 58 718 Conversion of CitruHine to Arginine The first of the three possibilities is depicted by the following equations, H (1, a) COOR. . C-COOH + oxidized hydrogen acceptor - I NH, (Aspartic acid or gl utamic acid) COOH. . . C-COOH + reduced hydrogen acceptor I NH (Dicarboxylicimino acid) .dJ 'j r{t a: ,,!)) .• 0 (1, b) I COOH .. . .. C-COOH + NH,-C- NH . . I ll NH , . COOR - (Ci trulli ne) 0 NH I I COOR . . ... C-COOH* + NH,-C . . .. COOH f (fl!; Ketodicarboxylic acid) (Arginine) * Here, a;s i'n Equations 2 to 4, the dicarboxylic acid is wrntten ii,s being converted to ' the corresponding keto acid derivative. This is done only for convenience. We have obtained no evidence regarding th.e: nature ' f the product arising from the deaminized . dicarboxylicamin8 ' ~cld. It i' possible that it is also decarboxylatecl in the oxidation; i.e.~''ttii? reacti6n mechanism is that suggested by Herbst (2) . Lil •J )ffj ,II, This mechanism was eliminated by the finding · tqaj;/r§Lr~i.nine formation ill inhibited under conditions in which oxidative deamination of the dicarboxylicamino acids proceeds up.chegk~d'. ,, Arginine is also formed from citrulline, ammonia, a1+d a-ketoglutarate or oxalacetate. These reactions are also iphibited by KCN or As203. The interpretation we have placed ?n these ,I~sults is that here also the reaction proper (i.e., arginine formation) is between citrulline and glutamic acid or aspartic acid; ,tJhs,t"the dicarbox-ylicamino acids were formed by reductive amination: prior to their combination with citrulline. It follows that the ''iiilino 59 H. Borsook and J . W. Dubnoff 719 acid derivatives of the dicarboxylicamino acids (which are formed spontaneously and are in equilibrium in water with their corresponding a -keto acids and ammonia) do not react with the citrulline. This evidence also therefore excludes the reaction mechanism of Equations 1, a and 1, b. The other two possible mechanisms are given by Equations 2, 3, and 4. H I COOH . . . C-COOH + COOH .. .. NH-C-Nllz (2, a) II I 0 NH2 H ,, I COOH. .... C-COOH -i I NH H I COOH . .. . . N-C-NH2 (2, b) I OH OH I +:± (2, c) COOH ..... C-COOH I NH H I . .N-CH-NH2 COOH . OH ·l) l, COOH. .... C.,..;-COOI-I + oxidi zed hydroge n acce ptor (3, a) I ll NH I j COOH ... . . NH-C:H---:-NH2 (3, b) -t ~ ,J COOH. ; .C- COOH + COOI-I. .... NH-0.---NH, + II • 0 T re;i duced hydrogen acceptor II NH 60 720 Conversion of CitrulJ.ine to ,.,Afgir1ine or OH I COOR ..... C-COOI-I -> (4, a) I NH I COOR .... ..,N H-CH-NH2 I .I (4, ,b) COOR ..... C-COOH NH2 I + COOR ... . . NH-CH-NH2 II 0 NH2 I (4, c) (4, d) COOR ..... NH-CH-NH, + oxidized hydrogen acceptor COOR . .NH -C-NH, + reduced hydrogen acceptor II NH In both mechanisms an intermediate compound between the citrulline and the dicarboxylic acid is postulated. According to the mechanism of Equations 3, a and 3, b this compound is decomposed by dehydrogenation into arginine and the a-keto acid derivative of the dicarboxylicamino acid. In the mechanism of Equations 4, a to 4, d the hypothetical intermediate compound is decomposed first and the resulting diamine group undergoes dehydrogenation to form arginine. If the mechanism of Equations 4, a to 4, cl were con ed, when citrulline and the dicarboxylicamino acid are present but arginine formation is prevented by As2O3, the keto acid should be formed nevertheless, but without an equivalent amount of ammonia. Ammonia and ke'to acid will be formed in addition.from the concurrent and independent oxidative deamination of the dicarboxylicamino acid. Accordingly the ratio t::,. NH3 to t::,. keto acid should be lower and the absolute amount of keto acid higher in solutions containing ' citrulline, dicarboxylicamino acid, and As2O 3than in the absence of citrulline; i.e., the reaction would proceed as far as Equation 4, b. It was found that the ratio t::,. NH3 to t::,. keto acid in the presence of As2O3 was the same with and without citrulline; nor was there an . absolute increase in keto"a1Cid in the presence of citrulline. 61 H. Borsook and J . Y.l . Dubnoff 721 The evidence appears to lead by exclusion to a mechanism of the type of Equations 2 and 3. Braunstein (3) postulated an intermediate compound analogous to that of Equation 2 as a preliminary to transamination . It must be emphasized that in both cases the existence of this hypothetical compound is, as yet, unsupported by any direct evidence. It is an inference only. Positive results were obtained when proline, hydroxyproline, ornithine, or lysine was used in place of aspartic acid or glutamic acid. All the other amino acids were negative. There is independent evidence that the first three amino acids named are readily convertible to g,uJamic acip, in the rat (4-7). It is reasonable to conclude, th(;)[rJ9re, that lysine also is converted to glutamic acid (or aspartic ~qi§),. Thi~ i the first clue, as far as we a1:e aware, regarding the path of degra~ii,tion of the carbon skeleton of lysine in the animal body. s. , In the ornithine cycle by which 1u,:e.a is formed in the liver from ammonia and carbon dioxide, one.i of"')~\ .the steps proposed is the conversion of citrulline to arginine (fil t:0 The mechanism for this con.version which we have found ini ij<,lJj.ey, i.e. the interaction of citrulline with aspart.ic acid or gh tfl,JP,ic acid, does not occur in liver. A study of arginine formation!iii' liver is in progre;s and will be reported later. Procedure ancl Preparations Usecl Most of the experiments were carried out with surviving rat kidney slices. In a few experiments guinea pig kidney wru ,µfed. The details of the technique have been described (9). 'J.', iR _?r three smal~ slices were suspended, in 4 ipl. of bicarbonate-:-tR ~1~1 ·'s solu1 t ion containing the substances under investigation.',£, ,l}e pH of the solution was 7.4 after equilibration with 95 per cent qxygen and 5 per cent CO2 . All the experiments were carried out at 38° for 1 or 2 hours. At the end of the experime:1tal period th~ _conten,1fi1 ~f t he reac1tion vessels, acidified to pH 6.0 w.1ith a drop of 0 ..5,.~1~C1,, the slices, J AN if\., .. and two 2.5 ml. washings with water were transfe~T~ )tp, test-tube, graduated at 10 ml. After 10 rsinutes in a b~-'.i.fm1y.rater bath tkr test-tubes were cooled, and the c9nteP:ts made \-PirRiNi1-<t mark with water, mixed by shaking, and filtered-. The t iss~e:- Jices were drie~ at 100° and weighed. ," 1, } ' 62 722 Conversion of Citrulline to Arginine Glycocyamine and arginine were then determined in the protein-free filtrates. A 5 ml. aliquot was passed through a column of permutit and the traces of glycocyamine remaining in the column removed with 5 ml. of 0.3 per cent NaCl. The glycocyamine was determined in the combined filtrate by the method we have described (10). The arginine was then eluted from the permutit and determined as described by Dubnoff. 1 Ammonia and keto acids were also determined in some experiments. These analyses were carried out on the protein-free filtrat~s which were not passed through permutit. The analytical :dibthod used for ammonia has been described (11). The keto acids we;·e determined as follows:. To 2 ml. of proteinfree filtrate ' 0'.3 ml. of a saturated solution of 2 ,4-dinitrophenylhydrazine in i N HCl was added, followed after standing for a few minutes py 1 ml. of 1 per cent NaOH. An intense, deep red color appears' immediately. This quickly fades and after 20 minutes a stable1fecl color remains wli~se intensity is proportional to the concentdtion of pyruvic aci'd, oxalacetic acid, or a-ketoglutaric acid present. A linear relationship is obtained between the concentration of'keto acid and the depth of color when it is measured at a wave-1e'ngtli of 0.525 µ, in 'the concentration range from O to. 2.0 mg. per cent. The color of higher concentrations is too deep to be measured accurately. \., ,y 1 . The use of 2 ,4-dinitrophel'.).ylhydrazine in alkaline solution for the colorimetric determination'. tifJpy ruvic and other keto aci<;ls has recently bJen reviewed by .Klefti: I2). The simple procedure we have employed, without preliminai-y extraction of the hydrazone, would givEJ positive values with sd6h substances as acetone and acetoacetic acid. Under our experimental conditions the latter two substances are not formed in measti1'able amounts. The controls (Table VII? showed that for our limited'purposes the simplified procedure was satisfactory. l( + )-Cit,;µ}line was used throughout. It w_as prepared by a comhination of the methods described by Kurtz (13) and by Gornall and Hunter '(14). The starting mater.ial was a cfonmercial preparation of Z( + )-ornithine. All the other 'amino acids used were prepared or purified either by Amino Acid Manufactures, University of California at Los Angeles, or in this laboratory. -We are in1 Dubnoff, J. W., J. Biol. Chern., 141, 711 (1941). 63 H. Borsool} and J. W. Dubnoff 723 debted to Professor H. B._,V\~kifyJ or a generous donation of glutamine. The naturally occ11,1trj11g isomers of the am,ino acids were used except when the dl form is specified. -;pyruvic acid was prepared by distillation of a commercial prod4 t _;, t}.i.y distillate was neutralized by NaOH, and the sodium salt crystallized and dried. a-Ketoglutaric acid was prepared by the method of Neuberg and Ri:il:~ ~_115). "\Ve are indebted to M r. Werner Baumgarten for this preparation. Ox;11lacetic acid was prepared by the method of Simon (16). The a-keto acid derivative of methionine was obtained by oxidation of dl:.methionine with d-amino acid oxidase prepared from hog kidney, ethereal extraction of the acid from the acidified, deproteinized soiution, evaporation of the ether, solu tion of the acid in water, neutrnlization with NaOH, and recrystallization from water. The acetaldehyde was a commercial preparation. Coenzyme I was prepared by th e method of Ohlmeyer (17), and . coenzyme II and adenosine triphosphate according to the prescription given by Warburg and Christian (18). The coenzyme pr<?parations, as used, were 60 per cent pme. These preparations we:re made by Dr. Norman IL Horowitz. R esults Table I is a typical protocol showing the formation of arginine from citrulline and glutamic acid or aspartic acid. There was always a significant increase in· arginine over the blank value when citrulliny ,,f;lone was added to the Ringer's solution. This arginine was for*P, presumably, from the added citrulline and aspartic ~cid or ~11tamic acid (see T able IV) which was p1;esent in the kidney sl:ices, ait the outset or formed in the slices during the experiment/ Nearly 3 times as much arginine was formed, however, whe~•'ei}\er aspartic acid or glutamic acid was added with the cit1:ul!,in/h,,, The two dicarboxylicamino a•cids w~re equally effective . . There was a small increase in glycocyamine whenever arginine fo~~tj}m occurred. It may be inferred from previous observatffi!Ri9 ~~tthat the amidine group of the arginine was transferred to · @Hui~ which was present in the slices initially or formed in the cl5urse of the e~periment. The qu antity of available glycine was ... •at t, f 64 724 Conversion of Citrulline to Arginine small ancl also limited because no more glycocyarriine was formed when the arginine was increased 3-fold. It was not feasible, because of the high solubility of arginine and its salts, to attempt to prove by isolating one of them that the substance we were measuring colorimetrically was arginine. This wa.'3 TABLB I Typical Protocol Showing Formation of Arginine by Sitrviving Rat Kidney Slices f rom Citntlline and Glutamic Acid or A sparlic Aci d Ringer's solution; 38°; 1 hour. Solu tion No . (!) · Tissue Citrulline, Olutamic Aspa rtic weight, acid, acid , 0.0025 " dry 0.005 M 0.005 M (2) (3) (4 ) (5) --- --,\ mg. - - - Arginine found (6) (7) 1ng. per cent g,n, fre sh tissue 0 .02 0.02 0.33 0 .31 0.04 0.04 1.07 0.89 0.03 0.04 0 .89 l'.1l2 0.3 0.1 1.8 1. 9 0.1 0.1 5.2 5.7 0 .1 0.1 5.8 5.4 --- --- mg. pl:T Glycocyam ine found (8) (9) mg. gm. fresh tissue - - - -- per cent mg. per 0.02 0.3 0.04 0.2 0 . 07 0 .4 + 0.07 0.4 + - I -;0.04 0.1 5 + 0.06 0 .2 6 + - ' 0.08 0.4 7 + + 0.07 0.4 8 + + 0.04 0.2 9 + -' 0.04 0.2 10 + 0.06 0 .4 11 + 0 :07 0.4 12 + .·, -r : The figures in 99htnins 6 and 8 give the concentrations ~f ,a.rgin.ine and of glycocyamine in the protein-free solutions after th ey had passed through fi fHfflt.1t. Th e fi gures rn • C o1umns 7 an d 9 ' are '· ·ob tame • d t 1rn co 1umns of •)permu by multiplying-Hi· ~ •ift Columns 6 and 8 respectively by the factors, 4/ 100 X 10/ 4 X 2 X 100@1/'(5 tX W) = 40/ W, where Wis the dry weight of tissu e in mg. The basis of these factors is as fo llows: .The volume of the original , undiluted :B,inger's solu tion was 4 ml.; it was diluted to 10 ml. for deproteiniz;i,tion; in the cour$e of sepa ration of the glycocyamine and elution from perrnutit the aliquot used was diluted with an equal volume of salt solution; the fresh weight of the sli ces is 5 times ~he dry weight. 1 2 3 4 3 .2 7.3 7 .2 6 .7 '11.7 ' 1 11.1 8 .3 6 .2 10 .3 11.9 6.1 7.5 :?' - : •, ' .;+ ;+ ~ T) :J. establi_sb,ed q,y tqr!f(:) i~depenclent pieces of evidence: the specificity of th,e color Fe~g~nt under the conditigns in whicµ we used it, the disappearance of this chromogenic n)J,¥~i;ial when subjected to the a~tioi;t· of argin~~,,<and the .formation, pf, ,glycocyamine in the highly sp;cific transamidination reaction with glycine . . 66 726 Conversion of Citrulline to Arginine different1 amino -?,c,ids and o( some other nitrogen compounds as con~rj;bl).t9i;s of the =NH group in the conversion of citrulline to argi:µ,ine, Glutamine was the only compound which was as effective as aspartic acid or glutamic acid. This may be taken to indicate the rapid hydrolysis of the added glutamine rather than that the amide itself is effective. An active glutaminase is present in TABLE II Proof from Action of Arginase Thal Arginine Was Main -Chromogen 1lfeasured by Colorimetric Method Employed The arginine values are measured in mg. per gm. of fresh tissue. Citrulline, 0.0025 M Aspartic acid, 0.005" Glutamic acid, 0.005 M - - - + - + + + - - + - + - + • Arginine present measured co lorimet,rically ~ Before action of arg inase After action of arginase 1. 4 3.3• 11.0 0.2 0.2 0 .6 0.5 0.2 0.2 s:s 1.2 1.0 Ringer'_s solution, 38°; 1.5 hours. Each of the six mixtures was carried through in quadruplicate to the end of the period of incubation with kidney sli ces; pairs of similar solutions were then combined. Each of the values given above therefore is an average of two completely separate but identical experiments run simultaneously from t he initi al incubation with kidney slices through. the 'digestion with arginase to the final colorimetric determination. The arginine was dete rmined before and after digestion of the protein-free filtrates w ith arginase . Dry arginase powder was prepared by the method of li).mter . and Dau phi nee (19) . 4 ml. of the protein-free filtrate were digested at pH 8.7 for 1 hour at 38° with 0.2 ml. of a 3 per cent soluti on of the arginase' powder. At the end of the hour the solutions were acidified to pH 6.0, boiled, and filtered. The arginine in the filtrates was then determined in the usu al ma nn er a ft er removal of the glycocyamine with permutit. kidney (20). Asparagine was much less activ,,e than aspartic acid (the asparaginase activity of kidney is very low), while acetamide, formamide, and nicotinamide were negative. The positive results with praline, hydroxyproline, and ornithine were to be expected. WeJl-r1"!?,lherbe and Krebs (4) al).d N eber (5) have proved that praline and hydroxyproline are converted to 67 H. Borsook and J. W. Dubnoff 727 glutamic acid by rat kidney slices. The figures in Table IV sh9w thaij the formation of arginine from citrulline was approximate! twice as fast with proline as with hydroxyproline. This is in accord with the findings of Weil-Malherbe and Krebs that glutamic acid is form ed twice as quickly from proline as from hydroxyproline. The conversion of l( + )-ornithine to glutamic acid was first suggested by Krebs (6) on the basis of indirect evidence. d( + )Proline and d( - )-ornithine gave the same oxidation prnduct, a-keto-8-aminovalerianic acid, after treatment with d-amino T A BLE III Glycine Trans ainidination with Arginine Formed froin Citrulline The results a re ex pressed as mg . per gm. of fresh tissue. Solution No. Metabolites used cyam1n c Glyco- Arginine "Total" arginine 1 2 3 4 5 6 Citrulline Aspartic acid Glutamic " Citrulline + as pa rtic acid + glu tamic " " 0 1.2 0.2 0 .2 2.6 2.7 0.4 2.0'' 0 .8 0 .6 5.9 6 .6 0.4 3.8 1.1 0 .9 9.8 10 .6 The different mixtures of meta bolites were firs t incubated at 38° with rat kidney slices for 2 hours. The citru lli ne , aspa rtic acid , and glutamic acid were initially 0.01 M. A 3 ml. ali quot of each mixture was then remove_d and added to 1 ml. of a cell-free extract of rat l;cid ney to, which glycine had bee n added to a concentration of 0.04 M. These mixtures -were then incubated at 38° with shaking for 6 hours. They were th en ·de.proteinized and analyzed for glycocyamine a nd a rginine. acid oxidase. l( - )-Proline was proved to be converted to glutamic acid. It was a reasonable surmise then that glutamic acid was formed also from l( + )-ornithine via a -keto-8-aininovalerianic acid in the same manner; viz., oxidation at the a-carbon atom to form the acid and amination of the a-carbon atom. Direct evidence of this conversion was obtained by Roloff, Ratner, and Schoenheimer (7) . These workers fed deutero ornithine to normal adult mice and lat er found not only deutero arginine but also deutero proline and deutero glutamic aqid. Data such as those in Table IV reveal that the conversion of l( + )-ornithine to glutamic acid occurs rapidly in rat kidney. 68 728 Conversion of Citrulline to Arginine In -view of the findings with proline, hydroxyproline, and ornithine it is a reasonable interpretation of the positive result with lysine (Table IV) that this amino acid also is converted to glutamic acid. 3 . !.lh:t d f, "cmo1 ai hb; T AB LE IV •r .911f Relative Effectiveness of Different Amino Acids and Other Nitrogen Compounds As Contributors of =NH Group in Con~ersion of Citruliine to Arginine Ringe r's solution; 38°; 1 hour. The citrnlline was ini t ially 0.0025 M; all other compounds 0.005 M. Except where t he dl form is specified, the na turally occurring isomer was used. The res ults a re expressed as per cent of "total" a rginine formed from cit rulline and glutamic aci d. " Total" arginine = a rginine+ glycocyamine X 1.49 . R ela tive speed of arg inine formation Substance Acetamide u Al anine Ammoni~ . Asparagin,e Aspartic acip Cysteine Forma mide Glutamic aci d Glutam ine Glutathion e Glycine Histidine Hydroxyproline -7 -8 ' 1 55 108 -4 - 13 100 101 46 1 0 23 Substance dl-Isoleucine Leucine Lys ine di-Methionine Nicotinic acid amide dl- N orleucine Ornithine di-Ph enylala nine Proli nc di-Serine Threo nine 't r Tryptophane ·,,,. Tyrosine Valine Relative speed of argmme format ion - 8 - 8 47 -8 -21 -18 42 -12 51 - 3 -14 -1 6 -6 5 Table V shows the inhibition of the interaction of citrulline and aspartic acid or glutamic acid by low concentrations of KCN, As203, or As20 5. This inhibition is abolit°'95 ,, per cent complete. 3 It may be ex pected from t heir structural similarity t hat the conversion of lysine to glutamic ac id fo llows a course a nalogous to that of ornithine. One pl:),thwaty for which t here is some ex perim_ental evidence is as follows: oxidat~ve deam ination of the a-amino group, follow ed by oxidative de,amination of the ,-amino group, f3 oxidation at· the ca rbon atom to form a -ketoglutaric acid, wh ich is reduced with t he addition of ammonia to glutamic acid. I .' 69 H . Borsook and J. W. Dubnoff 729 The cyanide inhibition is relieved to a considerable extent by the a-keto acid derivative of methionine, by oxalacetate, and by V TABLE Inhibition of A rginine Formation by Oxidation Inhibi tors Ringer's solution; 38°; 2 hours. The ini tial concJ1 tration of citrulline was 0.0025 M; of =NH donor 0.005 M. The results a re exp ressed as mg . of "total" arginine per gm. of fres h tissue . fl Uninhibited 0.001 M KCN o.901 n 1.3 2.7 12 .2 11 .2 0.9 1.1 1. 8 1. 7 0 '. 9 ' 1.1 t 1~1f 1.8 ,t.,. 1.8 Nd ·aimino acids added . . . . . . . . . . Citr !line. .. .. tt· + aspartic acid . . . . .. . + glutamic " ... .... .. " TABLE As,O, 0.0036 M As,O, ' 0 .8 1.1 1. 7 1. 7 VI Counteraction of Hydrogen Acceptors on Inhibi tion of Arginine Formation by KCN and As,0 3 Ringer's solution; 38°; 2 hours . The citrulline was ini t ially 0.0025 M; =NH donor /1;!1,Q. other metabolites 0.005 M. tt I)(, Aspa rtic acid = NH donor Unin- Glutnmic acid 0.001 M 0.001 " Unin- 0.001 " 0.001 M 1. 1 3. 1 12 0 0 .9 1.0 1.3 0.9 1.0 1.6 0 .6 ;) KCN As,O, hibitcd I<CN As,O, --- - - - - - - - - - - hi bited 1NJo amino acids added. . . . .. Ci t,rnlline alone . . . . ... . + = N H donor . . .. " + " " " + acetalclehyde. .. . . Citrulline + = NH donor+ a -keto acid deriva tive of methionine .. . . Citrullinc + = NH donor + oxalacetate. Citrulline + = NH donor+ pyruvate . . .... 1.1 3. 0 12 .2 0 .9 1.0 1.6 0 .9 1.1 1.0 0 .9 . 7.0 r.I 5.5 l I 7.2 3.1 1.1 3.1 1.2 6.7 l.· 4 .3 l 1. 9 ,r 2.2 1.3 pyruvate . These metabolites do not relieve t he inhibition by arsenite (Table VI). Our interpretation of these findings is that both cyanide and 70 730 Conversion of Citrulline to Arginine As2O3 (and As2O5) inhibit the formation of arginine by preventing the necessary dehydrogenation of the citrulline-aspartic acid (or glutamic acid) complex . The effect of cyanide is indirect; it prevents the oxidation of cytochrome. The flow of electron and of hydrogen transfer is thus blocked by the lack of an available acceptor. Arginine formation can be reinstituted, however, by the provision of electron and hydrogen acceptors other than the cytochrome-oxygen system which can function even in the presence of cyanide. They must be provided in sufficient concentration because their reoxidation is also blocked by cyanide and they cannot therefore function in a cyclic manner. The a-keto acid derivative of methionine, oxalacetic acid, and pyruvic acid served this purpose.. Acetaldehyde did not relieve the cyanide inhibition. This would exclude complex formation with the cyanide as an explanation of the positive results with the above keto acids. The arsenite inhibition, on the other hand, is not relieved by these metabolites because this inhibitor interferes with the process of dehydrogenation; i.e., the initial process of donation of electrons and hydrogens is blocked. Hence the provision of additional hydrogen acceptors affords no relief. In view of the participation of the pyridine coenzymes in the oxidative deamination of Z( + )-glutamic acid (21) and the evidence that an oxidation of aspartic acid or glutamic acid is involved in this formation of arginine, we tested the effect of coenzymes I and II and of adenylic acid pyrophosphate on the reaction. These substances. neither relieved the arsenite inhibition nor increased the rate in the absence of an oxidative inhibitor. The enzymatic activity of kidney slices is largely lost (about 90 per cent) when the kidney is homogenized into a cell-free suspension. One possible explanation for t his loss of activity was that the pyr"dine coenzymes were dissociated from the protein of the enzyme a result of the dilution consequent on the preparation of the cell-free suspension. If this explanation were correct, addition of one of the two pyridine coenzymes or of adenylic acid pyrophosphate might be expected to restore some of the lost enzymatic activity. This was not the case . It is possible that some other coenzyme is operative in this reaction. Another piece of evidence against the above explanation of the loss of enzymatic activity is that kidney tissue homogenized and as 71 H . Borsook and J. W. Dubnoff 731 tested without any added .fluid 1is no more active than when homo·genized an'cl s{1spended in 4 tiihes its volume of buffer solution . Table VII 'sh~ws: the sepa'ration of oxidative deamination of the dicarboxylicamino a!ci!dslffoci ·the process in which citrulline is converted to arginine. The latter reaction is inhibited by arsenite; the former is not. In fact ~n- inhibitor 'such as A.s20 3 which retards the disappearance of the products of deamirtation, the ammonia and keto acids, is necessary for the demonstration of oxidative deamination by tissue slices . .r, TABLE ·, VII 'l. . J~ nt~\ .... Separation of Conversion of Citrulline._ tR.t1-rgi_~ ·11,e fi;om Oxidative Deamination: ~I ,,.. , ,~,. •rrc111,,• • Guinea pig kidney slices were used· inger's· solution,; 38°; 2 hours . The citrulline was ini tially 0.0025 M; a,sp~;r,t~c acid either 0.005 or 0.05 M as indicated; As,O, 0.001 M. The res ults a re expressed as micromoles pe-r gm. of fresh tissue . Citrulline Aspa rtic acid, initial mola lity - + + + + + ·+ ; ,, - As,O, Arginine found - 0.9 1.0 1. 9 0 .9 12.1 6 .3 1. 7 1.3 1.0 1.2 1.0 1.0 + - 0 .005 0 .05 0.005 0 .05 0 .005 0 .05 0.005 0.05 + - + + - , + + K eto acid as py ru vic acid 0 0 0 0 0 1. 8 13.5 36.9 0 0 ~ ~, 10 .4 '.( I 35 .1 Ammonia 4. 3 10.4 7 .1 10 .9 8 .7 13 .6 . 25.5 54 .2 9.0 20.2 23. 7 54. 1 6 ammon ia t!. pyruvic acid 1.6 1. 4 1.9 1.4 Another piece of evidence which indicated that the two oxidative processes, arginine formation and deamination, are different, although they involve the same substrate (aspartic acid or glutamic acid), was obtained from a comparison of the effects of two different concentrations of aspartic acid, 0.005 and 0.05 M . The higher concentration of aspartic acid inhibited arginine formation, but the rate of deamination, in an absolute sense, was more than 3 times greater. If the reaction consisted in an exchange of the oxygen on the 1:-carbon atom of citrulline for the =NH group of 72 73f Conversion of Citrulline to Arginine the imino acid deri\;ative of aspartic or glutamic acid, the reaction should proceed with oxalacetic acid and ammonia or ketoglutaric acid and ammonia, even in the presence of cyanide, because the imino acids are formed spontaneously from the latter keto acids and ammonia. Table VIII shows that arginine is formed from citrulline and the products of deamination of the dicarboxylicamino acids under normal conditions; but the reaction is completely inhibited by cyanide. TABLE VIII Aspartic Acid 01· Glutamic Acid Formation i n Rat Kidney Slices; Attested to by Formation of Arginine from Citrnlline Ringer's solu tion; 38°; 2 hou rs. The ini t ial concentration of citrulline was 0.0025 Mj of glu tam ic acid or other metabolites a nd ammonia 0.005 M. The resul ts a re expressed as "total" arginin e formed per gm. of fresh 1 , ,I' , "' t issue . • Metabo lites in addition to eitrulline "Total" arg inine found I(' None .. Citrate + ammonia. Fumarate + " Glucose+ " Glutamate. a-Ketoglutarate + a mmoni a. Oxalacetatc + a mmoni a. Pyruvate + a mmoni a .. Succinatc + 2.4 2.9 4.8 I 3.7 10 .5 7 .9 4.8 2.6 4.5 Three independent lines of evidence concur , therefore, in excluding prior oxidative deamination of t he aspartic acid or glutamic acid as part of the reaction mechanism in the formation of arginine from citrulline. As stated above, another possible reaction mechanism would locate the oxidation on a derivative of the citrulline after it had updergone transamination with the dicarboxylic acid to form a CQmpound of the type 73 H. Borsook and J . W . Dubnof:f 733 Dehydrogenation of the diamine group would then yield arginine. If this were the reacti9n mechanism, the addition of citrulline in the presence of As 20 3 and aspartic acid or glutamic acid woul~, as discussed above, decrease the ratio t::.. ammonia to t::.. keto acid, because of an absolute increase in the keto acids, without a corresrmnding increase in ammonia. The data in ~able VII, eliminate this hypothesis. In the presence of As20a the absolute amount of keto acids and the ratio of t::.. ammonia to t::.. keto acid were the same in the presence as i_n ,the absence of ci.trulline. The evidence, b.)lj)l:)Xclusion, therefore points to the locus of the oxidation on a hypothetical intermediate compound of citrulline and the dicarboxylicamino acid. The reason that the ratio t::.. ammonia tot::.. pyruvic acid is greater than 1 is that As 20 3 does not completely inhibit the oxidatiop. of keto acids. ': Even in unpoisoned tissue the removal of ammo~ia.i,s less complete. than that of th~ . keto acids. Simil~r ob,s.e,rvati~·~s were made by Krebs (6). The formation- of arginine from citrulline by surviving kidney slices can be used as a test for potential sources"'of aspartic acid or glutamic acid . It was shown by this method that certain amino acids are converted in the kidney to one or the other of Jli~· e two dicarboxy licamino acids (Table IV). The action of tissue sli'ces in this respect parallels closely that-of the whole animal. 'Fln~arallel was also seen in the synthesis of hippuric acid (22) and/ &1creatine (23). The same method. ·a • be used as a test for pi:e£iirsors ,,H01 1 . of the carbon skeleton of the ~ic!l,rboxylicamin~, acids. , TJ!e reagents are citrulline, an excess of ammonia, and the metabolite in question. Some examples are shown in Table.VIII. Fulm:arate, glucose, a -ketoglutarate, oxalacetate,-and uccinate were definitely positive. A dubiously slight positif<?i. ,J!',e,sult was_ obtained with citrate. Pyruv,§l;te was ,neg~Jjye... ·)b1 q 1,, The mechanisms by which these non-nitrogenous metabolites may be converted to tb:e W91}~9I ililii1eJ1.Ii1i».<t,{:;t.cids are well known and need not be discussed ,her~•'1f'•)l-li§..:dq,t~1):µ TablE:J VIII are positive evidence that these changes o~m-1-r ~a,.dily in the kidney. The citric -acid cycle appe~s ,to .by qµanJ!iiJi31,-ti':'ely less important than simple oxidation ofi~~w;i.,nic acid ·a%ihits derivatives. • Tl!e. formation _of arginiit~·.n}ay be used a,s an indicator to ascertain the "preferred" directio!).s of some of the reversible prpGesses 74 734 Conversion of Citrulline to Arginine in the dynamic steady state of metabolic reactions in kidney slices. This test is based on the fact that, in the presence of an excess of citrulline, the formation of arginine is faster the greater the concentration of aspartic acid or glu tamic acid (at concentrations below 0.005 M). The following are two examples : The amount of arginine formed from citrulline in the presence of different amino acids a:nd a-ketoglutarate is an indication of the speed of transamination in the direction of glutamic acid formation; the difference in the amount of arginine formed from citrulline, a-ketogil!ltarate, and ammonia in the presence and in the absence of 'fl[ 'J11.' T A BLE IX ntss1~ ,,; Transamination in Rat K idney Slices 10 OJ h • ·, 1 t· '' 38° ; 2 h ours. Tl1e 1mtia • • • I concentrat10n • of c1•t ru 11·me . .1.1,mger s so ti 10n; ",{,lfiff6!.0025 M; of- 'otlier amino aci'd.s, keto acids, and ammonia 0.005 M. ',ttel results ai'e e;pressed as "total" a rginine formed per gm. of' fresh tissue. ;-.fi Meta boli tes in addition to citrulline "Total" arg nine Noii.e.' . . Glutamate . .. . ........... . . . . . .. . + pyruvate ....... . .. . . a-Ketoglutarate . . . ... , '· ; .. 0 . Ammonia..... . ..... . .. .. ... .. . . . . . . . .. . .. . .. . . . . .. . . a-Ketoglutarate + ammonia . . .......... . . " + pyruvate '1+ ammoni a .... . . . .. .. . . . . . . Alanine + a-ketoglutarate. di-Alanine + a-ketoglutarate .. . Valine + a-ketoglutafate. 2 .0 10.9 7 .8 3.1 2.2 4.6 found • 1. 9 2.8 2.1 1.6 pyruvate indicates whether ammonia nitrogen available for amino acid synthesis is bound preferentially as glutamic acid or as alani1ie. T able IX is a summa;ry of some· experiments along these lines. It is seen that glutamic acid contributes its amino group to pyruvate. On the other hand, amino nitrogen does not pass readily from l( + )-alanine, dl-alanine, and l( + )-valine to a-ketoglutarate. In this respect ammonia is more effective. We may infer then that the formation of glutamic acid from a-ketoglutarate is faster by reductive amination than by transamination. 75 H. Borsook and J. W. Dubnoff 735 Similarly, in the presence of ammonia, a -ketoglutarate, ancl-rJ}y: ruvate there was no evidence of the formation of any glutantic acid available for arginine formation; presumably the nitrogen•w'Ns tfound preferentially as alanine. " The clynaillic steady state in kidney slices appears therefore to include'a cYcle which favors th e formation of glutamic acid by reliluctive' a'mination, followed, if suitable keto acids are available, by ,t,fansamihation: toform other amino acids with the regeneration of the a-ketoglutarate". • T AB LE .: 1-i • X I tth-ibi tion)?'!J f<yruvcite ,o,t;. Arginine fi'o rmcition frqm Citn4lf,fteH, Ringer 's solu ti o,11i;: 3~~; 1 hou):. The ini tial conceq~rlj,tion qf- 05tr.u.Jline was 0.0025 M; 1oI •',othe~ amino• 1 ,._c· .T rls and pyruvate 0.0051, fM, ' ~) l. , 1 The results are -expressed as "t,otal" argin"ne formed per gm. o'f:' resh tissue. l'.f.lrI ' ' '<t, 1 o· Metaboli~es 1./-dded in asfrti,bJ.on _to citrulline I \ • J, None . .. . . ... .. ..... . Aspartate . . ....... . + pyruvate .. . Glutamate. ', + pyruvate . .. . " Lysine . . ..... + pyruvate .. ·,; . . . . . Ornithine.. . . . .... . .. . " .+ pyruvate .. . ·s· + pyruvate . . Praline. ~\3H . 5 -~TIB l 2.8 4-_ 1~iha 2.3~:i)~) . ... .. . , ? ?·; sri j , •· · · · il.-¥-10i 3.7 . 2. 4 ,: 2 .8!' 1 .'l ' The situation in' kid~ey slices appelus to be s1m.l'lar to that in the whole animal. Schoenheimer and his colleagues (24) observed tnat the most extensive exchange of normal 'nitrogen for N 15 -8ccuh·ed in the nj.trogen of aspartic. c'i'd and glutamiq aci<l whether t'hi ·N15 was in.troduced as ammonia or as· a single aminlflacid. O Thd' formb.tron of arginine from citruilme was inhibitelFby PY~ () . , ruvate wliether ·the donor of the =NH group was aspartic acid, glutamic acid, or a precursor of these dicarboxylicamino acids _(Table X),. This is in accord with the findings in Table IX t).iat pxruvate, is convei:ted to ,alanine at the expense of glutamic acid. -It'is' i:nterest'irig :tKat pyruvate is inhibitory under the normal cir- . 0 - 76 736 Conversion of Citrulline to Arginine cumstances of the experiment of Table X; yet in the presence of cyanide it permits arginine formation to occur, neutralizing to some extent the cyanide inhibition (Table VI). This latter fact suggests strongly that pyruvate does not inhibit the oxidation necessary for the formation of arginine from citrulline, that it is inhibitory under normal circumstances because ~t competes effectively with the citrulline for the .amino group of aspartic acjd or glutamic acid. More direct evidence is required, of course, to determine whether or not this hypothesis is correct.' DISCUSSION The work of Braunstein and Kritzmann (3, 25), of von Euler et al. (21), and of Cohen (26) established by indirect but strong evidence the ceJ1tral position of the dicarboxylicamino acids in the continual and rapid interchange of amino nitrogen which Schoenheimer and his cowork.ers demonstrated to occur in the body by direct evidence (24). 4 The mechanism of this interchange involves transamination and reductive amination. Another mechanism is involved in the formation (in the kidney) of arginine from citrulline in which the dicarboxylicamil).o acids also participate. Inst(lad of transamination, the reaction might be designated as transimination, except that the dehydrogenation of the amino group· of the dicarboxylicamino acid does· not occur prior to its reaction with the citrulline but while it is in combination with it. The discovery of an extremely active mechanism for converting citrulline to arginine in the kidney suggests a hitherto unsuspected source of the arginine which the kidney requires for, among other purposes, ~he transamidination reaction with glycine to form glycocyami,ne. It seem,s i,mprobable that a mechanism which can transform citrulline to .:1,i;ginine so quickly and act upon low concentrations of metabolites, less than 0.001 M, :fs inoperative in vivo, teleological as the argu~e~t. is. The question then. arises, what is the source of the citrulline? , The kidney cannot. cpnvert ornithine to citrulline. I ')'. 4 ~ It is interesting in retrospect that this interchange of amino nitrogen could have been inferred from the fact that ammonia exerts considerable nitrogen sparing action even in man. The continual_ and extensive synthesis and breakdown of protein was deduced from this and other data (27). 77 . H.. "'·Borsook and J. W. Dubnoff 73.7 ,. If one accepts the ornithi11~,urea cycle in ~he !iv.er as proposed .leY Krebs and Henseleit (8), the.Jiver.can hardly supply the citrullince, because in that cycle the existence of the citrulline is only transitory on its way to arginine. ( SUMMARY rG 1101 Citrulline is converted to arginine at a rapid rate by rntJand guinea pig surviving kidney slices, This property is almost completely lost when the cell structure is destroyed. ' -~ • r " 2.. Either aspartic acid or glutamic acid ip necessary for th.is ":te. . . . . J) , .'I(' ; . ! 1 act10n (m add1t10n to c1trullme). Jlifi , •. -.. 3. Proline, hydrnxyproline, lysine, and Qr\Zlithine may repla:ce the dicarboxylicamino acids in this reactiow.¥k Evidence is a:4iduced that they do so by .being ·converted firsu!vcl·igft'ltamic acid t{o'tf pos-_ . ', . ., \ ,, . "d) imrlqo.st• 1 . -n' wd :Jl s1bly bu~l~ss ~1kel:y, to as]ilart~c\)ac1_ . ,, ;1'";1-,,·,, , , . cf8 1:i1 . : 4. Arg1mue 1s f~qp.ed fro:n c1trultme anqfl;-k;~toglutanc,~~S\-3:llil ammonia or oxalacetic acid and ammonia. Evidence is presented that these products .0f oxidative deaclmatidnCafol.!reduced. by, fil;. ney slices o forip tRe parent dicarbox~cami '?>' acid§>. ~ 1 .&: rp' '• f ' '· h of argm1p.e . • rfrop-l, . ttt>-1·c;,,~u it 11·me r 1.s near1y; i), , ,a !,(• . 5 .,,<".·nw ' or. at10n\ co:m:pletel~ inhfil1ted by oxidatfve inhibitg{s, KCN, As203, 'A~i\, indicatiQ,g-&ttn ruudative stepis ·n tne ,.xeM.-tiu,n n1,enhanv,m. oTle cyanide inhibition is relieved in part by the a-keto acid tle1.,,ivative ,of m'ethio'n-in'e-;·'by 'oxala: etat , ,ali.dr:,b y'.'i)yr t,,ate. The inliibiti6h by As20 3 and As20 5 is not relieved by these metabolites. An interpretation of these findings is presented. 6. The oxidative step is not a dehydrogenation of the amino group of the dicarboxylicamino acid (to form the imino •group) prior to its reaction with citrulline. Evidence is _presented that this oxidation may be located at a hypothetical intermediate compound of citrulline an'd the dicarboxylicamino acid. 7. Arginine formation from citrulline can be used as an indicator of the "preferred" direction of some of the reversible processes in the metabolism of the cells. Some examples are presented. a: BIBLIOGRAPHY 1. Borsook, H., and Dubnoff, J . W., J. Biol. Chem., 138, 389 (1941). 2. Herbst, R. M., J. Am. Chem. Soc., 58, 2239 (1936). 3. Braunstein, A. E. , Enzymologi a, 7, 25 (1939). 4. Weil-Malberbe, H., and Krebs, H. A., Biochem. J. , 29, 2077 (1935). 78 738 Conversion of Citrulline to Arginine 5. Neber, M., Z . physiol. Chem ., 240, 71 (1936) . 6. Krebs, H. A., Enzymologia, 7, 53 (1939). 7. Roloff, M., Ratner, S., and Schoenheimer, R ., J . Biol. Chern., 136, . 561 (1940) . 8. Krebs, H . A., and Henseleit, K., Z . physiol. Chem., 210, 33 (1932). 9. Borsook, H., and Dubnoff, J. W., J. Biol . Chem., 132,307 (1940). 10. Dubnoff, J. W., and Borsook, H., J. Biol. Chem., 138,381 (1941) . 11. Borsook, H., J . Biol . Chem ., 110,481 (1935). Borsook, H ., and Dubnoff, J. W., J. Biol. Chem., 131 , 163 (1939). '· 1_, 12. Klein, D., J. Biol. Chem., 137 , 311 (1941). 13. Kurtz, A. C., J . Biol . Chem., 122 , 477 (1937- 38) . 14. Gornall, A.G ., and Hunter, A., Biochem . J ., 33 , 170 (1939). 15. Neuberg, C., and Ringer, M ., Biochem. Z ., 71 ,228 (1915). • • 16. Simon, L. S., Compt. rend. Acad., 137 , 855 (1903). ~: 't 17. Ohlmeyer, P ., Biochem. Z., 297 , 67 (1938). ,.l 18. Warburg, 0., and Christian, W., Biochem . Z., 287, 291 (19~6). 19. Hunter, A., and D auphinee, J . A., J. Biol. Chem., 86 ,627 (1929- 30). 20. Krebs, H. A., Biochem. J., 29 , 1951 (1935) . 21. von Euler, H ., Adler, E., Gunther, G., and Das, N. B'., Z. physiol. Chem., 264, 61 (1938) . ~· 22. Borsook, H ., and Dubnoff, J. W., J. Biol. Chem., 132,307. (1940). 23 . Borsook, H ., and D.ubnoff, J. W., J. Biol. Chem. , 132 ,559 t1940). 24. Schoenheimer, R ., and Rittenberg, D., Physiol. Rev., 120, 218 (1940). 25. Braunstein, A. E., and Kritzmann, M. G., Enzym~loiAa, 2, 129, 138 (1937) . ~ • 26 . Cohen, P . P ., Biochem. J ., 33 , 1478 (1939); J . Biol . Chem., 136, 565 (1940) . 27. Borsook, H., and Keighley, G. L., Proc. Roy. Soc. London, Series B, 118, 488 (1935) . • II CORNEAL VASCULA.RITY AND ITS RELATION TO ARIBOFLAVINOSIS 79 It is generally agreed that between the clinically recognizable vitamin deficiency states and the nonnal, there must exist wild or latent deficiency states. This finds practical expression in the wide difference between the so called minimum protective and optimum level of vitamin intake. According to Kruse ( 1) ,., the specific pathological process in a deficiency disease is characterized by velocity, intensity, and sequence. An acute process is rapid in its course and also rapid in responding to therapy; a chronic process is slow in its onset, progress, and response to therapy. Both the acute and chronic states may be divided into mild and severe degrees, but with suitable techniques one should be able to show a continuous gradation from normal to markedly pathological states. In ariboflavinosis this gradation has been said to be measured by the extent of corneal vascularization. A vascularizing keratitis of the cornea in vitamin G deficient animals was first reported in 1931 by Day, Langston, and 0'Brien(2). Later in 1939 Eckardt and Johnson (3) noted that all rats on a riboflavin deficient diet which survived to the 48th day shovted either keratitis or vascularization of the cornea. In the same year Bessey and Wolbach (4) produced a vascularizing keratitis in animals deprived of riboflavin which could be cured by supplying the vitamin in the diet. This manifestation preceded all other demonstrable lesions due to riboflavin deficiency. The first report on corneal vascularity in humans receiving insufficient riboflavin was presented in 1940 by Kruse, Sydenstricker, Sebrell, and Cleckley (5). All nine patients in their study had frank symptoms of riboflavin 80 deficiency. Nine had cheilosis, (fissures at the angles of the mouth), 8 had a characteristtc glossitis, and 3 had seborrheic accumulation at the nasola~ial folds. Examination with the biomicroscope revealed' that the earliest change was a superficial invasion of the cornea by capillaries from the anterior ciliary vessels at the nasal and temporal sides of the cornea. At a latter stage the capillaries invaded the sub- stantia propria.,and superficial and interstitial infiltration appeared. There was no regression on feeding niacin, thia;min, oevitamic acid, codliver oil, or crystalline vitamin A. However, riboflavin resulted in a decrease in size and activity of the vessels. Later in the same year this study was supplemented by a report (6) on 47 patients, all of whom were known to be riboflavin deficient. Besides the usual •ubjective symptoms of itching, burning, roughness, photophobia, and poor vision, 45 patients showed a circumcorneal injection due to congestion and proliferation of the limbic plexus. Thirty-seven patients showed actual invasion of the cornea by the capillaries arising from the limbic plexus. This vascularization developed first just beneath the epithelium, later at varying depths in the substantia propria, and finally just within the endothelium. Increased pigmentation of the iris which disappeared on treatment was seen in 19 cases. In most cases the congestion disappear- ed within 48 hours after treatment with daily doses of 5 to 15 mg. of riboflavin. If the vasoularization was more extensive, 5 to 18 days were required to empty the vessels. The authors stated that "aribo- flavinosis is possibly the most prevalent, apparently uncomplicated avitaminosis; it is possible also that it is more easily recognized than others on account of the specific lesions of the eye which occur 81 early in the period of deficiency.n Thus, by means of the slit lamp and biomicroscope all stages of the deficiency from mild to acute could be recognized. On the basis of this work the Conference on Methods and Procedure for Nutrition Survey (7) recommended the biomicroscopic examination of the cornea for the evaluation of the extent of riboflavin deficiency. This procedure was accordingly employed in the nutrition survey at the Lockheed Aircraft Corporation, Burbank, California. has already appeared (8). A preliminary report This paper is the final report of the applica- tion of the biomicroscope to the determination of riboflavin deficiency. The subjects of this survey were 1,173 white male volunteers from all the major departments of the plant. into two groups: ment of These were divided at random an experimental group which received a vi ta.min supple- Vitamin A (from fish liver oil) 50,000 r.v. Vitamin D (from fish liver oil) 800 r,v. Vitamin Bl (synthetic) 10 Mg. Riboflavin (synthetic) 10 Mg. Niacinamid (synthetic) 100 Mg. Ascorbic Acid (synthetic) 250 Mg. Calcium (Ca co 3 ) 500 Mg. and a control group which received the corresponding placebos. The initial examinations began in November, 1941 and ended in February, 1942. The final examinations on the 617 subjects remaining after one year began December, 1942 and ended April, 1943. The appearance of the limbus under the biomicroscope is shown graphically in figure 1. The fine capillaries which lie just under the FIGURE 1 0 limbus b two arcades C trunk ( concentric collateral d e streame r streamer t yp e arcade f white streamer 82 epithelium arise from the apices of the marginal loops in the conjunctiva and anas"t:4,mose to form a series of loops extending towards the iris. According to Kruse et al (5) there is no vascularity within the normal lirnbus. For the purpose of recording each sign was located according to the hours of the olook. Distances were measured with a micrometer eyepiece. In the following tables the effect of therapy on a number of signs which have been said to be indicative of ariboflavinosis is recorded. In most instances only enough consecutive cases were analyzed to give a statistically reliable result. If the extent of capillary invasion of the cornea is measured by totaling the number of hours occupied by capillary loops, the distribution shown in table I is obtained. There is no significant difference in the mean or in the distribution between the experimental and the control group. All subjects had some corneal vascularity, and the extent was marked in most oases. Table II shows the distribution of the maximum number of arcades in any hour within the cornea. There is a slight decrease in the mean between the first and second examination, but the change is in the same direction in both groups. This may have been due to a difference in grading in the two examina ti. ons or to unknown factors other than therapy, but the difference is not statistically significant. The division of the cornea into segments shows no difference between the two groups in the maximum nurnber of arcades (table III) or the extent of invasion to be found in a given sector (table IV). It has been claimed (5, 9) that the streamer type of arcades are particularly indicative of ariboflavin- 83 osis. These are active capillaries which extend some distance directly towards the iris and then loop back (figure 1). Often the arterial limb is much smaller than the venous limb, and the latter may be mistaken for a free ending capillary. The results recorded in table V sho.v that the mean of the total number of hours of arcades of streamer type of the experimental group is not lower than that of the control, nor is the extension towards the iris less (table VI). If more than one arcade is present, there is no difference in the length of the last arcade between the two groups (table VII). There is no difference between the two groups in the percentage of the total number of hours that are of streamer type (table VIII). Although we found no evidence for the existence of streamers in the sense of free ending capillaries, those capillaries which were not obviously loops were classified as streamers following the procedure of Kruse et al (5) although closer inspection always showed a fine connection between afferent and efferent vessels. The distribution of these so-called streamers is shown in table IX. In the course of the second examination a type of vessel was observed which was designated as a white streamer. These streamers, just under the epithelium are sometimes surrounded by a narrow whitish cone whose base is at the limbus and from whose apex the capillary extends as far as tre iris in a very fine loop (figure 1). seen under high magnification. Corpuscles are often The cloudy infiltration around the base of the capillary suggested a pathology that might be related to the conjunctiva! cloudiness of vitamin A deficiency. However, vitamin therapy did not decrease the incidence of these streamers (table X). 84 About 60% of the subjects were observed to have large trunk vessels (figure 1) for which the term concentric collaterals has been proposed by Vail and Ascher ( 10). According to these authors nconcentric collat- erals are engorged parts of the pre-existent limbal meshwork. Their function is to collect blood from all limbal loops of one entire limbal sector, or a part of them, and to return -this blood from the lirnbus to the larger conjunctival veins". There is no indication in our data (table XI) that the engorged condition leading to the establishment of concentric collaterals is in any way related to vitamin deficiency. the conclusions of Vail and Ascher. This confirms In some subjects a network of pig- mented deposits resembling the remains of capillary arcades were observed within the cornea particularly between the five to seven o'clock position. Ten cases were found among the controls and only three in the experimental group out of 617 cases, but the number ot cases is too small to be significant. The finding that corneal vascularity decreased with age (table XII) is just the reverse of what might be expected if corneal vasculari ty were indicative of vitamin deficiency, for a symptom of a chronic lack would be expected to accumulate with age. This trend has been confirmed in an independent survey conducted among a group of employees of Los Angeles County (table XIII). These results .an in direct conflict with the data presented by Scarborough ( 11) showing that vascularity was greater in patients over 50 years of age. Vail and Ascher (10) have pointed out that there is a normal pericorneal plexus of vascular arcades whose final loops under slit lamp illumination seem to be in the cornea but which are actually in the conjunctivoscleral wedge. This is more clearly 85 seen in figures 2 and 3. The overlapping of the cornea by the sclera creates a semi-opaque transition zone which makes the location of the true limbus difficult. Vascularization in this corneo-scleral trans- ition zone although actually not within the cornea proper has been considered pathological by Kruse et al. and has been used as a basis for other nutrition surveys and our own. However., the data in this study are still valid, for if a difference in corneal vascularity has existed, a different distribution would have been found between the experimental and the control groups regardless of the point of reference. For ex- ample, if the measurements in table ' IV are decreased by l mm~which is the average width of the transition zone, there is still no difference in the distribution of arcades longer than l mm. between the tv-10 groups. Similarly in table II the number of arcades is· . decreased if the true limbus • . 1,9. . used as the reference point, but the distribution remains the same. Retro ill LJ.tr :Ln8t i.on D1.rect Illuminstion Fi g ure 2 wi th Ch ang e i n Il l umin a tion direct il l rnr,inat ion limb us retro illumination limbus FIG Um~ 3 I nfluence of Type of Lighting on the Po sition of the Limbus 86 DISCUSSION In any examination for mild vitamin deficiency the very lack of corroborative symptoms places a greater demand on the specificity of the method employed. In the original study of Kruse et al (5) all the subjects presented a definite a.riboflavinosis syndrome. Since corneal vasculari ty was found in these subjects and regressed on treatment with riboflavin, it was tacitly assumed that all corneal vasculari ty was due to riboflavin deficiency. not This conclusion was.borne out by subsequent investigation. Many other conditions have been described that give rise to the same sign. Bessey and Wolback (4) have reported that vitamin A deficient rats show a vascularization of the cornea, and a similar condition has been reported in tryptophane deficient rats by Totter and Day (12) and by Albanese and·Buxchke (13). Corneal lesions have been observed in zino deficiency (14) and in sodium deficiency.(14). They have been observed in poisoning by thallium and on exposure to chemical vapor (9). Sandstead (15) believes that capillary invasion of the cornea may be more comm.only the result of exanthenatous diseases as measles. On the other hand, Mc- Creary, Nicholl~ and Tisdall (16) believe that ordinary irritants found in every day life - glare, smoke, dust, eyestrain - may be sufficient to maintain a state of vascularity. These authors present photographs which dramatically illustrate the marked engorgment of previously invisible vessels five minutes after irritation by a. weak soap solution. Gregory found that the injection of 200 mgs. of nicotinic acid ca.used a marked circurncorneal injection. The findings of Youmans (17) that vascularity 87 decreased in ·winter and increased in summer and that negroes show one tenth the vascularity of whites• point to the possible influence of' light on vascularization. Confirming this lack of specificity is the absence of improvement on therapy which in our present state of knowledge is still the best criterion for determining whether a given sign is typical of a vitamin deficiency. Sandstead found a 80-90% deficiency in 336 persons as judged by the slit lamp technique, but in a controlled experiment on 17 subjects no change was observed on giving 15 mgs. of riboflavin daily for sixty days. 34% of a group of 204 patients studied by Scarborough showed evidence of corneal vascularity. However, 8 patients who were given riboflavin exhibited no change after one month of this therapy. In a study on 41 individuals, McCreary, Nicholls, and Tisdall (16) found no consistent changes in corneal vascularity in either the placebo group or the group given 9 .9 mgs. of riboflavin daily for three months. Cn the other hand a decrease in corneal vascularity on riboflavin therapy has been reported only where subjects showing a definite ariboflavinosis syndrome have been selected for study (5, 6, 11, 18). In these cases regression occurred within a few weeks. It may be true as Kruse has stated that the reversal of a chronic deficiency state is a very slmv process, but if the time required to significa.ntl:f change the degree of corneal vasculari ty is beyond the period of one year of therapy employed in this survey, it is doubtful whether that change could be related only to the therapy employeq especiall;y in view of the lack of specificity of this sign. At least a slight trend should have been apparent if there were any relation between 88 00:aunon corneal vascularity and ariboflavinosis, or if only a fraction of the cases were due to ariboflavinosis • The data presented shQV no change Even whatever. ~if all corneal vascularity had been brought about by a previous riboflavin deficiency and had remained as scars, this sign would be of no value in the diagnosis of an existing deficiency. In such a case a better relationship might then be established between corneal vascularity and diet, but again no correlation has been found (10, 15, 19, 20). Some workers (10, 21) believe that there is never any true proliferation of capillaries except in corneal disease, and that the only changes that can be observed are the emptying or filling of preformed vessels with some increase in size as a result of engorgment. Changes of this type can hardly be expected to reflect a specific pathology but rather a non specific irritation or inflama.tory condition of which ariboflavinosis is but one case. Conditions that may bring about such a hyperemia have already been presented. The suggestion that everyday irritants as glare, smoke, dust, and eyestrain ~..ay maintain a constant state of engorgment of normally present capillaries is more in keeping with the widespread occurrence of this state among normal populations receiving adequate ·amounts of riboflavin. Norrr~lly the respiratory gases of the nonnally a.vascular cornea move by simple diffusion. The vascularizati on found in ri bofle.vin deficiency has been explained (3, 4) as a response to asphxia brought about by the lack of the oxidative catalyst, riboflavin. That this explanation may not be tenable, even in those cases of vascularity due to ariboflavinosi~ is suggested by the work of Philpot and Pirie (22). These workers analyzed the different parts of the eye and found the following average 90 CONCLUSIONS Corneal vascularity cannot be considered of any value for the diagnosis of mild riboflavin deficiency since (1) It is far too extensive in a large population receiving a relatively adequate diet. (2) There is no correlation between dietary intake and degree of vascularity. (3) There is no correlation with other symptoms characteristic of riboflavin deficiency. (4) There is no change on therapy in unselected groups. (5) The decrease with age is not indicative of a cumulative chronic process. (6) It may arise from many other causes. 91 SUivlMARY Six hundred and seventeen white male workers at the Lockheed Aircraft Corporation were exaffiined for corneal vascularity with ihe biomicroscope and slit lamp. There was no difference in the degree of corneal va.scularity between the control subjects and those who had received 10 mg. of riboflavin and other vitamins daily for one year. It was concluded from this and other work that ordinary corneal vascularity is not necessarily due to riboflavin deficiency and is of no value as a diagnostic sign. 95 TABLE I TOTAL NUMBER OF HOURS OF ARCADES (both eyes) : Ne- of ' hours 0 l 2 3 4 5 FIRST EXAMINATION l!ixperimenta.l GontroJ. .No. 01· per cent .No. or per cent cases of total cases of total l .5 13 18 31 28 21 •6 .6 1.2 .6 1.2 1.2 1.2 2.3 4.7 4.1 6.4 5.2 6.4 7.6 10.5 18.0 16 .3 12.2 1 0 2 5 3 4 3 3 10 7 4 10 11 14 12 18 25 35 32 1.0 2.5 1.5 2.0 1.5 1.5 5.0 3.5 2.0 5.0 5.5 7.0 6.0 9.5 12.6 17.6 16.l Total 173 100.3 199 100.5 9 10 11 12 13 14 15 16 17 18 1s, 20 21 22 23 24 Mean Sigma l 1 2 l 2 2 2 4 8 7 11 9 11 19~8 3.5 1 .6 6 7 8 SECOND EXAMINATION ~xperimenta. c.;ontrol 1Jo. or per cent 1.N o. 01 per cent cases of total cases of total 19.6 3.7 o. 2 1 0 2 3 2 4 5 3 5 9 7 8 14 9 1.2 .6 2 1 3 2 1.0 .5 1.5 1.0 .5 1.5 1.5 2.5 6.0 5.0 1.5 1.0 13 14 10 26 27 1.2 1.7 1.2 2.3 2.9 1.7 2.9 5.2 4.0 4.6 8.1 5.2 5.2 7.5 8.1 5.8 15.0 15.6 5 12 10 3 2 9 14 16 10 15 14 25 21 28 14.l 173 100.0 199 99.6 9 18.6 4.9 l 3 3 18.4 5.0 4.5 1.0 8.0 5.0 7.5 7.0 12.5 10.5 96 TABLE II MAXIMUM NUMBER OF ARCADES COMMON TO BOTH EYES No. of Arca des Fi rst Exami na t ion per pe r number cent number cent l!;xpe r imental Contr ol Second Examination per per nw:nber cent number cent Experimen'Eal Control 0 0 0 0 0 0 0 0 0 l 2 l 11 6 7 4 8 4 2 48 29 58 29 64 39 85 43 3 89 54 91 46 65 39 73 37 4 23 14 36 18 20 12 28 14 5 3 2 2 l 5 3 3 2 6 l 1 0 0 5 3 1 1 Total 166 101 198 100 166 100 198 101 Mean 2.9 2.8 2.8 2.7 Sigma o.77 0.93 1.m 0.8E 'i 0 113 0 87 0 99 2 100 Mean 2.7 2.8 113 87 100 99 113 87 1.6 0 2 0 0 0 0 6 1.8 l 8 0 l 0 l 5 Total - 0 0 1 9 24 7 0 l 0 l 4 2.0 1.s l 0 l 21 8 : 18 20 16 38 43 44 37 11 20 16 17 3 0 0 100 0 0 100 0 18 54 58 61 50 35 40 31 49 40 56 35 2 36 23 23 27 20 4 2 4 2 33 34 37 30 1 0 0 0 0 0 7 3 8 5 ... ~-. Per cent exper- coniment trol 0 No. of cases !!;xper- coniment trol Between the hours of 4 - 8 6 Per cent exper- coniment trol 3 No. of cases Exper- coniment trol Per cent exper- control iment No. of cases Exper- coniment trol Between the hours of 10 - 2 . -· - 0 Number of Arcades ....·--· Between the hours of 2-4 and 8-10 MAXIMUM NUMBER OF ARCADES COMMON TO BOTH EYES IN DIFFERENT SECTORS TABLE III -,J (0 0 0 0 0 113 o, 0 99 0 0 87 11.5 3.3 . 20-24 25-29 Total Mean Sigma . - -· 3.3 2.5 3.3 113 13.5 100 87 0 14.2 Qj 0 2 25 68 18 0 11.4 100 1 23 74 2 0 1 20 6 3 3 15-19 5 64 33 37 47 41 10-14 : 2 55 62 43 0 38 7 5- 9 8 5 100 0 2 22 60 16 0 Control NumPer ber cent Maximum extension between the hours of 10 - 2 Experiment NumPer ber cent o- 4 I 1 mm. =15 uni ts 6 Experiment Control Per - - NumPer ber cent ber 001.t Num- Maximum extension between the hours of 2-4 and 8-10 Distance in arbitrary units EXTENSION OF ARCADES FROM LIMBUS TABLE IV -- 99 87 , 3.5 10.s 1 1 3 61 32 1 1 1 3 53 28 l Experiment NumPer ber cent 3.4 10.4 ----- -··· ~. 113 0 0 8 65 35 5 ber Num- Control Maximum e xtensi on between the hours of 4 - 8 100 0 0 7 58 31 4 Per cent (() ():) 99 TABLE V TOT.AL l\TUllIBER OF HOURS OF STREAMER TYPE (both eyes) Experimental Control Hours Number Per cent Number Per cent o- 4 45 49 59 55 5- 9 27 29 32 10-14 12 13 10 15-19 6 7 3 20-24 3 3 3 3 Total 93 101 107 101 ; 31 9 ' 3 100 TABLE VI TOTAL EXTENSION OF STREAMER TYPE ARCADES Extension 1 mm. = 15 uni ts Experimental Control No. of cases Per cent of total o- 4 44 47 59 55 5- 9 28 30 32 30 10-14 12 13 10 9 15~19 6 7 3 3 20-24 3 3 3 3 Total 93 100 107 100 Mean 6.3 5.4 Sigma 4.4 4.6 No. of cases Per cent of total TABLE VII EXTENSION OF LA.ST ARCADE OF STREAMER TYPE ' Extension l mm.= 15 uni ts I Experimental Control No. of cases Per cent of total No. of cases Per cent of total o- 4 29 37 50 58 5- 9 41 53 27 31 10-14 7 9 9 11 15-19 l l 0 0 --- Total 78 100 86 100 101 TABLE VIII PER CENT OF ARCADES OF STREAMER TYPE Experimental Percentage Group I Control No. of cases Per cent of total No. -of cases 0-10 8 18 8 11-20 10 22 10 Per cent of total 15 18 '' 21-30 2 4 8 31-40 8 18 4 7 41-50 4 9 7 13 51-60 2 4 3 5 61-70 6 13 7 13 71-80 1 2 ~ 5 81-90 1 2 1 2 91-100 3 7 4 7 Total cases 45 99 55 100 .15 ... .... Mean 37% 40% ··· --· 102 TABLE IX TOTAL NUMBER OF STREAMERS (both eyes) Experimental Number of Streamers Control No. of cases cent . Per of total No. of cases Per cent of total : 0 56 64 64 l 5 6 10 10 2 10 12 8 8 3 5 6 0 0 4 1 1 8 8 5 3 3 7 7 6 5 6 1 1 7 0 0 3 3 8 · 1 l . 1 l 9 1 1 62 10 11 l Total 87 Mean 1.2 100 103 1.3 101 103 TABLE X WHITE STREAMERS (Total of both eyes) Number or cases Number of Streamers Experimental Control Per cent of total Experimental Control ! o- 4 58 54 62 50 5- 9 24 33 26 31 10-14 8 8 9 8 15-19 4 6 4 6 20-24 4 4 25-29 1 1 0 0 1 1 30-34 3 35-39 I Total 93 107 101 101 104 TABLE XI TRUNKS (Total hours both eyes) Number of case s No. of r Per cent of total Experimental Control Experimental Control 0 35 5 45 39 41 1 18 19 20 17 2 13 12 14 11 3 4 9 4 8 4 55 9 6 8 5 5 5 6 5 6 6 6 7 5 7 1 3 1 3 8 1 0, 1 0 9 1 0, 1 0 10 0 0 0 0 11 0 1 0 1 12 1 0 1 0 13 0 0 0 0 14 00 1 0 1 Total 90 110 100 100 Mean 1.9 Trunks 1.9 -- ' 224 19 Total Mean 2 10 38 85 22-24 2 70 23 52 19-21 10 47 17 17 39 19 43 16-18 4 30 7 7 20 10 22 13-15 5 23 14: 14 10 6 13 10-12 9 100 62 31 32 18 32 23 23 4 4 4 3 7 7- 9 14 28 2 2 5 5 1 ber Num- i i I I i Age: 5 1 1 2 0 cu.muper lative cent per cent 30-39 0 Number Age: 0 0 4- 6 0 0 Number cumuper ative cent • per cent 20 - 29 1- 3 hours No. of Age: 94 100 7 86 50 36 7 36 14 18 18 11 7 7 4 4 cumuper lative cent per cent 40 - 49 LOCKHEgD SURVEY ber Num- Age: , cumuper lati ve cent per cent 50 - 59 (Total number of hours both eyes - experimental and control) CHANCE OF VASCULARITY VHTH AGE TABLE XII Number Age: cwnuper lative cent per cent 60 - 69 I-' 0 CJl . 19 .. -- Average ' both surveys . . 84 17 101 20 - 29 17 i 67 ; 17 ! Ages: 20 Mean l 22-23 6 4 19-21 Total 1 16-18 13-15 17 15 30 3 5 5 4 101 91 74 57 44 8 16 13 16 13 TABLE xr.v 13 38 3 6 5 6 5 18 13 3 100 92 76 63 47 34 16 3 cumuper lative cent per cent 40 - 49 13 CF-ANGE OF VASCULA.RtTY WITH AGE 30 - 39 40 - 49 10 17 17 13 10 7 3 34 10-12 17 5 7- 9 17 5 17 5 ber Num- Arr.e: 4- 6 ber Nurn- cumuper la.tive cent per cent 30 - 39 1 ber Num- Ag:e: 1- 3 hours No. of 20 - 29 cumuper lative cent per cent Age: LOCKHEED SURVEY 13 13 32 3 6 4 4 5 6 2 2 Number Ar£e: 92 73 60 47 31 12 6 101 50 - 59 9 19 13 13 16 19 6 6 cumuper lative cent per cent 50 - 59 (Total number of hours both eyes - experimental and control) CHANGE OF VASCULA.RITY WITH AGE TABLE XIII 11 11 • 12 0 2 l 2 0 5 2 0 ber Num- 17 8 17 42 17 0 101 84 76 59 17 0 cumuper lative cent per cent 60 - 69 60 - 69 Af!.e: I-' 0 m 107 BIBLIOGRAPHY 1. Kruse, H. D., Milbank Memorial Fund (,:uarterly, 20, 245 (1942). 2. Day, P. L., Langston, w. C. and O'Brien, C. S., Am. J. Opthalmology, 24, 1005 ( 1931). 3. Eckardt, R. E., and Johnson, L. V., Arch. Ophth., 21, 315 (1939). 4. Bessey, O. A., and Wolbach, S. P., Jour. Expt. Med., 69, l (1939). 5. Kruse, H.P., Sydenstricker, V. P., Sebreil, W.R., and Clackley, H. M., Public Health Reports, 55, 157 (1940). 6. Sydenstricker, V. P., Sebrell, W. H., Cleckley, H. M., and Kruse, H. D., J. Am. Med. Assoc., 114, 2437 (1940). 7. Report of conference on methods and procedures for nutritional survey. 8. Public Health Reports 57, 189 (1942). Borsook, H., Alpert, E.,&Keighley, G. 1., Milbank :r.i:emorial Fund Quarterly, 21, 115 (1943). 9. Gregory, M. K., Brit. Med. Jour., 2, 134 (1943). 10. Vail, D. and Ascher, K. W., ll. Scarborough, H., Brit. Med. J., 2, 601 ( 1942). 12. Totter, J. R. and Day, D. 1., J. Nutrition 24, 159 (1942). 13. Albanese, A. R. and Buschke, W., Science 95, 884 (1942). 14. Follis, R.H., Day, R. G. and McCollum, E. V., J. Nutrition, 22, Am. J. of Ophthalmology, 26, 1025 (1943). 223 (1941). 15. Sandstead, H. R., Pub. Health Reports 57, 1831 (1942). 16. McRea.ry, J. F., Nicholls, J. V. V., and Tisdall, F. F., Canadian Med. Assoc. J;, 51, 106-110 (1944). 17. Y¢.urnan.s, J • B • 108 18. Tisdall, F. F., McCreary, J. F., and Pearce, H., Canadian Med. Ass. J., 49, 5 (1943). 19. Machelle., T. E. and :McDonald, P. R., Arn. J. Med. Sc., 205, 214 (1943). 20. Riggs, E., Perry, H., Patterson, J.M., Leeson, J., Mosiey, W., and McHenry, E.W., Canad. Jour. Publ. Health, May, P• 193 (1943). 21. Ferguson, w. J. N., Lancet, Apr. 1, 431 (1944). 22. Philpot, A. and Pirie, Biochem. J., 37, 250 (1943). 109 III ROOT BLEEDING 110 The Candidate wishes to thank Dr. F. W. Went for his direction and guidance in this stimulating problem. 111 The force within the root which causes an uptake of water from the soil and exudation from the cut stem is known as "root pressureu. Many theories have been proposed to explain the mechanism of this uptake of water. The passive forces of osmosis and imbibition are not sufficient to completely account for this phenomenon and there is growing evidence that water uptake is directly controlled by the living cell. According to Priestly (1920) the nature of the Casparian strip which is formed around the endodermal cells to force the passage of water from the cortical cells to the xylem through protoplasm points to the role of living cells in water transference. The effect of de- privation of oxygen and change of temperature on water uptake led Blackman (1921) to the belief that vital forces were involved. The need for oxygen for bleeding has also been shown by Brewig (1936) and by Grossenbacher (1938). The depression of water uptake by narcotics (Heyl, 1933) and by cyanide (Rosene, 1941 and Van Overbeek, 1941) also emphasizes the relationship of bleeding to respiration. Henderson '· (1938) has shown that respiration and absorption of water in corn roots run parallel and that energy is expended. The ability to express sap under pressure from a cut stem is decreased when the root is killed. (Kramers, 1932). A definite gradient is established in a living section of Sycamore root with water uptake by the morph ologically lower end and exudation at the upper end (James and Baker, 1933). A similar gradient is maintained by the living onion root (Rosene, 1941). These gradients disappear with the death of the root. Clearly sufficient evidence has been accumulated to indicate that energy provided by respiration is involved in bleeding and that the 112 rate may be expected to depend on the amount of respiratory metabolite present. It was suggested that the extent of translocation of sucrose from other parts of the plant to the root might be measured by the rate of bleeding. (Went, 1944). This paper is a report of attempts to estab- lish the conditions under which bleeding might be dependent on the sucrose supply. 113 EXPERIMENTAL Torm.to plants of the San Jose Canner variety which have been extensively studied by Went (1944) were used in this study. These plants were grown in flats in sand, and after three to four weeks were transplanted to four inch pots. They were watered daily with nutrient sol- ution. The activity of the roots is reflected in the amount of water taken up from the culture solution, in the pressure developed in a manometer attached to the top of the root, or in the rate of exudation of sap from the cut stem. Although these measurements are manifestation of the Sam.fl phenomena, they are not necessarily proportional. In this study the total amount of exudate in a given period was measured. The stems were cut about two inches above the sand just above the first node, connected to an inverted U tube filled with water by means of a short piece of rubber tubing, and the exudate collected in a vial or graduated tube as shown in figure 1. In all cases the bleeding was con- ducted at 25° C. and 60-80% humidity. Sugar was fed to the tomato plants by inserting a leaf top first into sucrose solutions contained in 30 ml. vials supported by small stakes as described by Went and Bonner ( 1943). Sucrose was analyzed by the method of Hassid (1938). The effect of feeding varying amounts of sucrose is recorded in Table 1. 114 TABLE 1 Plants approximately 24 inches tall; all leaves and stem above second leaf cut off; plants placed in dark room at 18° C. for 24 hours with leaves in sugar as shown in table. Average exudation of 10 plants in ml. Treatment 24 hrs. next 48 hrs. control 15.8 10.3 1% sucrose 1 leaf 15.l 9.1 1·<11 /0 sucrose 2 leaves 16·•6 9.3 2% sucrose l leaf 16.l 10.3 2% sucrose 2 leaves 16.l 8.7 115 There is no significant difference between the groups probably because the sucrose was not limiting. In spite of this the rate of b leading dropped off rapidly with time. In the experiments described in table 2 ., an attempt was made to deplete the amount of stored carbohydrate in the plants by increasing the dark period. Although the rate of bleeding dropped rapidly with the length of the dark period, the increase in bleeding in plants fed sugar over the controls was small. TABLE 2 Plants 20 inches tall. Average of ten plants in each group. 5% sucrose fed during entire dark period. Exudation in ml. per 24 hours after varying dark periods Dark period control one leaf in 5% sucrose 0 hrs. 14.8 14.4 24 hrs. 5.1 6.7 48 hrs. 3.2 In the followin e; experiment (table 3) it was found that removing all but one leaf decreased the bleeding rate over plants that had all leaves during the dark period, but this decrease could not be overcome by feeding sucrose. Lowering the temperature during the dark period apparently decreased respiration of carbohydrate and increased translocation of this material so that plants bled as well with one leaf in this period as the controls at 18° c. with all leaves intact. 116 TA.BI.E 3 Average of ten plants 15 inches tall. 24 hours dark period. Effect of temperature and presence of leaves Control one leaf in 5% sucrose All leaves intact Temperature 18° c. 7.2 ml./24 hrs. 7 .4 ml./24 hrs. 1 leaf remaining Temperature 18° C. 4.4 5.5 1 leaf remaining Temperature 7° c. 7.5 8.8 Further experiments were carried out in a~ attempt to decrease storage material and lower the bleeding of control plants. The most effective method was to expose small plants to high temperature, high humidity, and diffuse light (table 4). TABLE 4 Plants about 8 inches tall placed in greenhouse at 26.5° C. and 80% humidity for periods shown; then 24 hours in dark room while sucrose was fed. Nine plants in each group. The effect of light Control 2.2 Sunlight 48 hrs. indirect light 48 hrs 5% sucrose 0.8 Analysis for sucrose showed that plants under normal light had adequate amounts of sucrose which could not be greatly increased by the procedure used (table 5). 117 TABLE 5 Increase of sucrose of roots by feeding leaves Control 5% sucrose Bleeding per root 17.5 ml. 24.2 ml. Bleeding per gram of root 42.7 ml. 45.1 ml. Sucrose per gram of root 5.8 mg. 6.6 mg. In this experiment it was shown that the amount of bleeding was roughly proportional to the wei ght of the root (figure J). However, this relationship did not always hold. Smaller plants in which the stored material was reduced by high temperature and humidity showed little or no sucrose within the limits of the analytical method although the stems of the fed plants did show a sli ght increase (table 6). TABLE 6 Tomato plants 7-8 inches tall. 2 days at 26.5° c., 80% humidity Sucrose content of depleted plants Control 5% sucrose Bleeding per root 2.1 ml./24 hr. 2.4 ml./24 hr. Bleeding per gram of root 13.5 ml./24 hr. 15.3 ml./24 hr. Sucrose in roots o.o mg/gm o.o mg./gm. Sucrose in stems O.O mg/gm 1.5 mg./gm. Since it was clear that the roots were not getting sufficient sucrose by the procedure employed~ an attempt was IIB.de to increase the concentration in the roots by infiltrating the leaves in vacuo with the apparatus shown in figure 3. The leaf was placed in a vial in the 118 usual manner. A split rubber stopper was slipped over the petiole and pushed into the vial. Suction was applied and released. ~en properly infiltrated the ,leaves appeared dark green.) The stopper was then re- moved and the leaf left in the vial for 24 hours. The data in table 7 show that although the sugar reaching the roots was greatly increased the amount of bleeding was only slightly increased. This infiltration procedure was employed in all subsequent experiments where sucrose was supplied. It was nO\llr evident that sucrose was no longer limiting under these conditions. The finding of Skoog et al (1938) that auxin influences the periodic exudation of Pisum sativum suggested that this substance might now be the limiting factor in these experiments. show that this is the case. The data in table 8 It is seen that the sucrose concentration of the roots is lower when auxin is applied possibly because of increased utilization in the presence of auxin. TABLE 7 12 inch plants 3 days in darkness. Sucrose treatment during last day. Average of nine plants for bleeding. Average of three roots for sucrase determination. Increase:/ bleeding by infiltration Control 5% sucrose 13 15 5% sucrose infiltrated Average bleeding per root ml./24 hrs. Average bleeding per gram of root ml./24 hrs. 18 119 Increase in sucrose concentration by infiltration 5% sucrose infiltrated Control 5% sucrose Sucrose in stem mg./gm 4 3 40 Sucrose in root mg./gm 2 12 31 TABLE 8 12 to 20 inch plants 4 days at 25° C. and 80% humidity. Leaves in- filtrated and auxin applied to roots during a 24-hour dark period. Effect of auxin on b leading Bleeding first 24 hrs. second 24 hrs. Mgs. sucrose per root Control 1.5 1.2 0.08 I ndole acetic acid 10 mgs ./ liter 2.1 1.9 0.01 Sucrose 5% 2.8 1.7 0.87 Indole acetic acid plus sucrose 4.2 3.1 0.38 However, the followin g experiment (table 9) did not show this difference, nor was there any difference in concentration whethe r auxin was applied during the period that sucrose was fed or just at the time of bleeding. 120 TABLE 9 10 inch plants 48 hours at 26.5° C. and So% humidity. In experiment 4 auxin applied to roots while sugar was being absorbed. Bleeding avera ge of 6 plants. Bleeding first 24 hrs. Bleeding next 6 days Sucrose mg./100 mg. of root Control 1.7 2.2 0.04 Auxin 10 mg./liter 2.5 5.0 0.12 Sucrose 2.7 2.2 .22 Sucrose plus auxin 4.7 7.4 .24 Sucrose; auxin applied at bleeding time 3.2 6.0 .22 Normal plants; no pretreatment 7.9 5.8 .55 Normal plants of the same lot gave a larger exudate in the first 24 hours than the pretreated plants which had been given auxin and sucrose. This may have been due to the fact that more sucrose was present in the normal plants or to the action of othe r unknown factors. However, the bleeding _o f the sucrose plus auxin plants was si gnificantly hi gher :i:.n the second period than the normal plants. From the amounts of bleeding in the second period in both table 8 and 9 it is evident that auxin is more quickly depleted than sucrose and is more likely to become limiting under these conditions. In table 9, for instance, the bleeding of plants given only sucrose was not greater than the controls during the second period, while the bleeding of plants given auxin alone approached the bleeding of plants given both auxin and sucrose. Van Overbeek ( 1941) has shown tha t the active component producing root pressure is reversibly inhibited by cyanide. As determined in this 121 manner the active component produces about 25 - 50% of the total root pressure. In the present study a more severe poison, sodium arsenite, was employed (table 10) and the inhibition of bleeding was much greater. Since no attempt was made to reverse the inhibition the possibility of destructive changes in the cells or loss of semi-permeability is not ruled out. The fact that small starved plants approximate the lov. values of exudation in large arsenite treated plants suggest that this low value may be the true measure of the non-active bleeding. TABLE 10 Tomato plants 12.5 - 15 inches high. ing. Average of 12 plants. 24 hrs. in darkness before bleed- Sodium arsenite applied to roots. Effect of M/500 Sodium Arsenite on Bleeding Pre treatment None femp 8 days 26~C Humidity -so?, Control 24.2 ml./24 hrs. Arsenite 1.9 ml./24 hrs. o.s 122 DISCUSSION The problem of the utilization of the rate of exudation for studying translocation has necessitated a study of the factors involved in bleeding. The results obtained so far show that at least two factors are required - a food factor, sucrose, and a hormone factor, auxin. similar situation holds for the growth of the Avena coleoptile. A The region of rapid growth receives auxin from the apical region and the food factor from the basal region (Went, 1928, 1936). Under the con- ditions of the present study both factors are produced in the leaves under conditions of light and must be translocated to the roots 1 • footnote 1 Van Overbeek (1939) has reported that auxin is produced in isolated tomato roots, but our results indicate that the amounts must be too small to be non-limitin g in bleeding experiments. This depletion may be brought about by placing plants under conditions producing a high rate of respiration, by placing plants in the dark with a resulting decrease in formation of these substances, or finally by topping the plant and thus removing their source. A simultaneous limi~ ation of growth by both auxin and sucrose has also been found by ~onner (1934) and by Schneider (1938) in the Avena coleoptile, and by Fliry (1932) in Helianthus. Normally auxin is not limiting in the growth of tomato plant stems but removal of the tops increases the need for this factor (Went and Bonner, 1943). 123 It has not yet been possible in our experiments to raise the sucrose concentration of depleted roots to that of untreated controls. This ma.y be the reason that sucrose plus auxin did not increase the rate of bleed-air o ing to that of untreated controls. This may"be due to a decrease in other components necessary to bleeding or to irreversible changes brought on by the temporary deficiency. This is again reminiscent of the studies on Avena coeoptiles by Schneider (1938), who showed that a cdeoptile section depleted of one factor for any length of time never completely recovered even after the addition of that factor. The role of salt has not been investigated in this study, but it may be an additional factor involved in bleeding. Its probable impor- tance is suggested in the data presented by Grossenbacher ( 1938) who showed that the addition of full strength nutrient solution temporarily depressed the pressure developed by Helianthus roots but soon brought about a marked incre~se in pressure over the low salt control plants. Further studies by Skoog, Breyer, and Grossenbacher (1938) also show a marked influence of salt in the presence of auxin on exudation. The role of salt in the uptake of water by plant cells is still in dispute. Commoner:~ Fogel, and Mueller (1943) have shown a marked effect of added potassium in the presence of auxin on the water uptake of potato sections from hypertonic sucrose solution~and they believe that water uptake may be incidental to auxin directed salt uptake. These experimental results are in agreement with the data of Steward, Stout, and Preston (1940) on potassium induced uptake in potato disks, and Hoa.gland and Broyer (1942) on water uptake in barley. On the other hand Reinders {1938) and Van Overbeek (1944) have claimed that water is 124 taken up under the influence of auxin from distilled water alone. The increase in wet wei ght reported by the latter author was only about 10% and could not be increased by adding salt to the medium. The r ea.son for the discrepency is not clear, but the necessity for salt for any extensive water uptake is not ent i rely excluded by the distilled water experiments, for it is probab le t hat there was an initial loss of salt from the slices into the distilled water at the beginning of the experiment (Steward, 1932, 1933). The relationship between salt uptake and water uptake is not clear. It has been stated (Commoner et al, 1943) that the driving force of water uptake is the osmotic pressure of the actively absorbed salt. If this were so, the osmotic pressure of the expressed sap of potato slices should be somewhat greater in the presence of auxin. has clairred that this is not the case. Van Overbeek (1944) The converse view that salt is necessary to maintain the os motic pressure within the ce ll after the active uptake of water is also unlikely. The specific role cf potassium which stimulates water upta ke as compared with calcium whic h depresses water uptake (Stewart, Stout, and Preston; 1940) argues for a more direct role of potassiun1 than the mere maintainence of osmotic equilibrium. It seems more likely that vva.ter absorption and potassium accumulation are to some extent independent of each other but are both reflectiohs of the same respiratory process (Stewart, 1932). The results of Skoog et al, (H38) clearly support this view for the salt concentration in the exudate of auxin treated plants may be greater ot smaller than in the controls. It must, nevertheless, be realized that salt concentra- tion in the xylem is not necessarily a measure of the actual a.mount of 125 accumulated.(Hoagland and Broyer, 1942). That auxin activity is influenced by potassium is evidenced in the findings of Kogl, Haagen-Swit, and Erxleben (1933) that activity of auxin decreased sharply on purification and that activity was restored by the addition of KCl. The stimulation of auxin activity by potassium in the growth of the Avena coleoptile has been reported by Alban and Commoner (1941). Still, there re~ains the possibility that the action of potassium is simply physico-chemical. For instance, it has been of ovion shown that potassium shifts the distribution equilibrium,..between water and agar (Thirna.nn and Schneider, 1938). It is interesting to speculate on the mechanism through which a single substance like auxin can play a role in such complex and apparently diverse phenomena as growth, water uptake, protoplasmic stream~ ing, etc. Evidence for the relation between respiration and water up- take has already been presented. A similar relation exists between respiration and growth (Bonner, 1936) and between respiration and protoplasmic streaming (Thimann and Sweene;y, 1937). The specific respira- tory process affected by auxin in growth (Thimann and Commoner, 1942), protoplasmic streaming (Sweeny, 1941), root number and root length (Sweeney and Thima.nn, 1942), and salt uptake (Machlis, 1944) is said to be the four carbon acid respiratory system. Recently, Berger and Avery (1943) have presented evidence that it is the alcohol dehydrogenase system that is involved. Further experiments may relate root bleeding to this same respiratory system, and thus further correlate the diverse roles of auxin in plant metabolism with a single energy producing system. 126 It is very likely that the influence of auxin on growth is through the mechanism of water uptake. Since water can be actively absorbed even by flaccid cells in the presence of auxin (Commoner and Ma.zia, 1942), no osmotic forces need be postulated for growth. It is not unreasonable to assum~ that the excess pressure developed by active water (and salt) uptake may extend the cell until the secondary wall is laid down, and by its rigidity prevents furt he r growth. In this connection it has been found that low temperature, by preventing the d~position of secondary walis, allows growth to continue longer than at room temperature (Heyn and Van Overbeek, 1931) (Bonner, 1934). In the light of recent evidence it may be unnecessary to postulate an ipfluence of auxin on wall placticity (Heyn, 1940) or on the lm-rering of t he wall pressure (Van Overbeek, 1944). Auxin through the control of water movement may act no~r only in growth but also in non-growing tissue. It has been shown, for example, that auxin applied to the pulvinus of Mimosa will initiate turgor movements (Burkholder and Pratt, 1936). ro o + pr ~ H v f f It has been suggested that '~,. plays a part in refilling the xylem vessels which may become partly blocked with air. If this is so, the value of root pressure to the plant is not in the amount of water moved, which is relatively very small compared to the amount moved by transpiration, but in maintaining a continuous water column which is the very basis for movement by transpiration. Thus, bleeding is but one example of auxin controlled water relations in the plant. 127 SIDIIMARY The application of root bleeding to the determination of the amount of translocation of sugar has led to a study of the factors limiting this phenomenon. It has been shown that a respiratory substrate and auxin are involved. It was possible to deplete tomato plant roots of both of these factors by growing the plants in darlmess or at hi gh temperature and humidity. Under these conditions both factors became limiting. The sucrose concentration of the roots could• be raised by in' filtratin g the leaves, but t he values did not reach those of undepleted plants. This may be the r eason that roots from treated plants did not bleed as well as roo t s from normal plants. The possibl e relation of salt to water uptake and the role of water upta ke to cell elongation i s discussed. Fi g ure 1 Setup for Measurement of Bleeding Fi g ure Stopper for 2 Infiltration 30 X • Control Sucr o se •x X 25 ~ • •• • 20 co H ..c • • <;j1 C\l ':-15 • • .- rl s; ;,( X bD ~ •r-l ro (j) <D rl ,D 10 )( • CH 0 • .µ ~ '.:5 0 8 • 5 '~ 0--------------------------------. 0 10 20 40 30 50 Weight of Root mg . RELATIONSHIP BETWEEN BLEEDI NG hND ROOT WEIGHT F,i g ure 3 60 128 Bibliography Alban, H. G., and Commoner, B., 1941, Biol. Bull. 80:310. Blackman, V. H., 1921, -New Physiolo gist, 20: 106-115. Berger, J. and Avery, G. S., 1943, Science 98: 454-5 (1943). Bonner, IJ). and Went, F. W., 19(43. Bonner, J., 1934, Dissertation, Calif. Inst. of Technology. Bonner, J., 1934, Pro. Nat. Acad. Sc., 1934, 20, 393. Bonner, J., 1936, J. Gen. Physiology 20: 1. Bonner, J., Thesis C.I.T., Pasadena. Brewis, A., 1936. Jahn, Wiss. Bot., 82: 803-828. IA- Bif kho ide r, P.R. and Pratt, R., 1936, Amer. Jour. Bot. 22: 46-52. ~ Commoner, B., and Thimann, K. v., 1941, Jour. Gen. Physiol. 24: 279-296. Commoner, B., and Niaizt, D., 1942, Plant Physiol., 17: 682-685. Commoner, B., Fogel, S., and Muller, W. H., 1943, Amer. Jour. Bot. 30: 23-28. Fliry, M., 1932, Jahr~-Wiss. Bot. 77: 150-184. Grossenbacher, K., 1938, Plant Physiol. 13: 669. Grossenbacher, K., 1939, Amer. Jour. Bot., 26: 100. Hassid, M. z., 1937, Ind. Eng. Chem., Anal. Ed., 9: 228. Henderson, L., 1934, Plant Physiol., 9: 283. Heyn, A.N.J. and Van Overbeek, J., 1931, Proc. K. Akad. 1~etensch. Amsterdam, 34: 1190-1195. Heyn, A:r...N. J., 1940, Bot. Rev., 6: 515-574. Heyl, J. G., 1933, Planta 20: 294-353. Hoagland, D., and Broyer, 1942, Jour. Gen. Physiol., 25: 865-880. James, N. u., and Baker, H., 1933., New Phytologist 32: 317. 129 Kogl, F. A., Ha.a.gen-Smit, A. J •• and· Erxlebern, 1933, Zeit.• Physiol. Chem., 214: 241-261. MachlZs, L., 1944, Am.ir. Jour. Bot., 31: 183. Van Overbeek, J., 1939, Bot. Gaz., 101: 450. 1942, Amer. Jour. Bot., 29: 667-683. 1944, Amer. Jour. Bot., 31: 225-269. 1944, Annual Review of Biochemistry. Priestly, J. H., 1920, New Phytolo~ist, 19: 189-200. Ro$ene, H. F., 1941, Plant Physiol., 16: 19-38. Schneider, c. L., 1938, Amer. Jour. Bot., 25: 258-270. Skoog, T., Breyer, T. C.~ and Grossenbacher, K. R., 1930, Amer. Jour. Bot., 25: 749-759. Stewart, F. C., 1932, Protoplasma, 15: 29-58. 1933, Protoplasma, 18: 208-242. Stewart, F. c., Stout, P. P., and Preston, c., 1940, Plant Physiol. 15: 409-447. Sweeney, B. M., 1941, .l1Jner. Jour. Bot., 28: 700-2. Sweeney, B. M. and Thimann, K. V., 1942, J. Gen. Physiol. 25-841. Reinders, D. E., 1938, Proc. Kon. Aka.d. Wetensha.p, Amsterdam, 41: 820-831. Ruge, U., 1937, Zeit. Bot., 31: 1-56. Thimann, K. V., and Sweeney, B. M., 1937, Jour. Gen Physic., 21, 123, 1937. Thirnann, K. V. and Schneider, C. L., 1938, Amer. Jour. Botany 25: 270-280. Went, F. W., 1928, Rec. Trav. Bot. Neerl., 25: 1-116. Went, F. W., 1936, Biol. Zentral bla.tt, 56: 449-63. Went, F. w. and Bonner, D., 1943, Arch. of Biochem., 1: 439-452. Went~ F. W., 1944, Unpublished. 130 Went, F. w., 1944, Amer. Jour. of Bot., 31: '1135: J50.