S 587 
 .R6 
 Copy 1 
 
 THE REACTION BETWEEN AMINO-ACIDS AND 
 
 CARBOHYDRATES AS A PROBABLE CAUSE 
 
 OF HUMIN FORMATION 
 
 M. L. ROXAS 
 
 (From the Laboratory of Agbiculturax Chemistry op the 
 University of Wisconsin, Madison) 
 
 Reprinted prom 
 
 THE JOURNAL OF BIOLOGICAL CHEMISTRY 
 Vol. XXVII, No. 1, October, 1916 
 
c- 
 
 Reprinted from The Jodrnal of Biological Chumihtky, Vol. XXVII, No. 1. I9I6 
 
 THE REACTION BETWEEN AMESrO-ACIDS AND 
 
 CARBOHYDRATES AS A PROBABLE CAUSE 
 
 OF HUMIN FORMATION.* 
 
 By M. L. ROXAS. 
 
 (From the Laboratory of Agricultural Chemistry of the 
 University of Wisconsin, Madison.) 
 
 (Received for publication, July 31, 1916.) 
 
 The studj' of the black substances obtained when proteins 
 are hydrolyzed in strong acid solution is of great interest at the 
 present time on account of their bearing on the natural melanins 
 and on the quantitative determination of certain amino-acids 
 in proteins. Grindlej^' and his coworkers state that humin 
 nitrogen causes an error in the analj'sis for amino-acids of com- 
 mon foodstufTs when the Van Slyke amino nitrogen determination 
 is directly applied to them. This view on theoretical grounds 
 was also expressed by Hart and Bentley.- It is therefore very 
 important to know more about the structure and mode of forma- 
 tion of these compounds. 
 
 Mulder' was the first to show that albumins separate flooculi of a brown 
 or black color on being boiled with concentrated hydrochloric or sulfuric 
 acids. Hausmann-* made similar observations with globin. Samuelly" 
 pointed out that the formation of these "artificial melanins" or "melanoi- 
 dins" might be a secondary reaction between amino-acids and carbohy- 
 drates. Maillard^ conducted experiments along this line and found a 
 
 * The work described in this article forms part of a thesis submitted 
 in partial fulfilment of the requirements for the degree of Doctor of Phil- 
 osophy in the University of Wisconsin. 
 
 ' Orindley, H. S., and Slater, M. E., /. Am. Chem. Soc, 1915, xxxvii, 
 2762. 
 
 '- Hart, E. B., and Bentley, W. H., J. Biol. Chem., 1915, xxii, 477. 
 
 3 Mulder, G. J., in Mann, G., Chemistry of the Proteids, London, 1906, 87. 
 
 ■* Hausmann, W., Z. physiol. Chem., 1900, .xxix, 140. 
 
 ' Samuelly, F., Beitr. chem. Phys. u. Path., 1902, ii, 355. 
 
 'Maillard, L. C., Compt. rend. Acad., 1912, cliv, 66. 
 71 
 
72 Humin Formation 
 
 numlKT of them reacted with sugars. His experiments, however, were 
 carried on in aqueous solutions at a very high concentration and tem- 
 perature and it is doubtful whether under these conditions the reaction 
 is similar to what takes place in the formation of either the "natural" 
 or artificial melnnins. It will be shown lator in this paper that not all 
 the amino-acids found reactive by Maillard reacted at all at low con- 
 centration in water. Gortner and Blish' made the important discovery 
 that when trj-ptophane is boiled with sugar in 22.9 per cent hydrochloric 
 acid solution SO per cent of its nitrogen is converted into humin nitrogen. 
 They conclude from their experiments that tryptophane alone is responsi- 
 ble for humin formation. Grindley and his coworkers' disagree with this 
 conclusion since they found evidence that other amino-acids give the same 
 reaction. 
 
 In view of these conflicting statements and in the hope that the 
 study of the reaction between amino-acids and carbohydrates 
 would throw some light on the structure and mode of formation 
 of the humin substances, it was thought worth while to determine: 
 (1) Which amino-acids react with carbohydrates under a given 
 set of conditions. (2) Whether difi'erent sugars behave alike 
 toward the same reactive amino-acid. (3) What group of the 
 reactive amino-acids takes part in the reaction. 
 
 REVIEW OF THE LITEIUTUHE. 
 
 Udnlnszky' and Hoppe-Seyler' have shown that when sugar is boiled 
 with aci<ls humin substances are formed, and if boiled in the presence of 
 a nitrogenous material, the humins may also contain nitrogen. Udrdns- 
 zky found, for example, that glucose and urea boiled together in strong 
 hydrochloric acid solution formed humins which contained about 6.73 
 per cent nitrogen. 
 
 Saniuelly' was the first to study the behavior of humins, or "melanoi- 
 dins," towards oxidizing and reducing agents. lie prepared his "melanoi- 
 dins" from commercial scrum albumin according to Schmicdeberg's'" 
 method modified by himself. He subjected his product to the action of 
 
 ' Gortner, R. .\., and Blish, M. J., J. Am.Chem. Soc, 1915, xxxvii, 1030. 
 After this work was completed and t^ent for publication another article 
 by Gortner on humin formation appeared (J. Hiol. Chcm., 1916, xxvi, 
 177). In this article Gortner admits that amino-acids other than trypto- 
 phane may be involved in humin formation, which is in harmony with 
 the reoulls reported here. 
 
 » Idnlnszky, L. v., Z. physiol. Chcm., 1.S&8, xii, 33. 
 
 » IIop|)c-.Soyler, F., Z. physiol. Chem., 1SS9, xiii, 06. 
 " SchmiedclxTg, O., .Arch. cxp. Path. u. Pharm., 1897, xxxix, 1. 
 
i\I. L. Roxas 73 
 
 HI and PI, in a closed tube kept for S hours at 200-210°C. Among other 
 products he obtained pyrrol, as detected by the color reaction with pine 
 shavings, and either pyridine or piperidine or some derivative of either 
 one of these bases. Ammonia was also evolved in largo amounts. None 
 of the reduction products obtained as above gave indol or skatol on fusion 
 with alkalies, while the humins before reduction gave on similar treatment 
 an unmistakable odor of both. Samuelly also tried reduction with zinc 
 dust in a current of hydrogen. From this treatment he obtained pyridine, 
 pyrrol-like substances, skatol, and small amounts of an aromatic com- 
 pound of the benzaldehyde series. The same investigator prepared humin 
 substances from some amino-acids and glucose. He" heated for 18 hours 
 a mi.xture of 10 gm. of glucose, 50 cc. water, 15 cc. concentrated HCl (sp. 
 gr. 1.19), and sufficient amount of the different amino-acids so as to have 
 in the solution 0.7 gm. of nitrogen. He tried ammonium chloride, urea, 
 acetamide, glycocoll, aspartic acid, cystine, and tyrosine. In each case 
 he found nitrogen in the melanin. No attempt was made, however, to 
 determine whether the hiunin nitrogen formation was due to adsorption 
 or to a definite reaction. It is interesting to note that all of the humins 
 so prepared gave off pyrrol on dry distillation with zinc dust, no pyridine, 
 and on fusion with alkalies only the humin prepared from tyrosine pro- 
 duced an odor of indol. Nencki'- and Berdez" obtained similar results 
 with the natural melanins. By alkali fusion these authors obtained from 
 tumor melanin, indol, skatol, volatile fatty acids, hydrocyanic acid, suc- 
 cinic acid, and other unidentified products. Pyrrol was obtained on dry 
 distillation, and after heating the melanin to 300°C. for some time, upon 
 addition of an alkali pyridine was detected. 
 
 Gortner and Blish' heated pure zein, plus tryptophane, plus carbohy- 
 drate in 22.86 per cent HCl, and obtained 16.5 per cent of the total nitro- 
 gen of the mixture in the humin form. When tryptophane alone was 
 heated with sugar in acid solution 86.56 per cent of its nitrogen was found 
 in the hvunins. On the other hand, when histidine plus zein was heated 
 with acid only 0.51 per cent of the total nitrogen was found in the humins. 
 This amount was almost the same as that obtained when zein alone was 
 heated in acid. They did not try, however, heating zein, plus histidine, 
 plus sugar, in acid. Among the conclusions these authors draw from their 
 experiments are the following: (1) The humin nitrogen belongs to "no 
 amino-acids other than tryptophane." (2) "The reaction involved .... 
 is probably the condensation of an aldehyde with the — NH group of the 
 tryptophane nucleus." (3) Histidine can be eliminated "as a factor in 
 the formation of humin nitrogen." 
 
 Grindley and Slater' have tried to apply the Van Slyke amino nitrogen 
 determination directly to the anal3'sis of feedingstuffs. As is to be expected, 
 
 " Samuelly, Beiir. chem. Phys. u. Path., 1902, ii, 383. 
 '- Nencki, M., and Sieber, N., Arch. exp. Path. u. Pharm., 1887, xxiv, 
 17. 
 
 " Berdez, J., and Nencki, M., Arch. exp. Path. u. Pharm., 1886, xx, 346. 
 
74 TTuniin Formation 
 
 on account of the hifili carbohydrate content of these, the humin fraction 
 in their nitrogen distribution is very high, varying from 3. So per cent in 
 blood meal to 15.79 per cent in alfalfa hay, expressed as per cent of the total 
 nitrogen. In discussing the origin of these humin substances these investi- 
 gators disagree with the conclusion arrived at by Gortner and Blish, that 
 the humin nitrogen of protein hydrolysis has its origin exclusively in the 
 tryptophane nucleus, since they have obtained "results that clearly indi- 
 cate that in addition to tryptophane a numlwr of other amino-acids when 
 gently boiled with 20 per cent HCl for 24 to 30 hours in the presence of 
 pure glucose give humin nitrogen. Preliminary experiments show that 
 under the above treatment 4.7 to 6.3 per cent of the total nitrogen of lysine 
 and cystine respectively is separated as humin nitrogen." 
 
 Since Bourquelot and Bcrtrand discovered tyrosinase, this enzjTne has 
 received much attention from a great number of investigators. Only 
 that part of the work relating to the action of tyrosinase on the different 
 amino-acids and related substances will be reviewed here." 
 
 The effect of tyrosina.se on tyrosine is described by Bertrand." A solu- 
 tion of tyrosine to which an extract of fyrosin.ise is added first becomes 
 red, then inky black, and finally deposits a black precipitate. He proved 
 definitely that atmospheric oxygen is essential to the change by conduct- 
 ing experiments in vacuo and in the air. Von Fiirth and Schneider" used 
 the blood (heniolymph) of the pupse of a butterfly, Dciciphila elpenor. 
 They separated the enzyme from the other substances in the blood by frac- 
 tional precipitation with ammonium sulfate. It was found to give a yel- 
 lowish red substance with pjTochatechol ; with hydroquinone it gave a 
 red solution, which then became turbid and finally deposited a consider- 
 able brownish precipitate. It also acted on adrenalin, giving a dirty brown 
 color. Oxyphenylethylamine became yellowish brown and finally gave 
 an olive-colored precipitate. But tyrosinase has no action on casein 
 itself. The same authors isolated the black substance produced from ty- 
 rosine by the tyrosinase of Dciciphila puptc and determined its elemen- 
 tary composition. Below is given a comparison between the percentage 
 composition of this black substance and of tyrosine respectively:. 
 
 Black lubfltanoo (humin) 
 
 from tyranno. TjTosinc. 
 
 perctnt p<rcenl 
 
 C 55.66 59,60 
 
 H 4.45 0.08 
 
 N 13.74 7.74 
 
 26 37 26.58 
 
 " Bourquelot and Bertrand, G., Bull. Soc. Mycol., 1896, xii, 18. A com- 
 plete list of references up to the time of its publication is found in Kastle's 
 Oxidases, Hull. Hyg. Lab., 59. 
 
 '* Bertrand, G., Compl. rend. Soc. bioL, 1896, cxxii, 1215. 
 
 "Von Kiirth, O., and Schneider, H., Beitr. chem. Phys. u. Path., 1901, 
 i. 229. 
 
M. L. Roxas 75 
 
 In the formation of this "artificial melanin" from tyrosine there is an 
 increase in the nitrogen content froni 7.74 to 13.74. Such an increase is 
 only conceivable in one of two ways; either there is a breaking up of the 
 tyrosine molecule, or some other nitrogenous substance besides tyrosine 
 takes part in the formation of the melanin. The latter must be the case 
 since tyrosinase, being but a weak oxidizing agent, would be unable to 
 break down the benzene nucleus. The nitrogenous compound that took 
 part in the reaction must evidently have come from the tyrosinase prepara- 
 tion itself. This black product also yields a skatol-like odor on fusion 
 with alkali. In connection with the wide distribution of tj-rosinase in 
 both the vegetable and animal kingdom the following is quoted from 
 Kastle's monograph : 
 
 "Von Fiirth and Schneider are therefore of the opinion that probably 
 wherever melanotic pigments occur in the living tissues of the lower and 
 higher animals they originate as the result of the action of appropriate 
 enzymes on substances of aromatic nature. They point out in this con- 
 nection that Salkowski and Jacoby have shown independently that ty- 
 rosine results from the autolysis of various animal tissues. It would seem 
 likely, therefore, that in the formation of melanotic pigments two ferments 
 are jointly concerned: one, an autolytic ferment capable of splitting off 
 tyrosine or a similar aromatic complex from the protein molecule, and the 
 other tyrosinase, which transforms the tyrosine into melanin." 
 
 But one of the most interesting phases of the investigations on tjTosinase 
 is that relating to its effect on the products of protein degradation and re- 
 lated substances. Bertrand and Rosenblatt'' have found that this enzyme 
 acts equally well upon racemic and Z-tj-rosine. Chodat and Staub'' dis- 
 covered that albumoses do not give a red color with tyrosinase but glycyl- 
 tyrosine anhydride gives such a coloration. In a later article'' these 
 authors observed that glycine, leucine, and alanine retard the action 
 of tjTOsinase; that dipeptides such as tyrosine anhydride, and glycyl- 
 tyrosine anhydride produce yellow substances which do not become black 
 as does tjTOsine itself. When, however, glycine, leucine, or alanine is 
 present, a red coloration similar to that resulting from tyrosine is ob- 
 tained : glycyltyrosine anh}'dride with glycine gives a rose color changing 
 to bluish green; with alanine the color is deeper red, with leucine deep 
 brown. But their most striking discovery is that phenylalanine is not 
 acted on by tyrosinase. This, however, acts readily on p-cresol, less 
 readily on jn-cresol, and still less readily on o-oresol. In fact these same 
 authors observed that the enzj-me acts most readily on the para-homo- 
 logues of phenol. Amino-acids like glycine increase the rapidity of the 
 action of tyrosinase on jj-cresol, producing a \dolet color which ultimately 
 becomes blue. Bei'trand undertook to investigate the action of tyrosinase 
 
 " Bertrand, G., and Rosenblatt, M., Compt. rend. Soc. hiol., 1908, cxlvi, 
 304. 
 
 "Chodat, R., and Staub, Arch. Sc. Phys. Nat.; 1907, xxiii, 265; xxiv, 
 172. 
 
76 Humin Formation 
 
 from wheat bran on compounds analogous to tyrosine and to phenylalanine; 
 that is, compounds with and without the phenolic hydroxyl group. Thus 
 he found phenylalanine, phenylcthylamine, phenylmethylaniine, phenyl- 
 aminoaeetic acid, phenylpropionic acid, phenylacetic acid, alanine, and 
 glyeocoll produced no coloration at all. On the other hand the following 
 compounds with phenolic hydroxyl groups produced coloration as follows : 
 
 Tyrosine Grenadine-red, inky black. 
 
 p-Hydroxj-phenylcthylamine Grenadine-red, olive-black. 
 
 p-Hydroxyphenylmethylamine Orange-yellow, orange-red, clear ma- 
 roon. 
 
 p-Hydroxyphenylamine Orange, mahogany-red, brown. 
 
 p-Hydroxyphenylpropionic acid Orange-yellow, grenadine-red, brown. 
 
 p-IIydroxybcnzoic acid Rose, orange, yellow. 
 
 p-Cresol Yellow, orange, red. 
 
 Phenol Yellow, orange, red, brown. 
 
 He concludes, therefore, that tyrosinase acts only on those compounds 
 containing a phenolic group. 
 
 In 1907 Abdcrhalden and Guggenheim" published an interesting arti- 
 cle on the effect of tyrosinase from liiissula dclica on tyrosine, tyrosine- 
 containing polypeptides and other related substances. They observed 
 that glyeocoll, (/-alanine, (/-valine, /-proline, (/-serine, (/-/-isoserine, and 
 /-phenylalanine retard the action of tyrosinase on tyrosine only slightly un- 
 less present in very large concentrations. The largest concentration used 
 was molar; /-aspartic acid and (/-glutamic acid, however, even when pres- 
 ent at a concentration of 0.01 molar retard the action considerably. 
 The same authors found that the enzyme has no effect on diiodotyrosine, 
 /-phenylalanine, /-proline, or cystine. But /- and (/-tyrosine, homogen- 
 tisic acid, and tryptophane showed a color cliango. Particularly interest- 
 ing was the case of (/-tryptophane. The authors state that at first they 
 thought that the coloration with tryptophane may be due to traces of 
 tyrosine. They, however, used a very pure product. They repeated 
 their experiment but always came to the same result. Furthermore, 
 they tried the effect of tyrosinase on solutions of tryptophane-containing 
 polypeptides and found development of color. They therefore conclude 
 that this coloration must not be ascribed to the presence of traces of tyro- 
 sine. Still more interesting is the fact that neither indol nor skatol were 
 found to produce coloration. Abdcrhalden and Guggenheim in the same 
 article describe the effect of tyrosinase on polypeptides containing tyro- 
 sine. The color developed in these ctises is modified to some extent by 
 the nature of the amino-acid combined with the tyrosine in the polypeptide. 
 Addition of some .amino-acids were also found either to accelerate or to 
 retard the action of tyrosinase on the polypeptide. Thus proline accoler- 
 
 '• .'Vbderhaldcn, E., and Guggenheim, .\1., Z. physiol. Chctn., 1937-OS, 
 liv, 331. 
 
M. L. Roxas 77 
 
 atcs considerably the action of the oxidase on glycyl-^ty^osine anhydride, 
 while aspartic acid and glutaminic acid retard the action. On the other 
 hand halogen derivatives of the polypeptides were not acted upon by ty- 
 rosinase. The same authors also found, as did Bertrand, that tyrosinase 
 acts on phenol, giving a brown color, which was modified by amino-acids. 
 Thus glycocoll plus phenol gave a cochineal color, while proline and phenol 
 gave violet. The authors finally concluded that the amino-acids, when 
 present, apparently take part in the production of the pigment. In a 
 later article" these same authors point out that tyrosinase acts on adrena- 
 lin with the rapid production of a red color and ultimately dark red floc- 
 culi. All three isomers of adrenalin are affected with equal rapidity. 
 
 It is to be regretted that in none of the above cited contributions was 
 either arginine, histidine, or lysine tried. It is hoped that this omission 
 will be filled in the near future. 
 
 EXPERIMENTAL. 
 
 The fact that zein, when boiled with glucose in 22.68 per cent 
 hj'drochloric acid solution, increases its humin nitrogen from 
 0.56 to 1.8-i per cent indicates, as Grindley and his coworkers^ 
 suggested, that other amino-acids besides tryptophane take part 
 in nitrogenous humin formation. Only a small per cent of some 
 of these amino-acids may take part in this formation so that only 
 by working with the individual amino-acids is it possible to de- 
 termine whether the humin nitrogen was due to a definite reac- 
 tion or to an adsorption. Again it is only by working under 
 approximately the same set of conditions that it is possible to 
 detect differences in behavior between the different amino- 
 acids. The following procedure was, therefore, adhered to as 
 consistently as practicable. 
 
 The amino-acid, plus sugar, plus 50 cc. of water or hydrochloric 
 acid solution of the specified strength was heated for 48 hours 
 in a 300 cc. Kjeldahl flask on a sand bath. The flask was pro- 
 vided with a reflux condenser made from a large test-tube fitted 
 with cork and tubings for a current of cold water. After heat- 
 ing, the digestion mixture was neutrahzed with the calculated 
 amount of sodium hydroxide solution. The salt thus formed 
 coagulated most of the precipitate that may have existed in a 
 colloidal state in the solution. The mixture was then filtered into 
 200 cc. graduated flasks and the humin was washed repeatedly 
 
 '» Abderhalden and Guggenheim, Z. physiol. Chem., 1908, Ivii, 329. 
 
"^8 Iluniin Formation 
 
 witli boiling water until the flask was filled to the mark. This 
 amount of washing was found to be sufficient to remove almost 
 all of the adsorbed amino-acid which could be removed by this 
 treatment alone. The humin with the filter paper was then 
 Kjeldahled. The filtrate was either Kjeldahlcd or \'an-SIyked 
 or used for both determinations. 25 cc. portions were taken for 
 the Kjcldahl and 10 cc. portions for the Van Slyke determination. 
 
 The nitrogen content of the amino-acids was determined either 
 by Kjeldahl's or by Van Slyke's method or by both. The per 
 cent of nitrogen was the only thing used to establish the identity 
 and purity of the compounds. 
 
 The following amino-acids were furnished by Professor Hart: 
 
 found. TheoroUosI. 
 
 per imt pa- cml 
 
 j^*'"°« 15.70 15 75 
 
 ^y^""® 11.2-11.0 11.67 
 
 Ty°8''"' 7.67 7.72 
 
 Lysine hydrobromide 
 
 (2C,H„N,0, . HBr . H,0) 14.72 1432 
 
 Tryptophane 6 .44 (Amino N) 6 80 
 
 Phenylalanine* 895 g g^ 
 
 • The phenylalanine wius kindly furniahod by Dr. T. B. Osborne of 
 New Haven. 
 
 The following amino-acids were prepared : 
 
 ^'"'■■°--''- Found. ^''"*«°Th«,«.U«.. 
 
 IMT cent pa- oral 
 
 Leucine, from zein 10. 90 10.70 
 
 Proline, from zcin, also from KPlatin 12.20 12 17 
 
 „, (No amino N) 
 
 Cilutiuninic acid, from gliadin 7.17 75^ 
 
 Arginine (freeHrom gelatin 29 9.5 32 ^j 
 
 '^'."'"°-'^' 7.'70 8.04 
 
 Histidine dihydrochloride, from blood 17 2.5 I8 f 
 
 ^'""N 5-6 g-,- 
 
 In the preparation of the above amino-acids the directions 
 given in Abderhalden's Arbeilsnwthodcn were followed. The 
 per cents of nitrogen found for arginine and histidine respectively 
 were not quite up to the theoretical, but since the amino nitrogen 
 was almost one-fourth of the total in the arginine .sample and one- 
 third in the histidine, it was evident that the samples of both 
 
M. L. Roxas 79 
 
 these amino-acids were free from other amino-acids, their low 
 total nitrogen content being due to moisture. The nitrogen 
 determinations of these amino-acids were made on the same day 
 that the experiments on humin formation were started. 
 
 Due to the scarcity of material it was found impossible to 
 recrystallize some of the amino-acids in order to obtain as pure 
 a product as could be desired. 
 
 The results are shown in the following table. All the experi- 
 ments were in duplicate and average figures are given. 
 
 The results show that neither alanine nor leucine give humin 
 nitrogen. Glutaminic acid when boiled with sugar in 2 per cent 
 acid solution yields some humin nitrogen, but none in 20 per 
 cent acid. Attention is called to the fact that glutammic acid 
 on heating even at the concentration used seems to form pyrro- 
 lidon carboxyHc acid readily, as evidenced by the loss of activity 
 of its amino nitrogen in Experiments 13, 14, and 16. Such a 
 formation does not take place in strong acid. Phenylalanine 
 yields about 1.65 per cent of its nitrogen in the humin in 20 per 
 cent acid solution. Proline does not give humin nitrogen with 
 glucose with 20 per cent acid, but seems to react to some extent 
 with xylose and fructose in 4.15 per cent acid solution. Cj^stine 
 with 20 per cent HCl jdelds about 3.1 per cent humm nitrogen 
 and the noteworthy fact about this amino-acid is the deeply 
 colored filtrate it produced. The same was observed with the 
 filtrate from the tyrosine-sugar experiments. As much as 15 
 per cent of tyrosine nitrogen may be converted into humin nitro- 
 gen. 
 
 The cases of the three hexone bases are particularly in- 
 teresting. When boiled with sugars in 20 per cent HCl solution 
 all three yield some of their nitrogen as humin nitrogen. Ar- 
 ginine and lysine, with sugar, give more deeply colored filtrates 
 than histidine. If the deep coloration of the filtrate indicates 
 reaction, then it must be stated definitely that in the case of 
 cystine, tyrosine, arginine, and lysine in 20 per cent HCl, the hu- 
 min nitrogen is due to a reaction and not to an adsorption. An- 
 other fact that supports the contention that a definite reaction 
 is responsible for humin formation at least in the case of tyrosine 
 is that phenylalanine gives but little humin nitrogen. If this were 
 a case of adsorption, then there should probably be no differ- 
 
80 
 
 Humin Formation 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 fe 
 
 
 
 
 
 o 
 
 
 
 
 
 o 
 
 
 
 
 s 
 
 "cj 
 
 
 
 
 s 
 
 
 
 1 
 
 
 
 Ul 
 
 : 
 
 = 
 
 
 _« 
 ^ 
 
 be 
 
 
 es 
 
 
 5 
 g 
 
 
 
 
 
 <J 
 
 
 
 
 
 ^ 
 
 '" 
 
 - 
 
 "• 
 
 
 - 
 
 - 
 
 
 
 
 (S 
 
 
 
 
 
 
 
 
 , . 
 
 
 00 
 
 SS 
 
 oo 
 
 00 
 
 lO CO 
 
 CO « 
 
 
 - Ml 
 
 « 
 
 
 
 
 o 
 
 "5 m 
 
 CO n 
 
 
 E 
 
 « 
 
 re 
 
 TO 
 
 re 
 
 o 2 
 
 •"S" •^ 
 
 
 •3 ■gj.~-sS 
 
 E 
 
 
 i ^8 
 
 
 o 
 
 r^ 
 
 g 
 
 s 
 
 s 
 
 s s 
 
 
 " 4^1 1 
 
 E 
 
 o 
 
 o 
 
 O 
 1 
 
 o 
 1 
 
 o o 
 
 d o 
 
 
 - IH^ii 
 
 (» 
 
 s 
 
 00 
 
 o 
 
 § 
 
 S 
 
 s ^ 
 
 CO « 
 
 
 E 
 
 CO 
 
 TO 
 
 CO 
 
 CO 
 
 so 2 
 
 21 ■* 
 
 4; 
 
 „ •i.,5|§i 
 
 . 
 
 s 
 
 00 
 
 § 
 
 o 
 
 ■O M 
 
 S § 
 
 '3 
 
 
 E 
 
 to 
 
 CO 
 
 ;2 
 
 'T 
 
 CO 2 
 
 2 ■* 
 
 
 " Mlli 
 
 E 
 
 23 
 
 
 
 
 
 
 
 
 
 
 !•§ 
 
 ^ 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o o 
 
 o o 
 
 
 
 E 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o o 
 
 O o 
 
 
 Bg 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 S 
 
 
 s 
 
 
 + 
 
 
 : + 
 
 
 + : 
 
 
 
 
 
 S» 
 
 
 + 
 
 + 
 
 
 + 
 
 
 g 
 
 
 J ^ 
 
 
 g ■ 
 
 
 c 
 
 1 
 
 1 
 
 d 
 a 
 •3 
 
 < 
 
 gm. + 20 per cent HCl 
 gm. + 0.720 glucose + ! 
 t HCl 
 
 1 
 
 i 
 
 do 
 
 + a 
 
 J 
 
 i 
 
 d 
 
 + 
 
 i 
 
 O 
 B 
 
 c 
 
 a 
 
 i 
 
 o 
 4- 
 
 i 
 
 
 
 c 
 u 
 
 i 
 
 s 
 
 d 
 
 + 
 
 i 
 
 c 
 u 
 
 Leucine. 
 Km. + 4.1.5 per cent HC 
 Km. + 0.720 gm. gluco 
 
 c 
 
 2 ; 
 + g 
 
 
 
 
 11 
 
 c o t =■- 
 
 ^1 
 
 ^3 
 
 8 
 
 CO ro 
 
 "2 "5 
 
 -s" 2 •* 
 
 
 
 
 o o 
 
 d d 
 
 o 
 
 o 
 
 
 o o 
 
 o 
 
 
 1^ 
 
 
 
 
 
 
 
 
 
 
 &g« 
 
 
 — c* 
 
 CO f 
 
 u; 
 
 o 
 
 
 t* 00 
 
 es 
 
 
 iJE^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
M. L. Roxas 
 
 81 
 
 to 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 CO 
 
 CO 
 
 o. 
 
 t^ 
 
 CO 
 
 CO 
 
 CD 
 
 00 
 
 00 
 
 00 
 
 -* 
 
 ■^ 
 
 00 
 
 00 
 
 ,_, 
 
 ,_l 
 
 ,_! 
 
 t^ 
 
 t* 
 
 r* 
 
 
 
 (M 
 
 (M 
 
 
 
 
 (N 
 
 O) 
 
 (N 
 
 
 
 o 
 
 o 
 
 CO 
 
 o 
 
 (M 
 
 00 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 t^ 
 
 
 (N 
 
 O 
 
 
 Ol 
 
 (M 
 
 o 
 
 1 
 
 o 
 
 1 
 
 o 
 
 - 
 
 t^ 
 
 o 
 
 1 
 
 t~ 
 
 o 
 
 T 
 
 o 
 
 1 
 
 CO 
 
 
 o 
 
 s; 
 
 lO 
 
 o 
 
 o 
 
 <N 
 
 o 
 
 o 
 
 M 
 
 CO 
 
 to 
 
 "3 
 
 CO 
 
 CO 
 
 CO 
 
 00 
 
 00 
 
 •* 
 
 ^ 
 
 00 
 
 00 
 
 ,—1 
 
 ,_( 
 
 ,-4 
 
 t^ 
 
 c^ 
 
 r~ 
 
 
 
 C<1 
 
 C^l 
 
 '-' 
 
 '" 
 
 '-' 
 
 CN 
 
 03 
 
 (N 
 
 O 
 
 o 
 
 o 
 
 o 
 
 l^ 
 
 o 
 
 00 
 
 o 
 
 o 
 
 
 T)< 
 
 •* 
 
 CO 
 
 o 
 
 Tt< 
 
 00 
 
 
 >o 
 
 o 
 
 o 
 
 ^ 
 
 M< 
 
 00 
 
 (M 
 
 ■* 
 
 ,_, 
 
 -* 
 
 (^ 
 
 ■ OD 
 
 oo 
 
 
 
 c^ 
 
 fM 
 
 
 '-' 
 
 
 o-l 
 
 C-) 
 
 (M 
 
 
 CO , o o 
 
 Tf 3 o o 
 
 o ~ o o 
 
 + : + 
 
 o W 
 + : + e 
 
 2 3 
 
 ^ IM Q 
 
 a w a 
 
 to 5 
 
 2 & 
 
 o : 
 
 csi ri 
 
 d H 
 
 + c 
 
 2 S 2 
 
 r« 
 
 ^ 
 
 
 d 
 
 M 
 
 
 
 
 
 >. 
 
 !-i 
 
 
 
 f^n 
 
 
 
 
 
 j:5 
 
 o 
 
 
 Dh 
 
 (M 
 
 , , 
 
 
 + 
 
 
 + § S « 
 
 THE JOURX.VL OF BIOLOGICAL CHEMISTRY, VOL. XXVII, NO. I 
 
82 
 
 Huniin Formation 
 
 
 
 
 Urn 
 
 
 
 
 
 
 
 
 
 
 _o 
 
 
 
 
 
 
 
 
 "o 
 
 ei 
 
 
 
 
 
 
 
 V 
 
 _o 
 
 « 
 
 2 
 
 : 
 
 
 
 
 ■^ 
 
 "o 
 
 
 
 
 
 .a 
 
 
 s 
 
 >. 
 
 
 
 
 
 a 
 
 
 a 
 
 .s 
 
 , 
 
 , 
 
 , 
 
 
 E 
 
 
 c: 
 
 u 
 
 ■ 
 
 ' 
 
 " 
 
 
 & 
 
 
 CJ 
 
 , "~ 
 
 
 
 
 
 
 
 2 S 
 
 J 
 
 2 
 
 2 
 
 - 
 
 
 
 
 ^ CJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 b 
 
 
 
 
 
 
 - I^^ll 
 
 
 8 
 
 8 
 
 8 
 
 8 
 
 
 
 H g.a 
 
 
 
 
 '^ 
 
 
 
 
 e 
 
 ToUl 
 N 
 
 found 
 (col- 
 umn 1 
 -(-col- 
 umn 2). 
 
 
 S-: 
 
 <j « 
 
 '^ ^ 3»0 *» M*=^ *^ 
 
 
 
 t 
 
 S§ 
 
 s.i 2 
 
 O^C ••t-iO -Nl-C 
 
 2S.8 2SS.S slag 
 
 
 
 S is 
 
 E 
 
 
 
 
 
 
 
 < -o" 
 
 
 
 
 
 
 
 
 3-5a -iei 
 
 
 
 
 
 
 
 
 O-^'a .5- 
 
 
 
 
 
 
 
 tj 
 
 O 1 -^ 
 
 
 
 
 
 
 
 1 
 
 ° S2 oil 
 
 
 
 
 
 
 
 '.a 
 
 < — - _«, 
 
 
 
 
 
 
 
 § 
 
 
 
 
 CN 
 
 o 
 
 00 
 
 
 ~ Jz'tlill 
 
 
 O 
 
 
 
 o> 
 
 
 o 
 
 E 
 
 •^ 
 
 -r 
 
 n 
 
 CO 
 
 
 J. 
 
 H .SS=.o 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 a 
 
 
 ? 
 
 1 
 
 Hurain N 
 (Kjddnhl). 
 
 1 
 
 s 
 
 d 
 
 s 
 
 o 
 
 
 2| 
 
 8 
 
 d 
 
 
 
 
 
 + 
 
 + 
 
 
 + 
 
 
 8 ': 
 
 
 
 
 
 g 
 
 S 
 
 
 s 
 
 
 + : 
 
 
 
 
 
 
 o 
 u 
 _3 
 
 o 
 
 
 3 
 
 
 § : 
 
 
 
 
 
 "S 
 
 
 
 
 
 u 
 
 
 
 
 c 
 
 
 
 
 
 
 3 • 
 
 
 B 
 
 
 
 i 
 
 1 
 
 
 
 
 •a : 
 
 
 1 
 
 
 s i. 
 
 s^ 
 
 8 C 
 
 s S 
 
 i i 
 
 
 1 
 
 
 CLi " ■ 
 
 o ._ 
 
 c 
 
 is 
 
 , a 
 
 fix 
 
 ° c 
 
 o 
 
 
 
 
 -f 
 
 + £ 
 
 + 8 
 
 + 8 
 
 + w 
 
 
 
 
 
 , 1. 
 
 _: *■ 
 
 i- 
 
 •^ 
 
 
 
 
 C 
 
 E ^ 
 
 IS. 
 
 1^ 
 
 ii 
 
 
 
 
 
 
 lO 
 
 1(5 
 
 
 
 
 
 •c 
 
 •c ^ 
 
 115 ^ 
 
 ■ .■5 — 
 
 
 
 
 
 ~ 
 
 M T 
 
 ZI -fl" 
 
 Z^ I- 
 
 S o. 
 
 
 
 
 6 
 
 d 
 
 d 
 
 d 
 
 d 
 
 
 |P 
 
 
 8 
 
 ?i 
 
 S 
 
 
 S5 
 
 
 s 
 
M. L. Roxas 
 
 83 
 
 
 00 
 
 
 
 
 CO 
 
 
 
 
 « 
 
 g 
 
 
 ^ 
 
 
 
 
 ^ 
 
 
 
 
 M 
 
 t^ 
 
 -:(< 
 
 
 g 
 ^ 
 
 ■^ 
 
 g 
 
 
 
 
 3: 
 
 CO 
 05 
 
 t-. 
 
 g 
 
 CO lO 
 
 d> ci 
 
 CS) CO 
 
 per 
 cent 
 47,31 
 99.4 
 
 S S 
 0. u 
 
 to 
 
 
 
 
 2 
 
 
 
 
 
 
 
 o 
 
 1 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 5: 
 
 
 
 
 5 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 (M 
 
 
 
 
 00 
 
 g 
 
 
 ^ 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 .s s 
 
 CO 
 
 Ocdo 
 
 
 o — 
 
 rM 
 
 
 
 + 
 
 K 
 
 
 
 Fi 
 
 c 
 
 ID 
 
 bD 
 
 
 
 3W 
 
 
 o + 
 + 9 
 
84 
 
 Iluinin Format ion 
 
 ?5 S 
 
 
 1^ 00 5; 
 
 
 
 =z"e55 
 
 c. sS-s4 
 
 o c o 
 
 
 OJ — O M o 
 
 N o -r tc o 
 
 £J si d 2 ^ 
 
 s-£ 
 
 o o — 
 
 0-2 
 
 O 
 
 + K 
 
 
 a . ^ ^ 
 
 + c + 
 
 -r ci 
 
 -^ ^ O 
 
 s 
 
 s 
 
 + 
 
 08 
 
 u 
 
 + 
 
 c 
 o 
 
 2 
 
 3 
 
 
 
 
 ec 
 
 O 
 
 
 
M. L. Roxas 
 
 85 
 
 o 
 
 CI 
 
 
 § 
 
 
 
 
 g 
 
 
 
 
 8 
 
 § 
 
 o 
 
 o 
 
 s 
 
 s 
 
 
 00 
 
 
 
 
 C5 
 
 
 
 
 
 C<1 
 
 
 (M 
 
 
 a o 
 ft o 
 
 
 
 
 
 00 
 
 o 
 
 p. 
 
 G ■ 
 
 
 
 
 
 8 
 
 
 o 
 
 
 
 
 § 
 
 
 
 
 
 o 
 
 5S 
 
 S- 
 
 o 
 
 
 o 
 
 
 
 
 ■* 
 
 
 
 
 
 -^ 
 
 o 
 
 '^ 
 
 o 
 to 
 
 
 o 
 
 
 
 
 § 
 
 
 
 
 o 
 
 o 
 
 o 
 
 00 
 
 
 s 
 
 
 
 
 
 
 
 g 
 
 
 
 
 Tt< 
 
 cc 
 
 CO 
 
 CO 
 
 o 
 
 
 o 
 o 
 
 
 
 
 o 
 
 
 
 
 o 
 o 
 
 o= 
 
 00 
 
 CO 
 
 
 
 
 
 
 
 s 
 
 
 
 
 ^ 
 
 (M 
 
 !M 
 
 c^ 
 
 o 
 
 
 ^ 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 
 
 
 
 
 
 
 " =^ c 
 
 + K 
 
 
 K "Si 
 
 ago 
 aco M 
 ° o ^ 
 
 + + g 
 
 a s o 
 
 bO bC p. 
 
 s 
 
 6J0 
 O 
 
 Kg 
 
 OK 
 
 lO ft 
 
 o o 
 
 + "^ 
 
 a a 
 
 ^ •^' 
 
86 
 
 Ilumin Formation 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 it 
 
 c 
 
 
 
 i 
 
 
 
 s 
 
 
 : 
 
 o 
 
 k- 
 
 
 
 
 o 
 
 
 i 
 
 
 
 Q 
 
 
 
 .Ml 
 
 
 
 . 
 
 
 . 
 
 a 
 
 o 
 
 T3 
 
 
 
 
 S 
 
 B 
 
 
 £ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 "3 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 
 
 
 
 g 
 
 
 5 
 
 
 
 : 
 
 
 - 
 
 - 
 
 
 
 
 - .S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _fc 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 h- 
 
 Hi 
 
 E 
 
 
 c 
 
 ■« 
 
 
 
 O 
 
 
 
 c 
 
 
 ■>f 
 
 P5 
 
 
 
 
 t?5 
 
 
 Total 
 N 
 
 found 
 (col- 
 umn 1 
 -(-col- 
 umn 2). 
 
 Ck 
 
 
 
 
 
 1^ 
 
 * fc 
 
 c 
 
 1 
 
 r u 
 
 ^^°..^ 
 
 »C 
 
 CD t. 
 
 c 
 
 
 o< X |_ -g 
 
 
 E 
 
 
 
 
 
 s 
 
 o a 
 
 ZJ 
 
 g 
 
 § a 
 
 sisg as 
 
 si 
 
 s s. 
 
 g 
 
 
 2§ S.S 
 
 
 .S ^8 
 
 
 
 s 
 
 
 
 s" 
 
 
 
 1 
 
 
 M 
 
 ^ 
 
 
 
 
 o 
 
 A 
 
 ezs| 
 
 1 
 
 
 P) 
 
 
 
 d 
 
 
 
 ^- 
 
 
 d 
 
 c< 
 
 
 
 
 d 
 
 
 < -3 = 
 
 
 
 
 
 
 1 
 
 
 
 1 
 
 
 
 
 
 
 
 1 
 
 
 S^J^as 
 
 Ck 
 
 
 ^ 
 
 
 
 g 
 
 
 
 § 
 
 
 s 
 
 o 
 
 
 
 
 s 
 
 
 U-^a -- 
 
 E 
 
 
 « 
 
 
 
 gi 
 
 
 
 g 
 
 
 - 
 
 ^ 
 
 
 
 
 M 
 
 
 
 
 
 5 
 
 
 
 s" 
 
 
 
 
 
 s 
 
 g 
 
 
 
 
 K 
 
 
 (» 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 E 
 
 
 
 
 
 1^ 
 
 
 
 S 
 
 
 = 
 
 00 
 
 
 
 
 w 
 
 " 
 
 Total 
 
 trat« 
 (Kiol- 
 dahl). 
 
 E 
 
 
 
 
 
 
 
 
 1 
 g 
 
 
 {3 
 
 53 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 .•* 
 
 
 
 
 
 
 c 
 
 
 
 a 
 
 
 
 s 
 
 
 
 B 
 
 
 
 a 
 
 
 :ir^ 
 
 
 
 
 o 
 
 
 
 o 
 
 
 
 c 
 
 
 
 c 
 
 
 
 » 
 
 
 2 
 
 
 
 
 
 
 
 CJ 
 
 
 
 c 
 
 
 
 &.< 
 
 
 
 o 
 
 
 ■§■5 
 
 i 
 
 
 i 
 
 I 
 
 
 s 
 
 u 
 
 
 
 
 8 
 
 r* 
 
 s 
 
 
 
 Sg 6 
 
 
 i^ 
 
 E 
 
 
 c 
 
 a 
 
 
 d 
 
 B. 
 
 
 c 
 
 Q. 
 
 d 
 
 d 
 
 
 
 W O- 
 
 
 i'6 
 
 
 
 
 Cv| 
 
 
 
 CO 
 
 
 
 w 
 
 
 
 £i 
 
 
 
 " ° 
 
 
 
 
 
 CS 
 
 
 
 
 
 
 
 
 
 so 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^ 
 
 
 
 ^H 
 
 
 
 ^^ 
 
 
 
 d 
 
 
 
 1^ 
 
 
 
 
 + 
 
 
 
 + 
 
 
 
 + 
 
 
 
 i 
 
 S 
 
 
 
 8 
 
 ■f 
 
 
 
 
 
 
 S" 
 
 
 
 
 
 
 3 
 
 
 
 + 
 
 S 
 
 + 
 
 O 
 
 3 
 
 
 c 
 
 3 
 
 
 
 
 C 
 
 1 
 
 
 d 
 
 
 
 U 
 
 o 
 
 
 
 q 
 
 
 
 1 
 
 1 
 
 
 c 
 C. 
 
 i 
 
 q 
 
 
 
 
 H 
 
 
 + 
 
 
 
 (N 
 
 ^ 
 
 
 ci 
 
 o 
 
 
 O 
 
 q 
 
 
 
 
 u 
 
 
 
 
 
 
 -f- 
 
 hH 
 
 
 + 
 
 
 
 + 
 
 + 
 
 
 H 
 
 + 
 
 K 
 
 
 
 
 i 
 
 i 
 1 
 
 
 8 
 
 c 
 
 & 
 
 
 i 
 
 8 
 
 c 
 c 
 
 
 1 
 
 
 
 
 5> 
 
 c 
 
 
 
 
 d 
 
 
 
 d 
 
 
 
 d 
 
 
 
 d 
 
 d 
 
 
 
 d 
 
 
 
 |l-- 
 
 
 1^ 
 
 
 
 
 
 
 § 
 
 
 
 s 
 
 "5 
 
 
 
 S 
 
 
 
M. L. Roxas 87 
 
 ence in behavior between tyrosine and phenylalanine. In aque- 
 ous or very weak acid solution arginine, histidine, and lysine 
 evidently react with sugar as indicated b}' the highly colored 
 solutions produced and by the loss of activity of a large fraction 
 of their amino nitrogen. Thus, when arginine plus glucose 
 was boiled in water there was a very deep coloration of the solu- 
 tion (Experiment 39), and at least 25 per cent of the amino nitro- 
 gen became inactive towards nitrous acid. Lj^sine behaved 
 similarly (Experiment 42), 17 per cent of the amino nitrogen 
 becoming inactive towards nitrous acid. Histidine acted like- 
 wise (Experiments 47 and 51), 16.2 per cent of its amino nitro- 
 gen becoming inactive. These facts show that in the cases of 
 histidine and arginine the a-amino nitrogen takes part in the 
 reaction. In the case of lysine it is difficult to establish which 
 amino group is reactive, since at the time the amino nitrogen in 
 the filtrate was determined the temperature in the laboratory 
 was about 35°C. and at this temperature it was found that both 
 the a- and the e-amino group of lysine react with nitrous acid in 
 5 minutes, as may be seen in the amino nitrogen determination 
 of the filtrate (Experiments 40 to 42). It is to be noted that in 
 these cases some loss of nitrogen also took place. It may be 
 that during the reaction some ammonia was given off. 
 
 The result with tryptophane is in agreement with the work of 
 Gortner and Blish' in that a greater portion of the tryptophane 
 nitrogen is converted into humin. The strength of the acid used 
 here and the different procedure followed may account for the 
 difference in the per cent of tryptophane nitrogen foimd in the 
 humin which according to the above named authors was 86 per 
 cent while in these experiments only about 71 per cent was ob- 
 tained. Due to a lack of material it was impossible to repeat the 
 experiment with tryptophane. 
 
 In order to determine which atomic groupings in tyrosine, 
 cystine, and tryptophane were responsible for humin formation, 
 the humin from each one of these amino-acids was dissolved 
 in 0. 1 N alkali and Van-Slyked. It was believed that if the amino 
 groups . in this humin remained mtact they should still give the 
 nitrous acid reaction. The results are as follows: 
 
88 Hiimin Formation 
 
 Humin nitroffon. Reactive with HNO9. 
 mg. nif/. 
 
 Tyrosiiu" 2.360 2.4>5 
 
 Cystine 0.974 0.88 
 
 Tryptophane 13 820 190 
 
 I''ioni these results it must be concluded that in the case of 
 tj'ioisine and cystine it was not the amino group that reacted 
 with sugar to form humin but some other group, probably the 
 (OH) in tyrosine and the (S-S) in the case of the cystine. If this 
 were the case, then the cystine would presumably undergo re- 
 duction before reacting with the sugar. 
 
 In order to determine whether, as Gortner and Blish suggested, 
 the furfurol obtained from sugar was responsible for the reaction. 
 Experiments 54, 55, and 56 were performed as follows: 
 
 Exporinieiit Humin N. Poroeotof 
 
 No. m). tolal N. 
 
 54 0.2 gni. cystine + 2 cc. furfurol + 20 
 
 per cent HCl 7.00 32.0 
 
 .•i.'i 0.2 gin. tyrosine + 2 oc. furfurol + 20 
 
 per cent HCl 8 40 55.0 
 
 .">fi 0.2 gin. nrginine + 2 cc. furfurol + 20 
 
 percent HCl.. 12 7.5 21.5 
 
 These results tend to show that the furfurol formed from sugars 
 under the influence of acids may to a great extent be responsible 
 for humin formation. 
 
 .\s to the effect of the different sugars on the reactive aniino- 
 acid. Kxperiments 20, 21, 22, 29, 38, and 46 show that xylose 
 ant! fructose give higher results than glucose as a rule. This is 
 to In- expected, if it is admitted that furfurol or some other sim- 
 ple aldehyde is the active substance in these reactions. 
 
 DISCUSSION OF RESULTS. 
 
 Some evidence is given which shows that the a-amino groups of 
 argminc, histidine, and tryptophane take part in the reaction 
 with sugars. On the other hand, the a-amino groups of alanine 
 and leucine arc unai>lc to give the same reaction. Glutaminic 
 acid and [ihenylalanine, although giving some humin nitrogen, 
 Ukewisc furnish no indication of reaction. At least for the pres- 
 ent it may be admitted that the humin nitrogen in these cases — 
 
M. L. Roxas 89 
 
 glutaminic acid and phenylalanine — was due to adsorption and 
 not to a reaction. It was also shgwn that in tj'rosine the reactive 
 group is presumably the (OH) and surely not the a-NHs. In 
 cystine, as shown above, the a-amino group remained intact, 
 so that presumably the reaction was with the mercaptan group. 
 The question may then be asked: Why are the a-NH^ groups of 
 arginine, histidine and tryptophane more reactive than those 
 of the other amino-acids? An attempt to explain this difference 
 in behavior of the amiuo-acids towards carbohydrates, based 
 on the present work and on some of the contributions reviewed 
 in the first part of this paper, is here offered. It is generally 
 stated that the properties of a compound are functions of its 
 structure. It is, therefore, to the structure of these amino-acids 
 that we must look for an explanation of then- different behavior 
 towards carbohydrates. The structural formulas of histidine, 
 tryptophane, and arginine are given below. 
 
 CH 
 
 NHz 
 
 /-\ 
 
 / 
 
 N N— H(b) (c) 
 
 NH=C 
 
 H^N 
 
 \(f) 
 
 1 
 
 NH-CH2 
 
 [-C = C— CHo— CH 
 
 \ 
 
 (a) 1 
 
 ( 
 
 COOH 
 
 / 
 
 Histidine. 
 
 NHa— CH— CHj 
 
 
 (h) \ 
 
 
 COOH 
 
 
 Arginine. 
 
 CHo 
 
 CH-COOH 
 
 NH 
 
 (d) 
 
 Tryptophane. 
 
 Several investigators have advanced the idea that humin 
 formation is dependent on the presence of labile hydrogen in 
 the amino-acid molecule (SamueUy, Grindley and Slater, etc.). 
 Evidently judging from the results of the present work the two 
 hydrogens of the a-amino groups of alanine and leucine are not 
 
90 
 
 Ilumin Formation 
 
 labile enough to give condensation products with carbohydrates 
 at least under the conditions of these experiments. 
 
 In histidine, arginine, and tryptophane, however, there are 
 other labile hydrogens (a, b, c, d, c, f). The positions of these 
 labile hydrogens with respect to the a-amino group are very favor- 
 able for ring formation. The reaction with a carbohydrate 
 or finfurol may very well be thought of as taking place as follows: 
 
 Histidine. 
 
 CH 
 
 H 
 
 I 
 C-R 
 
 4- 
 
 /CH; 
 
 ■Xc/" 
 
 N N'^0 h;.NH 
 
 .NH— C-^ v.. 
 
 CH II |\x)OH 
 
 HC=Cy /C 
 
 ^ch/ 
 
 < 
 
 COOH 
 
 CJI O H;NH 
 
 /\ 
 
 NfH \ / x:ooH 
 
 /H 
 
 H 
 
 _^(d.i};nh 
 
 I 
 
 R 
 III 
 
 |^CXX)H 
 
 I 
 
 X/X/oJi o h;nh 
 
 NH --il-' 
 
 IV 
 
 Arginine 
 NH,v/N— CH, — CH, 
 
 J-- ' 
 
 ^ ill \ /CH, 
 
 tiw^ >oh;nhc< 
 
 H— C COOH 
 
M. L. Roxas 91 
 
 The following facts tend to support the idea of ring formation : 
 
 1. The intense color of the products. 
 
 2. Miss Homer-' in her work on the condensation products 
 of tryptophane with aldehydes, speaking of the action of glyoxal 
 on this amino-acid, states: 
 
 "Taking into consideration the necessity of the presence of an oxidiz- 
 ing agent and also the fact that the substance produced is intensely colored 
 it is highly probable that in this reaction, besides the simple aldehyde 
 condensation .... there has also been elimination of hydrogen 
 accompanied by complex ring formation." 
 
 3. The fact that pyridine was obtained by Samuelly from his 
 "melanoidins," was at one time used as an argument to indicate 
 that a pj'ridine nucleus was found in proteins. This idea has 
 been disposed of by Emil Fischer's work on proteins, but the fact 
 remains that pyridine is found in the humin formed from pro- 
 teins. This occurrence may be explained by Reactions II and IV 
 thus; 
 
 yCHU y „ I /Cnd , „ 
 
 4. The action of tjTosinase on tyrosine tends to support the 
 idea of ring formation. Tyrosinase produces coloration with 
 tryptophane but not with indol, skatol, or glycocoU. There- 
 fore, the formation of the highl}' colored product requires the 
 peculiar structure of tryptophane. This formation may be con- 
 sidered as taking place in the manner described above. 
 
 The differences in behavior between histidine, arginine, and 
 tryptophane may again be referred back to the differences in 
 theii- structure. Tryptophane being already a complex com- 
 pound with a benzene and a pyrrol ring may form an insoluble 
 four-ringed compound with furfurol, which is extremely resist- 
 
 -' Homer, A., Biochem. J., 1913, vii, 111. 
 
92 Iliiiiiiii Forinatiftii 
 
 ant to the action of acid. This will explain why tr.vptophane 
 is converted into huniin almost (|uaiitit;itiv<ly. (hi the other 
 hanil furfurol may form with hi.stidiiie and arjj;iiiinu products 
 which are still more or less soluble and in the presence of strong 
 acid may be hydrolyzeil back to the free amino-acids. Thus 
 as in the case of Rlutaminic acid, no formation of a ring com- 
 pound takes place in strong hydrocidoric acid solution. This 
 view will explain why in weak acid or in aqueous solution both 
 histidine and arginine react more readily to form colored products 
 than in strong acid solution. 
 
 It is not claimed that the reaction given above gives the actual 
 structure of the melanin molecule, since no evidence is available 
 to indicate what happens to the rest of the molecule of the amino- 
 acids during the reaction with sugars. This theory on humin 
 formation is given here in the hope that it may serve as a guide 
 for future work on the structure of these compounds. 
 
 No evidence was found in the present work to explain Samu- 
 elly's finding that when the huniin obtained from sugar plus 
 tjTOsine was fused with alkalies, an odor of indol was obtained. 
 It might have been possible that the tyrosine used contained 
 traces of tryjitophane which would explain the production of 
 indol. Likewise the fact that pyrrol was obtained from his 
 "mclanoidins" can be traced back to tiie presence of the trypto- 
 phane nucleus in them. 
 
 Almost all of the experiments recorded in this paper were done 
 with single amino-acids. There was found evidence (Experiment 
 31) to show that the reaction would be different, at least in the 
 case of cystine and tyrosine, if other amino-acids were present 
 in the reaction mixture with sugar. If cystine and proline were 
 boiled together in the presence of glucose and 20 per cent HCl 
 a larRcr amount of cystine nitrogen disappeared in humin forma- 
 tion than when cystine was boiled alone. The same was true 
 when tyrosine and proline were boiled together. It would, 
 therefore, be interesting to study the behavior of mixtures of 
 ilifferent amino-acids when boiled with sugars, both in acid 
 and aqueous solutions. .Vbderhaldcn and Guggenheim'" work- 
 ing with tyrosinase along this same line already concluded that 
 other amino-acids, when present, apparently take pari in the 
 jiroduction of the pigment. 
 
M. L. Roxas 93 
 
 CONCLUSIONS. 
 
 1. Alanine, leucine, phenylalanine, and glutaminic acid may be 
 eliminated as important factors in humin formation, when sub- 
 jected to the treatment used in these experiments. Proline, 
 however, under certain condition.s may b!> involved in humin 
 formation. 
 
 2. The following amino-acids were responsible for humin 
 formation, and in digestions, with 20 per cent HCl plus sugar, 
 the proportion of their nitrogen disappearing was: Tyrosine, 15.0; 
 cystine, 3.1; arginine, 2.33; lysine, 2.62; histidine, 1.84; trypto- 
 phane, 71.0 per cent. 
 
 3. Xylose and fructose were as a rule more reactive than 
 glucose. 
 
 4. Arginine, histidine, and lysine reacted with sugars more 
 readily in weak acid or aqueous, than in strong acid solutions. 
 
 5. Arginine, histidine, and tryptophane reacted with loss in 
 reactivity of their amino nitrogen towards nitrous acid, but 
 tyrosine and cystine reacted without any such loss. 
 
 6. A possible mode of reaction is suggested. 
 
 It is with pleasure that the writer acknowledges his obligation 
 to Professor E. B. Hart, Chief of this Department, for giving 
 him this problem, and for his many valuable suggestions during 
 its execution. 
 
LlbKHKT Ul- i_unon.t->-. 
 
 002 672 082 5 
 
 THE VWAVtRLY PRESS 
 •ALTIMOna. u. •■ *.