Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The terminal groups of the human and bovine [gamma]-globulin Lay, Woo-Pok 1955

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1955_A6_7 L2 T3.pdf [ 3.93MB ]
JSON: 831-1.0106334.json
JSON-LD: 831-1.0106334-ld.json
RDF/XML (Pretty): 831-1.0106334-rdf.xml
RDF/JSON: 831-1.0106334-rdf.json
Turtle: 831-1.0106334-turtle.txt
N-Triples: 831-1.0106334-rdf-ntriples.txt
Original Record: 831-1.0106334-source.json
Full Text

Full Text

THE TERMINAL GROUPS OP THE HUMAN AND BOVINE U-GLOBULIN by WOO-POK LAY A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of Biochemistry We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF SCIENCE* Members of the Department of Biochemistry. THE UNIVERSITY OF BRITISH COLUMBIA August, 1955 ABSTRACT Sanger's dinitrophenyl method has been applied to human tf-globulln. One mole each of aspartic acid and of glutamic a c i d have been i d e n t i f i e d as the free c(-amino terminal residues. In bovine V-golubin, free amino groups are located on fi v e different*amino acid residues, namely, aspartic acid, glutamic acid, serine, alanine and va l i n e . Each i s present i n sub-molar quantity i n d i c a t i n g that the protein Is heterogeneous with respact to N-terminal groups. By a combination of the carboxypeptidase and of the hydrazinolysis methods, one mole each of serine and glycine have been established as the C-terminal amino acids of human tf-globulin. The free carboxyl terminal groups of bovine V-globulln also have been found to be serine and glycine. Carboxypeptidase erperiments indicate that there i s no difs - 3 ; ference i n the release of free amino acid from bovine ^-glo-b u l i n and human V-globulin. The C-terminal sequence of these proteins i s probably the same. In one chain, serine may be followed by leucine and valine. ACKNOWLEDGEMENT The author wishes to express his gratitude to Dr. W.J. Polglase for his interest and advice in con-nection with this work. He also wishes to thank the National Research Council for personal assistance in the form of research grants. HISTORICAL TABLE OF CONTENTS Page 1 EXPERIMENTAL 23 A. Ultracentrifuge Studies . . . . . . 23 B. Identification of N-terminal amino acid residues • 26 I. Preparation of dinitrophenyl derivatives of amino acids • • 26 II. Preparation of N£-2:lj.-dinitrophenyl-L-lysine • 28 III•. Preparation of 2:l|.-dinitrophenol . . . . . . . 30 IV. Preparation of dinitrophenyl -globulin • • • • 30 V. Hydrolysis of dinitrophenyl -globulin • . . • 32 VI. Chromatographic identification of DNP-amino acids • 32 VII. Regeneration of free amino acids from DNP-derivatives 3^ VIII. Quantitative determination of DNP-amino acids • 36 IX. Identification of N^-2tif-dinitropheny-L-lysine, l\l C. Identification of C-terminal amino acid residues by carboxypeptidase . . . . • • • • • lj.1 I. Preliminary studies kl II. Identification by two dimensional chromatography ij.2 D. Identification of C-terminal amino acid residues by hydrazlnolysis kk-I. Qualitative determination kl I I . Quantitative determination . kl DISCUSSION 51 SUMMARY 59 BIBLIOGRAPHY TABLES I. Sedimentation measurement of human 0-globulin (light component). • 21+ II. Sedimentation measurement of human TS-globulin (heavy component). 2$ III. Preparation of DNP-amino acids. 29 IV. A mide content of human and bovine ^-globulin. 31 V. Quantitative determination of N-terminal residues i n human ^-globulin. 39 VI* Quantitative determinationoof N-terminal residue In bovine tf-globulin. lj.0 VII. Quantitative determination of C-terminal residues of human and bovine tf-globulin by hydrzainolysis. 50 PLATES I. Sediment st ion diagram* 27 II. Chromatogram of N-terminal amino acid residue in human ^ -globulin and Chromatogram of N-terminal amino acid residues i n bovine ~6-globulin. 35 III. Chromatogran of free amino acids regenerated from DNP-amino acids from human 0-globulin, and Chromatogram of free- amino acids regen-erated from DNP-amino acids from bovine "8-globulin. . 37 IV. Chromatogram of free amino acids released by carboxypeptidase. l|-3 V. Two dimensional chromatogram of free amino acids released by carboxypeptidase i n human ~6-globulin. h-5 VI. Two dimensional chromatogram of free amino acids released by carboxypeptidase In bovine "6-globulin. lj.6 VII. Two dimensional chromatogram of free amino acids from hydrazinolysis of human """6-globulin. ij.8 HISTORICAL The study of proteins occupies a central place i n biochemistry since protein is the indispensable substrate for the chemical activity of l i v i n g matter. Life i s unique-ly characterized by i t s association with protein. Biolog-i c a l structures contain keratin, which forms hair, feather, wool, n a i l and hoofs; collegen which makes up the large part of tendons, ligaments and other similar connective tissues. In addition structural proteins form the r i g i d framework of most cells and c e l l membranes, which play an essential role in the permeability and metabolism of the c e l l s . Even more important i s the role of protein i n the dynamic aspect of l i f e . A l l enzymes, which catalyze spe-c i f i c a l l y most of the chemical transformations essential for normal cellular function, are either protein alone or pro-tein associated with an additional, or prosthetic group. There are hormonal proteins that are key regulators of met-abolic processes; contractile proteins of muscle fiber which can transform chemical energy into mechanical work; antibody proteins which counteract harmful and foreign agents; and pigment proteins such as hemoglobulin and flavoprotein that serve important carrier functions. That this l i s t can hardly be exhausted emphasizes the v e r s a t i l -i t y and significance of protein. In fact, the simplest biological entity that possesses the faculty of multiplic-i t y , viz., the virus, i s composed entirely of nucleoprotein. 2 No wonder Bergmann terms protein "the chemical requisite for l i f e , " Attention has, consequently, been focused on the problem of the detailed structure of protein in an attempt to provide the chemical basis for the understanding of the manifold variety of physiological functions exhib-ited by these substances* The early years of protein chemistry were be-sieged with extreme d i f f i c u l t y . The large size of the pro-tein molecule renders most of the conventional techniques of organic chemistry useless. The number of atoms in the molecule, particularly carbon and hydrogen, i s so numerous that elementary analysis cannot reveal the exact number with certainty, and even i f this were possible, the data so obtained would not be interpretable. It Indicates only that as a class, a l l protein contain nitrogen. Proteins have no characteristic melting point to establish their identity or purity. Most mysterious of a l l , proteins are extremely susceptible to alternation of environmental con-ditions and easily undergo changes in various properties». Denaturatlon may be brought about by merely shaking an aqueous solution of protein. As a result of such complex-i t i e s , progress was made very slowly and painfully during the period of nearly a hundred years since Mudder introduced the term "protein" i n l61|4« As a group, this is the least understood of the constitutional t r i o - protein, fat and carbohydrate. 3 The work on proteins of the nineteenth century consisted mostly of chance isolation of amino acids from animal and plant materials* It was shown that when pro-tein was subjected to the action Of boiling acids or alka-l i e s , a mixture of simple crystalline compounds could be isolated* These are* with the exception of proline and hydroxyproline, a l l el-amino acids with the general formula: R-CH(NH2) -COOH* Some twenty different animo acids have been identified since 1820 when Braconnot Isolated the f i r s t amino acid glycine from a hydrolyzate of gelatin* It be-came of the greatest importance to discover how these amino acids were combined with each other i n the protein molecule* A. milestone was reached at the turn of the century by the formulation of the peptide theory independently by Hofmeister and Fisher (1)* They postulated that the protein molecule was built up of a long chain of amino acids i n which the <JL-amino group of one amino acid was joined to the 'it-carboxy 1 group of another, and so on, with the elimination of a mol-ecule of water for each amide bond formed* Amino acids were, therefore, recognized as the key building units in the pro-tein molecule just as elementary carbon, hydrogen, etc*, were known to be the building stones of simple organic com-pounds* It must be pointed out that other functional groups of the amino acids might also be available for bond forma-tion i n a protein molecule* The €•amino group of lysine and the side chain carboxy1 group of aspartic acid and glutamic acid might conceivably take part i n branched peptide link-age* The possibility that the hydroxyl group of serine and threonine might condense with the carboxyl group of another amino acid to form an ester linkage must also, be considered* Nevertheless, the peptide theory has served as an exceed-ingly useful working hypothesis ever since i t s inception* To an organic chemist whose principal interest i s the elucidation of the molecular structure of a given com-pound, the f i r s t step i s quantitative elementary analysis« Similarly, structural studies of protein begin with quan-tit a t i v e determination of each amino acid in the molecule* This indeed, has been the main activity of protein chemists during the period of four decades from 1900 to 19lf0* Since, as mentioned above, a l l amino acids are o(-amino acids, many of the properties of an individual acid are common to all© Therefore, the estimation of as many as twenty different amino acids, quantitatively, from a protein hydrolyzate i s no easy task* The various methods discovered were, at best, tedious and required large amounts of pure proteins* In most cases, one had to be content with separation into classes of amino acids* Among the older methods were the fractional d i s t i l l a t i o n of amino acid esters by Fisher (2); extraction of neutral amino acids with n-butanol by Dakin (3); the fractional precipitation of the basic amino acids with phosphotungstic acid and silver ion by Kossel (lf-5); and the partition of total nitrogen in protein hydrolyzate 5 by Van Slyke (6). Technical d i f f i c u l t i e s , however, are so great with these procedures that they seldom account for as much as 80% of the total nitrogen in the protein* A decisive advanc e i n protein chemistry came i n 19ifl - 19kh when Martin and Synge (7-8) introduced the tech-niques of chromatography into amino acid analysis* The principle of chromatographic analysis involves the partition of the solute between two liquid phases, one of which i s im-mobilized by being held on an inert surface. The amino acids are extracted by the solvent at different rates as i t runs down the surface, and are thus separated from each other* By means of paper chromatography i n two directions, i t is possible to separate i n a single operation a l l the amino acids present in a protein hydrolyzate. Inspired perhaps by this success, the following years saw the discovery of nu-merous, ingenious methods, capable of very high resolution and adaptable to minute quantities. Among them, to name only a few, are the ion exchange analysis of Moore and Stein (9); the fractionation by countercurrent distribution of Crelg (10); and the ionophoretic separations i n various forms (11). With the combination of the above methods, i t is possible to account for more than 99% of the total n i -trogen of a protein in the form of amino acids, and hence no appreciable amount of other unknown substances can be pre-sent ln the molecule. Although a precise amino acid composition i s 6 indispensable for a complete description of a protein, i t reveals very l i t t l e as to the nature and function of the pro-tein molecule* Depending upon the arrangement of the amino acid residues i n the polypeptide chain, and the folding of the chain, i t is conceivable that two proteins with the same amino acid content may have entirely different charac-t e r i s t i c s * For example, both |3-lactoglobulin and pepsin have the same molecular weight and qualitatively the same amino acid composition, and yet the former is a typical pro-tein while the latter is an enzyme* Obviously, many of the physical and chemical properties are closely associated with the distribution of functional groups along the chain and on the surface of the protein molecule* The next step i n the elucidation of protein structure would be, therefore, the reconstruction of the sequence of amino acid residues i n the molecule* A. logical point to begin with would be the terminal amino acids i n a polypeptide chain since they differ from other residues in the molecule by their free amino or free carboxyl groups* These functional groups can then be "labelled" by the introduction of an organic radical, which is used for isolation and identification* An ideal reagent should have the following properties (12): 1) It should condense with the terminal residue under mild conditions without secondary affect to the rest of the protein molecule* 2) The condensation product should be unaffected 7 by hydrolysis under conditions sufficient to break a l l pep* tide bonds i n the protein molecule* 3) The condensation product should be readily isolated and identified* The earlier reagents used were those which have been applied to form characteristic acyl derivatives of the amino acid alone* For the amino group, Fisher (13) applied naphthalene-sulfonyl chloride to the free amino group of a peptide and isolated the acid resistent naphthalenesulfonyl amino acid from the hydrolyzate* Benzenesulfonylation (llj.) and ben-zoylatlon (15) have been used i n the same way* Goldschmidt (16) employed hypobromite i n an oxidative reaction in which the terminal acid was oxidized to carbon dioxide and a n i t r i l e with one less carbon. Bergmann (17) found that the free amino group reacted readily with phenylisocyanate, and the phenylhydantoin derivative could be identified from the hy-drolyzate. Barger and Tutin (18) coupled 2 ^ ^ - t r i n i t r o -toluene with the free amino group and isolated the dinitro-t o l y l derivative of the terminal amino acid* It is true that a l l these methods are very useful i n the elucidation of the structure of simple peptides such as anserine, carosine and synthetic peptides* Most of them have, however, numerous disadvantages which limit their usefulness* They require usually rather drastic reaction conditions, or the separation of the labelled product presents d i f f i c u l t i e s , and the yield is usually very low. Consequently, these methods have seldom been applied to proteins. In fact, up to 19k£» the only free 8 amino groups known to be present i n proteins were the fc-amino group of lysine i n gelatin (llj.) and the cd-amino group of phenylalanine In insulin (19)« The year 1945 saw the beginning of a new era i n the structural studies of protein when Sanger Introduced his now classical dinltrophenyl method (20-21). The success hinged upon the discovery'of the incomparable reagent 2:l+-dinitrofluorobenzene which is as useful in protein chemis-try as F i s h e r ^ phenylhydrazine i s in sugar chemistry* In 1910 Abderhalden (22) had described the synthesis of a ser-ies of dinltrophenyl amino acid derivatives by the coupling of the amino acid with 2:lj.-dinitrochlorobenzene* Later he (23) attempted to use this reagent for the identification of the terminal groups from a p a r t i a l hydrolyzate of s i l k fibroin. Due to the presence of anhydride i n the hydroly-zate and the d i f f i c u l t i e s i n separation of the reaction products, he failed to identify the amino acid with the free amino groups* Sanger found that 2:i|.-dinitrochlorobenzene reacts with the free amino group only when heat i s applied to the reactant mixture. This tarings about a certain amount of hydrolysis of the protein. He tried the more reactive analogous fluoro-compound, which turned out to be a great success* 2:ij.-dinitrofluorobenzene was found to react quan-t i t a t i v e l y with the free amino groups of proteins in slightly alkaline mediums at room temperature without fi s s i o n of the 9 peptide bond to give the yellow dinitrophenyl-protein* The dinitrophenyl-protein can be hydrolyzed In boiling d i -luted acid in which the bond between the 2:lj.-dinitrophenyl group and the amino group of protein i s resistant* A l l the amino acids which bear a free amino group in the original protein are now liberated as the 2:i|.-dinItrophenyl amino acid derivatives: The substitution of a dinitrophenyl group at the amino group of an amino acid eliminates dipolar character. The dinitrophenyl amino acids, therefore, behave like typical organic acids* They can be readily extracted into organic solvents and thus separated from other free amino acids pres-ent in the hydrolyzate* The presence of the dinitrophenyl group also confers upon the compound a distinct yellow color which is useful In following the fractionation of a mixture of such compounds by chromatography* Furthermore, dinitro-phenyl amino acids have a characteristic absorption at 350 «yx independent of the amino acid residue which fa c i l i t a t e s their quantitative measurement. Besides reacting with the free dUamino group of proteins, 2:i f-dinitrofluorobenzene also reacts with the £-amino group of lysine, the phenolic group of tyrosine and the imidazolyl group of histidine* , A l l of these mono-substituted compounds are, however, water 10 soluble and do not interfere with the separation of the d i -nltrophenyl derivative of the free oL-amino acid* The intro-duction of the dinltrophenyl group into amino acid, on the other hand, complicates their separation by paper chroma-tography* Due to the strong absorption affect of the aro-matic ring, special solvent systems must be used to prevent ta i l i n g on the chromatogram* Blackburn's isoamyl alcohol-phthalate buffer system (2lj.), and Biserte and Osteux's toluene-monochlorohydrin-pyridine-aramonia mixture (25) are among the most useful solvents. The dinltrophenyl method can be applied not only to the identification of the N-terminal residue but also to the determination of the sequence near the N-terminal* From the dinltrophenyl protein, a series of dinltrophenyl peptides may be isolated* On complete hydrolysis, each dinltrophenyl peptide yields certain amino acids in common and other amino acids that are different for each peptide* For example, Sanger (26) isolated from the B-chain of insulin the follow-ing dinltrophenyl peptides (DNP-peptide): DNP - phenylalanine DNP - phenylalanyl - valine DNP - phenylalanyl - valyl - aspartic acid DNP - phenylalanyl - valyl - aspartyl - glutamic acid. The sequence phenylalanyl - valyl - aspartyl - glutamic acid could thus be ascertained.. An extension of this approach to 11 the examination of the entire mixture of peptides formed upon partial hydrolysis of insulin permitted the elucida-tion of the structure of a large number of fragments, from which the complete picture of the Insulin was eventually reconstructed (27 - 3°)» The success of this method at once stimulated further research for better reagents* An ideal reagent would be one i n which cleavage occurred exclusively at the terminal peptide bond* By repetitive application of the same procedure, the entire sequence of amino acids could be worked out i n a straightforward manner* Bergmann (17), i n -deed, has described the use of phenyl!socyanate for such a purpose in 1-927* Edman (31) found, however, that the ease of the reaction paralleled the ease of ring closure to the hydantoin, and the phenyllsothiocyanate is preferable i n this respect: o o s . o o s \ S o " o It w i l l be noted that the formation of hydantoin is cat-alyzed by acid but does not require the presence of water* 12 By carrying out the cleavage i n an anhydrous, inert solvent (nitr©methane), no other peptide bondsalong the chain would be affected* A selective cleavage is thus obtained through the l a b i l i z i n g effect of the phenylcarbamyl group* The re-sidual peptide recovered can then be subjected to a second cycle* In a modified form suggested by Fraenkel-Conrat ( 3 2 ) , this method has been applied to a number of proteins success-f u l l y . For the determination of the f ree carboxyl amino acid residue in protein, a number of methods are available. Schlack and Kumpf (33) described in 1926 the f i r s t method for the identification of the C-terminal residue by reaction of the benzoylated peptide with ammonium thlocyanate, and subsequent isolation of the thiohydantoin derivative. Abderhalden i3k) coupled benzylamine with the free carboxyl group and identified the benzylamide derivative from the hydrolyzate. Bergmann and Zervas (35) worked out a proced-ure in which the benzoylated peptide was converted, through i t s hydrazide and azide, to benzylurethane, and then hydro-genated to give the C-terminal amino acid as i t s next lower aldehyde. More recently, other methods have been reported; the formation of acylureas by reaction of the. peptide with di-i|.-tolyl-carbodiimide ( 3 6 ) , the reduction of the C-termi-nal residue to an oL-amino alcohol by means of lithium a l -uminum hydride (37) or sodium borohydride ( 3 8 ) , and the anodic oxidation of the C-terminal residue to an amino 13 methoxyl residue that may be hydrolyzed to an aldehyde* Most of these methods, with the exception of the formation of thiohydantoin and the reduct ion to a l c o h o l , have never been appl ied to p ro te in because of various t e c h n i c a l l i m i -t a t i o n s * In 1952, Akabori (39 - W described what seemed to be a simple procedure for the charac te r i za t ion of car-boxyl terminal anino acids i n p ro te in * I t i s based upon the w e l l known fact that acylamino compounds are e a s i l y de-composed by hydrazine in to acylhydrazine and amine* When peptides or proteins are t rea ted with anhydrous hydrazine, a l l the amino ac id res idues , wi th the exception of the free carboxyl te rmina l , are converted to the hydrazide d e r i v -at ives * The C-terminal amino acids are , however, l i b e r a t e d as such i n the reac t ion mixture* o - o o HWrtH-C.HSH-C.n-t HK-CH-C- HH-Ctt-COOH -T Ik NHi-C-H-COOH j**" t O 0 H *" KM The free amino acid released can then be Identified either by precipitation of the hydrazides as their insoluble d i -benzal derivatives or by conversion to the mono substituted dinltrophenyl derivates. This method has been applied successfully to lxysozyme, taka-amylase A, ovalbumin and other proteins. A modified procedure, using ammonolysis instead of hydrazinolysis, has been proposed recently by Carpenter (iH). None of the chemical methods for characterization of the C-terminal residue, however, are as useful as the end-group assay by carboxypeptidase. The gentleness of reaction conditions, the small amount of substrate required, and the ease with which the released amino acids can be Identified are some of the advantages in this enzymatic procedure. The use of carboxypeptidase for this purpose i s based upon the assumption that the specific requirement of the enzyme toward proteins i s the same as that toward synthe-t i c substrates. A delineation of the reaction of the enzyme on model peptides is thus essential for the interpretation of the results on protein (1+2 - 1+3)• Carboxypeptidase i s the pancreatic enzyme 15 demonstrated f i r s t by Waldschmidt-Leitz (ifif) in 1934* and later crystallized by Anson (lj.5)* It i s an exopeptidase which catalyzes the hydrolysis of the C-terminal peptide bond of a wide variety of acyl dipeptides and longer poly-peptides of the following general structure: 0 o R,- t -W-CH -C.-NH-CH-COOV\ The terminal free carboxyl group is absolutely essential for the action of the enzyme as shown by the complete resistance of the peptide amide (lj.6)* The next most important factor i s the nature of the side chain of the C-terminal residue (R-^). It greatly influences the rate of release of terminal amino acids from a polypeptide chain* From the studies of numerous synthetic compounds, i t appears that the most sen-sitive substrates are peptides having C-terminal amino acid residues which bear an aromatic side chain, e.g., phenyl-alanine, tyrosine, and tryptophan. Following in order of decreasing rate are the longer aliphatic residues, and then the small aliphatic amino acids* It has been demonstrated (ij-7) that, In the case of carbobenzoxy dipeptides, phenyl-alanine i s s p l i t off from the C-terrainal position about five thousand times more rapidly than i s glycine* The presence of charged groups in the vicini t y of a susceptible peptide bond, such as peptides with acidic or basic C-terminal res-idues, decreases the reaction rate, while terminal proline or hydroxyproline are for a l l practical purposes entirely 1 6 resistant to enzymic hydrolysis. The nature of the side chain (R2) of the amino acid which contributes i t s car-boxy 1 group to the susceptible bond also influences* a l -though to a lesser extent, the rate of hydrolysis* It has been shown that the rate of release of terminal phenyl-alanine i s considerably decreased by the presence of an ad? jacent glutamyl (lj.8) or prolyl residue (lj . 9 ) . The length of the peptide residue ( R 3 ) does not appear to affect the reaction rate* Studies on synthetic substrates show, therefore, that carboxypeptidase w i l l release terminal amino acids stepwise from the C-terminal end along the chain u n t i l some structure barrier occurs* If, in a polypeptide chain, the amino acids were arranged i n order of decreasing rate of hydrolysis from the C-terminus, the relative rate of their liberation would be related directly to their sequence i n the chain* I f , however, the sequence were inversed at a certain point, the liberation of the preceding amino acid at the point of inversion would become the rate-determining step for the succeeding one, and the following amino acids would be hydrolyzed at the same rate* Carboxypeptidase was f i r s t used for end-group assay in 1 9 3 0 by Grassmann ( 5 0 ) , who identified glycine as the C-terminal amino acid of glutathione. Almost twenty years later Lens ( 5 D introduced i t s use into protein end-group determination. He identified alanine as one of the C-terminal 17 residues of insulin. Having now established the N- and C-terminal se-quence, one can proceed to the determination of the arrange-ment of amino acids within the polypeptide chain. This is done essentially by the characterization of the numerous small fragments obtained from partial hydrolysis of the pro-tein, either by acids, bases or by enzymes. A. picture of the original molecule can then be reconstructed by assembling the fragments like a jig-saw puzzle. It should be pointed out that the number of peptides i n the hydrolyzate i s ex-ceedingly large for even the smallest protein molecule* Furthermore these peptides differ l i t t l e in their proper-t i e s . Hence their separation, l n spite of the modern techniques mentioned above, is extremely laborious and d i f -f i c u l t , and has been successful only i n one or two cases. To duplicate the feat of Sanger with a typical protein 5 to 20 times the size of the Insulin molecule Is obviously im-possible for the time being. The ..numerous peptide frag-ments produced by partial hydrolysis give l i t t l e hope for the reconstruction of the original sequence even i f their characterization has been achieved. The emphasis, however, is upon technical progress. As pointed out by Martin ( 5 2 ) , the amino acid composition of any protein was not known wl th certainty in 1938, and an approximate amino acid analysis of a simple protein was a major task* Today, thanks to the Ingenuity of Sanger, we know the detailed structure of the 18 insulin molecule* Furthermore, the total amount of insulin which he used for the entire work on structure is comparable with the amount that was once required for a single approxi-mate amino acid analysis* The present paper is concerned with the determi-nation of end-groups of human and of bovine tf-globulin by some of the methods mentioned above* Early research in the f i e l d of immunology has shown that antibody resides in the globulin fraction of the immune sera* By studies of the pur,e antibody, isolated from the antigen-antibody precipitate, i t has been established now that antibody Is true globulin and not a component loosely bonded to the globulin molecule as previously assumed. Anti-body globulins di f f e r , however, from the normal tf-globulin, as well as among themselves, by their biological specificity. Each antibody combines only with that antigen responsible for i t s formation* The possibility also exists that a l l tf-globulln fractions have antibody function* tf-globulin i s , therefore, of interest from i t s important role i n immunol-ogical reactions* Valuable information may be obtained from a study of i t s terminal amino acid sequence with respect to the following points: 1) the homogeneity of the "tf-globulin fraction, 2) the theory of antibody formation, and 3) the reactive groups of the antibody-antigen complex* 19 It i t well known that -globulin (Fraction II of Cohnfs scheme) from pooled plasma shows extreme heterogen-eity by a variety of c r i t e r i a such as solubility test, electrophoresis or ultracentrigugation (53 - 5k) • When human "^-globulin is dialyzed against d i s t i l l e d water, a euglobulin Is precipitated leaving the pseudoglobulin in solution. The latter, however, is by no means pure, for on further dialy-sis i t produces more euglobulin. More euglobulin may also be obtained by f i r s t precipitating the pseudoglobulin with a l -cohol or salt, and then redializing the precipitate so ob-tained. In fact, Oncley and co-workers (55) have demonstrated that the subtraction of "^-globulin, obtained by repeated partial precipitation, may vary a thousandfold i n s o l u b i l i t y . Inhomogeneity is also found i n the eletrophoresis apparatus. Human tf-globulin has been shown to have a continuous spec-trum of mobilities of components with an isoelectric midpoint at pH 7»3» With their eleetrophoretic-convection apparatus, Cann and Kirkwood (56) were able to obtain a partial separ-ation of bovine V-globulin Into four subfractions which possess different isoelectric points. In addition immune serum usually contains a new fraction known as T-globulin or tf-globulin ( 5 7 ) , waich migrates between Y- and ^-globulin. Studies with the ultracentrifuge also point to the poly-dispersity of the "tf-globulin fraction. Gncley and co-workers (58) have shown that normal ^-globulin from pooled sera con-sists of a principal component with molecular weight of 7S, and variable amounts of large molecules with sizes varying from 9 to 20S. 20 With a l l these manifestations of heterogeneity, a more precise chemical check on the species of molecule in V-globulin is desirable* Porter (59) found by the dinltro-phenyl method, in contrast to the physiochemical measurement, that rabbit tf-globulin has only alanine as Its N-terminal amino acid* Van Vunakis (60) showed subsequently, however, that at least seven different terminal amino acid residues, each in fractional amounts, are present in human V-globulin. More recently, McPadden and Smith (61) reported only three N-terminal residues, i.e., glutamic acid, aspartic acid and serine, in the same protein. Putnam (62) confirmed this finding, but, in addition, found small amounts of an un-identified DNP-derivative in DNP-human ^-globulin. Smith ( 6 l ) reported five free N-terminal residues i n bovine ^-globulin. These results suggest that both human and bovine "^-globulin may be heterogeneous with r espect to terminal amino acid residues. Three modern theories have been proposed for the formation of antibody, Haurowitz and Breinl (63) advanced the theory that the antigen molecules interfere with the for-mation of globulin by the normal protein synthesizing, tem-plate with the result that the normal process of globulin formation is modified and a globulin of a different shape, i.e., the antibody, is produced. Burnet and Fenner (61+) argued, however, that the protein synthesized system i t s e l f is modified. The antigen acts as a sort of inducer to the enzyme system so that a new globulin Is formed by adaptive 21 enzymes. Pauling (65), however, assumed that the formation of antibody involved merely the refolding of the preformed normal "^-globulin molecule in the presence of the antigen into a new configuration complementary to that of the an-tigen. Pauling's hypothesis, therefore, differs from the f i r s t two in that It implies no difference i n amino acid composition and sequence in the molecule between normal tf-globulin and immune tf-globulin. An important piece of evi-dence in establishing the Interrelation between normal "tf-globulin and specific antibody was obtained by Porter (59), who compared the N-termlnal sequence of the two molecules* He found that both normal rabbit "^-globulin and antioval-bumin (hen's) possess the same N-terminal pentapeptide se-quence, namely, alanyl-leucyl-valyl-aspartyl-glutamyl. This result has recently been confirmed by Smith (66), who com-pared normal rabbit "^-globulin with different antlpneumo-coccal antibodies. Thus, normal Tf-globulin and specific antibodies do not differ i n the arrangement of amino acid i n the peptide chain, at least for a portion of molecule. The result seems in line with Pauling's theory that antibody arises from an "unspecified" globulin by refolding of the peptide chain. As to the reactive groups of the antibody-antigen complex, very l i t t l e is known about the forces that are i n -volved in such a complex (67). They are, in a l l probability, extremely complicated. Electrostatic attraction of a charged group, van der Waal's forces, hydrogen bonding and steric 22 hindrance may a l l play a part* Landsteiner has found that polar groupings, particularly carboxyl groups, sulfonic acid radical or basic quanternary ammonium groups, deter-mine the serological specificity of synthetic antigens. The surface of globulin probably also possesses many polar residues. It seems lik e l y that the main forces in point of strength are the coulombic forces between charged groups* Nothing is known about what these groupings are. Porter (59) compared the amount of unreacted imidazolyl group of h i s t i -dine in normal rabbit ^-globulin, rabbit antiovalbumin and ovalbumin-antiovalbumin precipitate. He found this group In the specific precipitate failed to react with 2:lj.-dinitrofluorobenzene to the same extent as in the normal Y-globulin or antiovalbumin. This seems to imply that the imidazolyl group of histidlne forms part of the combining group of the antibody-antigen complex* 23 EXPERIMENTAL  Material The human tf-globulin preparation (Lots 3 3 - 2 and 3 3 - 3 ) was kindly supplied by Connaught Laboratories, University of Toronto. Except for dialysis to remove gly-cine in the buffer, i t was used i n the dinitrophenyl and the carboxypeptidase experiments as such without further purification. In the hydrazinolysis experiment, the dialyzed protein was precipitated with ethanol at 0°C and dried l n an Abderhalden apparatus at 78°C for several hours. The bovine "tf-globulin powder (Fraction II) was a commercial product obtained from the Armour Company. The carboxypeptidase was a three time recrystal-line suspension obtained from Armour Laboratory. A.. Ultracentrlfuge studies. Human "tf-globulin was dissolved in 0.1fj>M-NaCl, and made up to 1 per cent protein concentration. The solu-tion was transferred to an 0.8mm analytical c e l l , and run at 5 0 , 7 4 ° r.p.m. i n a model E Spinco Ultracentrlfuge. Aver-age temperature of the rotor during the run was 21.2°C. Pictures were taken at 8 minute intervals during the 2^-hour run starting when the rotor reached f u l l speed. 2k TABLE I Sedimentation measurement (Light component) Magnification factor (M.P.) • 2.19 Distance between reference edge and center of rotor (R) = 5»725 cm. Distance between peak and Picture reference log X No. edge (D). D/fy[.F.. X=R+D/M.P. 10 Time (min.) 1. 0.12 0.24 0 .055 5.780 0.7619 8 2. 0.110 1 - - 5 . 8 3 5 0.7660 16 3. 0 .35 0.160 5.885 0.7698 24 0 .50 0.228 5.953 0.7747 32 5 . 0.59 0.269 5.994 0.7777 40 6. 0.71 0.81 0-321* 6.049 0.7817 48 7. 0.370 6 .095 0.7850 56 8. 0.92 0.420 6.145 0.7885 64 9 . 1.06 0.484 6.209 0.7930 72 10. 1.18 0.539 6.264 0.7969 80 11. 1.30 0 .594 ...&-0565O 6.319 0.8001 88 12. 1.1*2 6.375 O.8045 96 13. Ik. 1 .55 0.708 6.433 O.8084 104 1.69 0.772 6.497 0.8127 112 1 5 . 1.81 0.827 6.552 0.8164 120 16. 1.93 0.882 6.607 0 .8200 128 17. 18. 2.09 0 .955 6.680 0.8248 136 2.21 1.020 6.745 0.8290 144 25 TABLE II Sedimentation measurement (heavy component) Distance between peak and Picture reference log X No. edge (D). D/M.F. X=R+D/M.F. 10 Time (min.) 0.1+39 6.16k 0.7899 1*8 0.534 6 * 2 5 9 0.7965 56 0.598 6.323 0.8009 61+ 0.707 6.1+32 0,8081+ 72 0.790 6.515 0.8139 80 0,876 6.591 0.8189 88 0.961; 6.689 0.825l(. 96 1.055 6.78O 0.8312 101+ I . 150 6.875 0.8372 112 1.2L0 6.965 0.81+29 120 6. 0.96 7. 1.17 8. 1.31 9. 1.55 10. 1.73 11. 1.92 12. 2.11 13. 2.31 Ik. 2.52 15 . 2.71 26 The sedimentation constant was calculated by the following formula: G £ 1t,_- to Where W is the angular velocity of rotation; X 2 and X^ are the distance from the axis of rotation to a point in the c e l l where the protein solution has the maximum concen-tration gradient, i,e«, the boundary, at time tg and t^ respectivelyi. Two peaks were shown on the sedimentation diagram* A major component, consisting of about three-fourths of the protein, had a sedimentation constant of 6 * 7 S . A minor component was made up of heavier material ( 1 0 * 3 S ) . B. N-Terminal amino acid residues. I* Preparation of dinitrophenyl derivatives of amino acid. A l l amino acids, with the exception of the basic amino acids and tyrosine, can be converted into their DNP-derivative by the same procedure (20 - 21). A typical ex-ample is given below: DL-Aspartlc acid (0.k2g, 3 millimoles) was dis-solved in a 10 per cent sodium bicarbonate solution (lk ml.). To this was added an ethanol solution (28ml.) con-DIAGRAM 1. Sedimentation diagram of human ^-glo b u l i n i n the u l t r a c e n t r i f u g e . The protein concentra-t i o n was 1 per cent. The f i r s t picture (on the l e f t ) was taken 5 l minutes after the max-imum speed (50,7^0 r.p.m.) was reached; the subsequent pictures were taken at 8 minute i n t e r v a l s . The arrow indicates the d i r e c t i o n of r a d i a l migration. 28 taining 2:h-dinitrofluorobenzene ( l . l g , 6 millimoles). The mixture was then shaken for two hours at room tem-perature. At the end of the reaction, the solution was concentrated under reduced pressure to remove ethanol. The residue was dissolved i n water, and the excess dinitro-fluorobenzene removed by extraction with ether. Upon acidificat ion of the aqueous solution with diluted hydro-chloric acid, a yellow precipitate of DKP-DL-aspartle acid was obtained. It was recrystallized twice from hot water. Yield? O.L5g., mp., 196°C. II. Preparation of N^-2:k-dinitrophenyl-L-lysine. Since lysine has two amino groups, both of which can react with dinitrofluorobenzene, the ot-amino group must be protected i n the preparation of e-DNP-L-lysine ( 2 2 ) . L-Lysine (0.5g») was dissolved in water (10 ml.). Gupric carbonate was added slowly to the boiling lysine solution. Excess cupric carbonate was f i l t e r e d off from the dark blue solution. The residue was washed with a small amount of water. To the blue f i l t r a t e , excess sodium bicarbonate (ca.l .5g«) was added, followed by an ethanol solution (20 ml.) containing 2:k-dinitrofluorobenzene (1.5 g»)» The mixture was shaken for two hours at room temperature. The yellowish green precipitate was f i l t e r e d off, dissolved i n water (5 ml.). A sufficient amount of N hydrochloric acid was added to produce a clear solution. 2 9 TABLE III. Preparation of DNP-Amlno acid. DNP-Amino acid, DL-Alanine DL-Aspartic acid L- Glut ami c ac i d Glycine L-Hydroxy proline DL-Iaoleuclne DL-leucine L-lysine (Di-DNP) DL - Methionine DL - Phenylalanine L-Phenylalanine L-Proline DL-Serine DL-Threonine DL-'^'ryptopan This work. 1 7 2 - 1 7 3 1 9 6 syrup 2 0 3 syrup l k 7 - l k 8 1 2 5 - 1 2 7 168 1 1 7 2 0 8 . 1 8 6 - 1 8 7 1 3 6 - 1 3 7 197 1 7 7 - 1 7 9 1 7 5 - 1 7 6 DL-Tyrosine (DI-DNP) 1 7 3 - 1 7 k DL-Valine 1 8 0 m.p. ( 9 6 ) Literature. L 1 7 8 ( 2 2 ) , 1 7 3 ( 6 8 ) 1 9 6 ( 2 1 ) , 1 8 9 ( 6 8 ) 2 0 5 ( 2 2 ) , 193 (68), 2 0 3 (69) 1 1 6 ( 2 2 ) , 1 6 8 - 1 7 2 ( 6 8 ) , 1 7 k 1 7 5 (69) 2 0 3 ( 2 2 ) , 1 3 3 ( 6 8 ) l k 6 ( 2 2 ) , 1 7 3 ( 6 8 ) , 1 7 0 -1 7 1 ( 6 9 ) 1 1 7 ( 2 2 ) , 1 0 7 - 1 1 2 ( 6 8 ) 2 0 k - 2 0 6 ( 6 8 ) 1 8 6 ( 2 1 ) , 1 8 9 ( 6 9 ) 1 3 7 ( 2 2 ) , 1 3 8 ( 6 9 ) 199 ( 2 2 ) , 1 8 6 - 1 8 8 ( 6 8 ) , 2 0 0 - 2 0 2 ( 6 9 ) 1 5 2 ' ( 2 2 ) , 1 7 8 ( 6 9 ) L 1 7 5 ( 2 2 ) , 196 - 1 9 8 ( 6 8 ) , 2 2 1 (69) 8 k ( 2 2 ) , L92-98 ( 6 8 ) , 1 7 8 1 8 2 . (69°) 1 8 5 ( 2 2 ) , 1 8 3 ( 6 7 ) , 1 8 k ( 6 8 ) _ 30 Hydrogen sulfide was bubbled through the ice-cooled solu-tion for two minutes to precipitate the copper Ion, A, trace of charcoal was then added, and the warm solution f i l t e r e d immediately. The orange f i l t r a t e was concen-trated to a syrup under reduced pressure* Upon the addi-tion of water, a yellow substance crystallized* This product was recrystallized from 20 per cent hydrochloric acid. Yield: 0.54 S«» m.p., l87°C. III. Preparation of 2;4-dinitrophenol. 2:4-dinitrofluorobenzene (0.4 g.) was refluxed In boiling 5 per cent sodium solution for ljj hours. Yellow crystals began to precipitate as soon as the orange solution was cooled. The precipitate was f i l t e r e d off and dissolved in a small amount ofiwater. 2:4-dinitrophenol precipitated on acidification with H hydro-chloric acid. It was recrystallized from ethanol-water. Yield: 0.28 g., m.p. 110 - 112°C. IV. Preparation of dinltrophenyl 2f -globulin ( 7 0 ) . Human o'-globulin (0*4 g.) and an equal weight of sodium bicarbonate were suspended in water ( 4 ml.). A solution of 10 per cent (W/V) 2:4-dinitrofluorobenzene i n ethanol (8 ml.) was added, and the mixture shaken for two hours at room temperature. At the end of the reaction period, the yellow precipitate of dinltrophenyl ^-globulin 31 was collected by centrifugation, and washed with water (3X), ethanol (5X), and f i n a l l y with ether (3X). The product was dried i n air for two days before being used in further experiments. Yield: 0 . 4 8 g . Bovine tf-globulin was prepared i n a similar way. Prom 0 . 5 g . of bovine iJ-globulin, a yield of 0 ,55 g« of dinltrophenyl "tf-globulin was obtained. The amount of protein i n the dinltrophenyl tf-globulin was estimated by determination of the amide content of the air-dried preparation. Amide determin-ation was carried out after heating the sample for four hours i n a sealed tube with 2N hydrochloric acid at 100°C. The hydrolyzate was neutralized with 2N sodium hydroxide to thymolphthalein, and d i s t i l l e d In the micro-kjeldahl apparatus• TABLE IV Amide content of human and bovine ^-globulin. Amide Content Human "2 '-globulin Bovine "o'-globulin Protein M P -Derivative T>NP-Protein 1 0.992$ 0.780$ 0.731$ 1.024$ 2 0.984 0.812 0.725 0.987 3 1 .025 0.787 0.752 1.000 Average 1.000 0.793 0.738 1.010 0 32 The results Indicated that 100 mg, of DNP human Y-globulin contain 79*3 mg* of human ^-globulin, and that 100 mg* of DNP-bovine Y-globulin contain 73*07 mg. of bovine Y-globulin. V. Hydrolyais of dinitrophenyl Y^globulln* Both DNP human Y-globulin and DNP bovine Y -globulin were hydrolyzed i n the same way: DNP-Y-Globulin (100 mg.) was hydrolyzed complete-ly by refluxing i n an oil-bath for sixteen hours with 6N_ hydrochloric acid (10 ml.). The hydrolyzate was cooled and transferred into a separatory funnel, diluted twice with water, and extracted with ether (3 X 15 ml.). The ether ex-tract was washed with a small amount of 2$ hydrochloric acid and evaporated to dryness. The fraction contains a l l the mono-substituted ether-soluble DNP-amino acids. The aqueous solution, which contained the ether insoluble DNP-amino acids was also evaporated to dryness under reduced pressure for further studies*. VI. Chromatographic identification of DNP amino acids. During the hydrolysis of DNP-proteins, dinitro-phenol was formed as a decomposition product* Since large amounts of this substance interfere with the chromatographic 33 separation of a mixture of DNP-amino acids on paper, i t must be removed prior to identification. The cold finger sublimation technique of Mills (70) was used: The ether extract of DNP-V-globulin was trans-ferred to the suction tube of the sublimation apparatus and concentrated to dryness. The inner tube was then f i l l e d with dry ice and ethanol. The apparatus was evacuated with a water-suction pump, and held at 100°C for l£ hours, during which time a large part of the dinitrophenol was removed from the ether extract. For qualitative analysis, chromatography on buffered paper, according to the method of Blackburn ( 7 1 ) , was used. Whatman No. 1 paper was soaked i n potassium acid phthalate buffer at pH6.0 (50 ml. of 0.2M potassium acid phthalate plus 45*5 ml* of 0.2N. sodium hydroxide, d i -luted to 200 ml. with d i s t i l l e d water), and dried at ioom temperature. The solvent, tertiary amyl alcohol, and 10 per cent ethanol In butanol were also saturated with the same buffer. The chromatogram was allowed to equilibrate for several hours, before the solvent was introduced into the trough at the top of the Jar. (a) DNP-Human tf-globulin. The residue from the ether extract, after sub-limation, was dissolved i n a small amount of acetone (0.5ml.) 34 and applied to the buffered paper. Authentic samples of DNP-amino acids were run along side. With tertiary amyl alcohol as solvent, four yellow spots were located on the chromatogram at R f 0*1, 0.14, 0*69 and 0.91. The f i r s t two spots corresponded to those of DNF-aspartic acid and of DNP-glut'amic acid, respectively. The third spot was identified as 2:4-dinitrophenol. The last spot, running at the solvent front, was probably 2:4 dinltroaniline, an-other common decomposition product of the hydrolysis* (b) DNP-Bovine V-globulin. On the paper chromatogram with tertiary amyl alcohol as solvent, five spots were observed i n addition to dinitrophenol. They were, in order of decreasing Rf: DNP-valine (0.73)» DNP-alanine ( 0 . 6 l ) , DNP-serine (0.35), DNP-glutamic acid (0.12), and DNP-aspartic acid (0.07). VII. Regeneration of amino acid from i t s DNP-derivatlve. In order to confirm the above findings, the free amino acids were regenerated from their DNP-derivatives by the method of Mills (72): The residue from the ether extract was dissolved i n saturated barium hydroxide solution, and hydrolyzed i n a sealed tube at I4.O - 150°C for two hours. After cooling, the barium ion was precipitated with a piece of dry ice and the precipitate removed by centrifugation. The supernatant liq u i d was taken to dryness and dissolved i n a drop of water. (A) o 0 o o O o o z z < O i Q_ Z Q I o I Q . Z Q (B) o o O 0 0 0 o o o 2 CL z Q Q . z o z z 3 Q . < i Q_ Z Q o o o So CHROMATOGRAM 2: Chromatogram of the DNP-derivatives of N-t e r m i n a l amino a c i d s . (A) DNP-amino a c i d obtained from DNP-human Tf - g l o b u l i n . (B) DNP-amino a c i d obtained from DNP-bovine tf-globulin. Both chromatograms were r e -solved i n t e r t i a r y amyl a l c o h o l on Whatman No. 1 paper p r e v i o u s l y sprayed w i t h phathalate b u f f e r , pH6. 36 * The amino acids were identified by paper chromatography using butanol-acetic acid-water (4*1*5), or phenol-water in the presence of potassium cyanide* (a) DNP-Human Y-globulin* Glutamic acid (R f 0,22) aid aspartic acid (0*14) were identified on a paper chromatogram using butanol solvent* (b) DNP-Bovine TC-globulin. Since the butanol solvent could not separate serine and glycine satisfactorily, phenol saturated with water was used i n an atmosphere of 0*3 per cent ammonia, and in the presence of a trace of potassium cyanide, Valine (R f 0*73), alanine (0*57)» serine (O.38), glutamic acid (0*21) and aspartic acid (0*14) were identified* VIII* Quantitative determination of N-terminal  residues* For quantitative measurement, column chromato-graphy on buffered celit e 545 by the method of Perrone (73) was used* The celite was ground in a mortar with suitable buffer solution, and enough water-saturated sol-vent was added to give a f l u i d slurry* The mixture was poured into a glass tube (lXl5cm^ having one end closed with a rubber stopper* The mixture was then thoroughly homogenized by rapid strokes of a perforated metal plun-ger* A small layer of celite was pressed onto the bottom (A) (B) O 0 o o o o o o o o o o o o Q. Z 3 3 _l < * Q. > < 3 a z co - i UJ 3 < o «o CHROMATOGRAM 3: Chromatogram o f f r e e amino a c i d s regen-e r a t e d from DNP-amino a c i d s . (A) human tf-globulin. The chromatogram was r e s o l v e d i n b u t a n o l - a c e t i c acid-water ( k : l : f ? ) . (B) Bovine ^ - g l o b u l i n . The chromatogram was developed w i t h phenol-water (8:2 W/W) i n the presence o f 0.3$ ammonia and a t r a c e o f potassium c y a n i d e . 38 by means of a slow stroke.followed by rapid strokes of the plunger to homogenize the rest of the suspension* This process was repeated until a column of suitable height had been prepared* After removal of the stopper at the bottom and of excess solvent at the top, the column was ready for operation* (a) Human tf-globulin. DNP-Human tf-globulin (200 mg*) was hydrolyzed in 6j£ hydrochloric acid (5 ml.) in a sealed tube at 100°G for twenty-four hours. The hydrolyzate was diluted twice with water, and extracted four times with ether (20ml.)• The combined ether extract was evaporated to dryness, dis-solved i n a small amount of water-saturated chloroform-ether mixture (50:50 v/v), and applied to the top of the chloroform-ether column with a bent-tip dropper* The c o l -umn was made up of 2g* of ce l i t e , buffered With 0.5 ml* of phosphate-citric acid buffer at pHlj. (0.2M-Na2HP0l|.+0.2M-c i t r i c acid). The column was then developed with water-saturated chloroform-ether solvent* Three bands separ-ated on the column* Each of the bands was eluted, and the eluted evaporated to dryness, dissolved i n 1 per cent s>dium bicarbonate (50 ml.), and the optical density com-pared with a standard DNP-aspartic acid solution at 350m^ ti. using a Beckman Model B spectrophotometer. The yellow b i -carbonate solution was then acidified with diluted hydro-chloric acid, and extracted with ether to recover the DNP-39 amino acid for identification by paper chromatography. The fastest running band was shown to be 2:k-dinitrophenol, the second one was DNP-glutamic acid, and the third one was DNP-aspartic acid* TABLE V. Quantitative Determination of N-Terminal Amino Acid Residues i n Human #-Globulin. DNP-Amino acid. Optie* at 35< i l density ) Myj. Mole of amino acid per mole protein (M.W. 160,000) 1 2 3 1 2 3 Aver-age. Glutamic acid. Aspartic .acid. 218 173 209 157 212 180 0.7 0.56 0.67 0.51 0.68 0.55 0.68 0.55 (b) Bovine -globulin. DNP-Bovine ^-globulin (500 mg.) was hydrolyzed 't i n 6N.-HC1 (5ml.) at 100°C for twenty-four hours i n a sealed tube. The hydrolyzate was diluted with two volumes of dis-t i l l e d water and extracted with ether (5X20 ml.). The ether extract was dried, dissolved i n a small amount of chloroform-ether mixture (50:50 V/V), and applied to the 4 0 chloroform-ether column buffered at pEl+. Pour bands separated. They were, i n order of the i r appearance at the bottom of the column, ( 1 ) fast-moving DNP-amino acids and dinitrophenol, (2) DNP-glutamic acid, ( 3 ) DNP-serine, and ( 4 ) DNP-aspartic a c i d . The fast-running band was evap-orated to dryness, dissolved i n a l i t t l e chloroform, and re-chroraatographed on a chloroform column buffered at pH6 ,5 ( 0 . 2 5 M phosphate buffer) to separate dinitrophenol from DNP-valine and DNP-alanine. Each of the elutes from t h i s chromatogram was taken to dryness, dissolved i n 1 per cent codium bicarbonate ( 2 5 ml.), and the o p t i c a l density read at 350 mu. TABLE VI. Quantitative Determination of N-Terminal Amino Acid Residues i n Bovine ft-Globulin. DNP-Amino acid* O p t i c a l density at 350 mp. Mole of amino acid / mole of proteino Aver-age. 1 2 1 2 Aspartic acid 0 . 1 1 2 0 . 1 0 2 0 . 0 7 9 0 . 0 7 1 0 . 0 7 5 Serine 0 . 1 1 4 0 . 0 9 5 0 . 0 8 0 . 0 6 7 0 . 0 7 4 Glutamic Acid 0 . 1 5 4 0 . 1 3 2 0 . 1 1 0.092 0 . 1 0 1 Alanine 0 . 0 9 8 0.109 0 . 0 7 0 . 0 7 6 0 . 0 7 3 Valine 0.192 0 . 1 5 7 0 . 1 3 0 . 1 1 0 . 1 2 kl (Optical density of a standard DNP-aspartic acid solution containing 20 va.jx was 0 . 3 1 2 ) . IX. Identification of e-DNP-^lysine. The aqueous solution of the DNP Tf-globulin hy-dro ly sate was evaporated to dryness under reduced pressure* The residue was dissolved in a small amount of water, and studied on paper chromatography with 10 per cent ethanol in butanol as the developing solvent* A spot corresponding to 6-DNP-lysine (Rf O.ij.8) was observed in both human and bovine "tf-globulin. C. Identification of C-terminal amino acid residues by  carboxypeptidase. I. Preliminary studies* Since carboxypeptidase may be contaminated with •Q traces of chymotrysin, the enzyme was f i r s t incubated with diisopropyl fluorophosphate which is a specific inhibitor P for chymotrysin but does not affect carboxypeptidase (7k)• A The diisopropyl fluorophosphate was kept as a 0*1M sol-ution in isopropanol. For use, i t was brought to 0.00i(M by the addition of 0*1M phosphate buffer at pH8. In a typical experiment, carboxypeptidase sus-pension (o*l ml. containing 0.81|. mg. of enzyme hb.trogen) k2 was run into the inhibitor solution (2 ml*) and incubated at 37°C for fifteen minutes* The solution was then added to the human tf-globulin substrate (160 mg* i n 5 ml. of 0.15M sodium chloride, adjusted to pH8), and the mixture was incubated at 37°C. Samples (1ml.) were withdrawn at various times* The enzyme was inactivated by increasing the acidity of the solution to pH3* Free amino acids i n the sample were separated from the large molecular weight materials by shaking the solution for one hour with Amber-l i t e IR-120 (H* form). The supernatant solution was de-canted, and the resin was washed three times with d i s t i l l e d water* The amino acids were then eluted from the resin wl th $M ammonia* The elutes were evaporated to dryness under reduced pressure, and identified by paper chromato-graphy using butanol-acetic acid-water (k:lt£) as solvent. Results from this experiment showed that leucine, valine, threonine (and/or glutamic acid), and glycine (and/ or serine) appeared in the f i r s t five minutes of proteolysis* II . Identification of C-terminal groups by two  dimensional chromatography. Since many amino acids were released by carboxy-peptidase in a short time interval, and the butanol sol-vents were unable to separate the pairs of amino acid threonine-glutamic acid and glycine-serine, two dimensional chromatography must be used to establish their identities* k3 o o Q 8 LEU ILEU o o o O o o o o TRY VAL TYR o o 0 OOO O o 0 ooCD ALA THR-GLU GLY - SER ASP o o o o LYS CHROMATOGRAM k: Paper chromatogram of the free amino acids l i b e r a t e d from human ^ - g l o b u l i n by carboxy-peptidase at various time i n t e r v a l s . The incubation time i s given along the absissa; a mixture of known amino acids was run along side. The chromatogram was developed with butanol-acetic acid-water ( k : l : £ K 4 4 The same experiment as described above was repeated, and •the amino acids liberated were studies by two dimensional chromatography, using as the f i r s t solvent 80$ phenol i n the presence of 0*3$ ammonia and a trace of potassium cya-nide, and, as the second solvent, butanol-acetlc acid-water ( 4 : 1 : 5 ) . (a) Human -globulin. The chromatogram revealed nine ninhydrin< spots during five minutes of proteolysis corresponding to the following amino acids: aspartic acid, glutamic acid, ser-ine, glycine, threonine, alanine, lysine, valine and leucine* Four of these amino acids, namely, leucine, valine, serine and glycine were present in the highest concentration as judged by the intensity of their ninhydrine color. The same nine amino acids were released from human ^-globulin subjected to one hour proteolysis with car-boxypeptidase. (b) Bovine tf-globulin. Four amino acids, namely, serine, glycine, valine and leucine were liberated during five minutes of hydrolysis by carboxypeptidase. In one hour, the following seven anino acids appeared: glutamic acid, serine, glycine, ala-nine, lysine, valine, and leucine; with serine and glycine present in the highest concentration. D. Identification of C-terminal amino acid residues by hydrazlnolysis. KB O O o o CD G L Y H h-> > 5 i Q O < 8 < * PHEWOL-WATER ( 8 : 2 , W/W ) CHROMATOGRAM 5: Chromatogram of free amino acids released from human Y - g l o b u l i n by carboxypeptidase. The incubation time was one hour at 3 7°C Lexicine, valine, serine and glycine gave the most intense ninhydrin spots, Phenol-water was used as the f i r s t solvent i n the presence of 0.3$ ammonia and a trace of potassium cyanide. 46 R o o •LY • EN / \ > > I -< O < 111 % • _ i o 3 m R f - P H E N O L - W A T E R (8 = 2 , W / W ) CHROMATOGRAM 6; Chromatogram of free amino acids released from bovine "^-globulin by carboxypepti-dase » The incubation period was f i v e minutes at 37°C. Phenol-water was used as the f i r s t solvent i n the presence of 0.3$ ammonia and a trace of potassium cyanide. kl I. Qualitative determination. -Globulin (160 gm.) was hydrazinolyzed with an-hydrous hydrazine (ca. 1 g.) at 100°C for six hours i n a sealed tube. The reaction mixture was evaporated i n vacuo over concentrated sulphuric acid to remove excess hydrazine. The residue was then dissolved i n 10 ml. of water and passed through a column (1 x 15 cm.) of Amberlite IRC-50 (regener-ated by washing with 2N sodium hydroxide, water, 2N hydro-chloric acid, and water) at the rate of about 1 ml. per minute. The column was then washed with water, and a total of 20 ml. of effluent was collected. The effluent was evaporated to dryness, and analysed chromatographically. (a) Human ^-globulin. Two dimensional chromatography showed the appearance of five ninhydrin^ spots, namely, serine, gly-cine, glutamic acid, threonine and alanine. However, the f i r s t two spots had about ten times the intensity of the others. (b) Bovine ^-globulin. Two dimensional chromatograms revealed the same amino acids as observed with human Y-globulin. II. Quantitative determination. The effluent from the ion-exchange column was k6 Cf _THR BLY SER > > cr UJ < 5 o o o 2 < CO — PHENOL - WATER ( 8 = 2, W/W) CHROMATOGRAM 7: Chromatogram of the reac t i o n products re-s u l t i n g from hydrazinolysis of human tf-glo b u l i n . Glycine and serine were present In about 10 times the amount of the other amino acids. Phenol water was used as the f i r s t solvent i n the presence of 0.3$ ammonia and a tract of potassium cyanide. k9 evaporated to dryness, and dissolved i n 5 per cent sodium bicarbonate. The bicarbonate solution was added to 20 ml. of a solution of 0,1+ g. of dinitrofluorobenzene i n ethanol. The mixture was shaken i n the dark for two hours. The solution was then concentrated under reduced pressure to remove ethanol; the residue dissolved i n water and the ex-cess dinitrofluorobenzene removed by extraction with ether. After the aqueous solution had been acidified wl th dilute hydrochloric acid, the mixture was extracted with ethyl acetate (1+ X 20 ml.). The combined ethyl acetate extracts were re-extracted with 2 per cent sodium bicarbonate sol-ution. By this treatment the carboxyl-terminal DNP-amino acids were extracted into the bicarbonate solution, while the DNP-hydrazide derivatives which passed through the ion-exchange column remained i n the ethyl acetate layer. The combined bicarbonate solution was then acidified and ex-tracted with ethyl acetate. After evaporation of the ethyl acetate, the yellow residue was transferred to a chloroform-ether ( 9 : 1 0 , V/V) column buffered at pHlj.. Pour bands sep-arated: a fast running one was dinitrophenol, the second band was DNP-glycine, the third one was DNP-serine, and the slowest one was probably DNP-glutamic and DNP-aspartic mono-hydrazides which were not removed by the counter-current extraction. DNP-glycine band was purified on a second chloroform-ether column. After concentration to dryness, the elute from each band was evaporated to dryness, redissolved in 1 per cent sodium bicarbonate ($0 ml.), and read at 350 mu. 50 TABLE VII Quantitative determination of C-terminal by Hydrazinolysls. Protein, DNP-amino acid. 350 ny. 1 2 Mole of am-ino acid/ mole of protein. 1 2 Aver-age. Human JT-globulin Serine glycine 0.152 0.176 0.161 0.170 0.49 0,56 0.51 0 .54 0.50 0 .55 Bovine o'-globuHn serine glycine 0.21L2 0.181 0.252 0.194 0.77 0.58 0 .80 0.62 0.79 0.59 51 DISCUSSION The dinitrophenyl method clearly indicates that the free U-amino groups of human ^-globulin are those of aspartic acid and of glutamic acid. Quantitative measure-ment shows that aspartic acid i s present to the extent of 0 .55 mole and glutamic acid 0.7 mole per mole of protein assuming a molecular weight of 160,000 for human tf-globulin. These values are not corrected for hydrolytic breakdown of the DNP-amino acid. It has been determined by Porter and Sanger (22) that the recovery value of the pure DNP-aspartic acid i s about 60 per cent when hydrolyzed In boiling 5»7N hydrochloric acid for twenty-four hours and that of DNP- ' glutamic acid i s 75 per cent when hydrolyzed for twelve hours. The amount of DNP-amino acid recovered also varies with the presence of protein i n the hydrolyzate, and us-ually the lower value is obtained. Hence, by taking this correction factor of hydrolytic breakdown into consider-ation, human tf-globulin gives one mole each of aspartic acid and of glutamic acid as the N-terminal amino acid res-idues. This finding is In contrast to the earlier report of Van Vunakis (60) who found at least seven free amino acid N-terminals i n the same protein. McPadden and Smith (61) i n more extensive studies of the ^ -globulin, found that different subfractions of the human "tf-globulin have different amounts of N-termihal residues: one mole each of aspartic acid and glutamic acid i n subfraction II-3> but one mole of aspartic acid and two moles of glutamic acid i n subfraction I I - l and II-2. In both cases small amounts of serine were also observed. Putnam (62) reported that Fraction II contains, i n addition to aspartic acid, glutamic acid and serine, small 52 amounts of an unidentified DNP-derivative. It seems, there-fore, that different preparations of human V-globulin may contain different kinds and amounts of N-terminal residues* The major components would be those molecules with aspartic acid and glutamic acid as the N-terminal amino acid* Studies with bovine tf-globulin show that It i s even more heterogeneous than the human "tf-globulin with respect to N-terminal residues* Free ol-amino groups are located i n five different amino acids, namely, aspartic acid, glutamic acid, serine, alanine and valine* Bach of these Is present (taking into consideration hydrolytic breakdown) to the extent of about 0 .1 to 0*15 mole per mole of protein (M.N. 160,000). Since the sum total of the N-terminal amino acids i s s t i l l less than a molar quantity, s t i l l other N-terminal amino acids may be present i n undetectable amounts. These results suggest, therefore, that the preparation analyzed consisted of a number of molecules with similar pbysiocochemical prop-erties but different N-terminal amino acids. Unlike the determination of the N-terminal amino acid by the dinitrophenyl method which gives unequivocal re-sults, the methods for characterization of C-terminal amino acid are, as a whole, less satisfactory. Both the enzymatic and the hydrazinolysis methods used in this ezp eriment have their inherent disadvantages, and the interpretation of re-sults can hardly be arrived at without the correlation of the two procedures. When human V-globulin i s subjected to the action of carboxypeptidase for a very short period of incubation, a large number of free amino acids are liberated. Two d i -mensional chromatography revealed the presence of nine amino acids with five minutes of proteolysis. The amino acids 53 are, aspartic acid, glutamic acid, serine, glycine, threonine, alanine, valine, leucine and lysine. The large number of amino acids liberated would indicate that more than one open chain is present i n human -globulin, and that no proline or hydroxyproline (which are both resis-tant to hydrolysis by the enzyme) occurs near the C-terminus. Furthermore,since the same amino acids are liberated i n three hours of hydrolysis as are liberated i n five minutes, and since no amino acid released approaches a molar quantity, even after long hours of hydrolysis; one must conclude that the amino acid at the carboxyl end i s spl i t off at a slow rate. The liberation of.the C-terminal amino acid would thus be a rate-limiting factor. The amino acid adjacent to the C-terminal amino acid residue would be expected to appear as soon as i t s carboxyl group becomes free, so that before the terminal residue i s hy-drolyzed off completely, large amounts of succeeding amino acids along the chain would be liberated. Of the amino acids appearing on the chromatogram, serine, leucine and valine give the most intense ninhydrine color while the other six amino acids present gave weaker colors. If we assume that human ^ -globulin has two free o(-carboxyl terminal residues, i.e., two peptide chains, since we found two free N*terminal residues, two po s s i b i l i t i e s would arise. In the f i r s t one, any two of serine, leucine, and valine might be the C-terminal residues followed by the less prominent amino acids of the second group. I f this were 5 4 the case, quantitative yields of the terminal amino acids should be obtained. In the second possibility, one of the three amino acids serine, leucine, or valine would be the C-terminai residue of one chain followed by the other two. The hydrolysis would be terminated by a resistant amino acid accuring i n the fourth position of the chain. The second chain would have glycine as the C-terminal residue, which would be liberated very slowly by carboxypeptidase, followed by a great number of succeeding amino acids each in i t s release limited by the rate of liberation of glycine. Prom the carboxypeptidase method alone, therefore, no clear decision can be reached with regard to the C-terminal amino acid of human "^-globulin. Although bovine fl'-globulin has more free amino terminal residues than human tf-globulin, the free amino acids liberated by carboxypeptidase were the same for both bovine and human ^ g l o b u l i n . During a five-minute pro-teolysis, only four amino acids are s p l i t off, namely, ser-ine, leucine, valine and glycine. This seems to indicate that qualitatively, less C-terminal residues than N-terminal residues are present in bovine V-globulin. I f one assumes that both bovine ^-globulin and human V-globulin have the same C-terminal residues, the result supports the deduction that serine, leucine and valine belong to one chain while glycine belongs to the other. Those chains with 55 glycine as C-terminal residue wouJ.d liberate only one amino acid while i n the other non-glycine chain three amino acids would be liberated in the same time interval by carboxy-peptidase. Hydrazinolysis i s the second method chosen to ascertain the nature of free carboxyl terminal residue i n both proteins because this method gives free amino acid as the end product f a c i l i t a t i n g both qualitative and quanti-tative determination. It has been found by Locker (76) that the method w i l l f a i l i f the terminal amino acids are arginine, cysteine, cystine, asparagine or glutamine. For-tunately human and bovine I f-globulin do not posses these as C-terminal amino acids as demonstrated by the carboxy-peptidase experiments. When human o'-globulin i s subjected to hydrazinolysis for six hours, and the free amino acids liberated identified by two dimensional chromatography, a total of five spots are observed; serine and glycine give very intense colors with ninhydrin, while small amounts of alanine, threonine and glutamic acid are also v i s i b l e . It has been shown ( 7 6 ) , in the case of tropomyosins and actin, that hydrazinolysis also liberates small amounts of amino acids from positions other than the C-terminal, particu-l a r l y glutamic acid, serine and alanine. These "black-ground'1 amino acids arise, probably either through decom-position of the hydrazides during hydrazinolysis or through hydrolysis by a trace of water i n the hydrazine used. In 56 view of the fact that only two N-terminal amino acids are found by the dinltrophenyl method, only two chains are i n -dicated for human ^-globulin. The results of the carboxy-peptidase experiments indicate that the C-terminal amino acids of human ^-globulin can be serine, leucine, valine or glycine, Hydrazinolysis yields glycine, serine and traces of alanine, threonine and glytamic acid. Therefore, i t seems that serine and glycine are the two C-terminal residues of human Y-globulin, A quantitative measurement was made by conversion of the free amino acids to their dinitrophenyl derivative. It was found that about half a mole of each of the C-terminal amino aeid residues (gly-cine and serine) was present. Ohno (77) has determined the recovery value of leucine i n the presence of lysozyme, and found that 60 per cent of the leucine remained un-changed after ten hours of refluxing. Hence by taking into consideration the decomposition factor, one mole each of serine and glycine are liberated as C-terminal amino acids of human ^-globulin. From the results of the enzymatic method, this conclusion also lead to the suggestion that serine i s probably followed by leucine and valine along the chain* When hydrazinolysis was applied to bovine tf-globulln, the same results were obtained* although the amounts of serine and glycine recovered were slightly higher than for human ^-globulin. This result confirms the find-ings of the carboxypeptidase experiment. Since at least 57 five different N-terminal residues were identified i n this case, some of the peptide chains must have identical C-terminal amino acids* Having established the end groups in both Y-globulinsj attempts were made to deduce the amino acid se-quences. The N-terminal DNP-peptides were isolated by counter-current extraction of the partial hydrolysate with ethyl acetate, sodium bicarbonate solution, and ethyl acetate. The f i n a l acetate extract should contain a l l of the ell-N-terminal DNP-peptides, In the ease of insulin, where the ratio of lysine to ot-amino residues is two to four, this operation offers clear cut separation of the DNP-tA-amino peptide from €-DNP-lysine peptides. In human Y-globulin where the ratio i s about one hundred to one, the f i n a l acetate estracts are found to be contaminated heav-i l y with 6-DNP-lysine peptides* Ftirthermore, since two open chains are found i n human ^-globulin, the terminal peptide mixture i s more complicated than that of rabbit V-globulin Investigated by Porter (59) where only one chain was found. It seems that, for proteins with more than one chain, the polypeptide chains must be separated f i r s t be-fore characterization can be attempted* Since the action of proteolytic enzymes on pro-teins would be expected to be more specific than that of acids or alkalies, dinitrophenyl globulin was subjected to proteolysis by pepsin, cbymotrypsin and trypsin i n the hope 58 that a few "labelled" terminal peptides could be easily isolated* None of the crystalline enzymes, however, would attack the dinltrophenyl-tf-globulin. The introduction of the dinltrophenyl group does not seem to offer any stero-chemical barrier to the action of these classical proteo-l y t i c enzymes because both DNP-insulin ( 7 8 ) and DNP-ribonuclease ( 7 9 ) are hydrolyzable by these enzymes* The inactivity of the enzymes in this case must be attributed to the insolubility of the large DNP-^-globulin molecule. Finally, attempts have been made to fractionate the fragments of peptides obtained from enzymatic hydrolysis of human ^-globulin by the continuous electrophoresis apparatus* The hydrolyzate separated from the point of application into a continuous spectrum i n the paper between the two poles with no well-defined individual bands* When the effluent from each "cut" was studied by one dimensional chromatography, a complicated ninhydrin pattern was ob-tained* No single peptide could be obtained from the chrom-atogram* Indeed, an even more elaborate but unsuccessful experiment on a smaller protein (erytalbumin) was reported by Boulanger and Biserte ( 8 0 ) . The exceedingly large num-ber of fragments derived from a protein of the size of tf-globulin would probably preclude any attempt of fraction-ation in the Immediate future* 59 SUMMARY 1* Sanger's 2:i|-dinitrophenyl method has been applied to human V-globulin. One mole each of aspartic acid and of glutamic acid have been identified as the free ol-amino terminal residues* The amino group of lysine is also avail-able for reaction with 2tii-dinitrofluor©benzene. 2* In bovine Y -globulin, free amino groups are located on five different amino acid residues, namely, aspartic acid, glutamic acid, serine, alanine and valine. Each i s present in sub-molar quantity indicating that the protein i s also heterogeneouswith respect to N-terminal groups. 3. By a combination of the carboxypeptidase and the hydrazinolysis methods, one mole each of serine and gly-cine have been established as the C-terminal amino acids of human V-globulin. 4* The free carboxyl terminal groups of bovine V-globulin also have been found to be serine and glycine* Since bovine ^-globulin possesses at least five different N-terminal amino acid residues, some of the molecules must have the same C-terminal amino acids. IL* Carboxypeptidase experiments indicate that there is no difference i n the release of free amino acid from 60 bovine Y-globulin and human V-globulin. The C-terminal sequence of these proteins i s probably the same. In one chain, serine may be followed by leucine and valine. BIBLIOGRAPHY 1. Vickery, H.B., and Osborne, T.B., Physiol. Rev., 8, 393 (1928). t ~ 2. PIsher, E., Z. physiol. Chem., 151 (1901). 3 . Dakin, H.D., Biochem. J., 12, 290 (1918). k. Eossel, A., and Kutscher, P., Z. physiol. Chem., 31, 165.(1901).. 5* Kossel, A., and Edlbacher, 3«, Z. physiol. Chem., 110, 2kl (1920). 6. Van Slyke, D.D., J. B i o l . Chem., 9, 185 (1911). 7. Gordon, A.H., Martin, A.J.P., and Synge, R.L.M., Biochem. J., ^ 7, 79(19k3). 8. Consden, R., Gordon, A.H., and Martin, A.J.P., Biochem. J., ^8, 22k (19kk).. 9 . Stein, W.H., and Moore, S., Cold Spring Harbor Symposia Quant. Biol., lit, 179 (19k9). 10. Creig, L.C, Cold Spring Harbor Symposia Quant. B i o l . lit, 2k (19k9), .. 11. Svensson, H., Advances i n Protein Chem., L, 2 5 l (19W. . . 12. Pox, S.W., Advances i n Protein Chem., 2, 156 ( 1 9 k 5 ) . 1 3 . Fisher, E., and Bergell., P., Ber. chem. Ges., 36 , 2592 (1903)* Ik. Gurin, S., and Clarke, H.T., J. Bio l . Chem., 107, 395 (193k). 15* Abderhalden, E., and Heyns, K., Z. physiol. Chem., 202, 37 (1931). 16. Goldschmidt, S., Wiberg, E., Nagel, P., and Martin, K., Ann. Chem. k56, 1 (1927). 17. Bergmann, M., Mlckeley, A>, and Kann, E., Ann Chem. k58, 56 (1927). 18. Barger, G. and Tutin, P., Biochem. J. 12, k02 (1918). 19* Jensen, H., and Evans, E.A. Jr., J. Bi o l . Chem., 108, 1 (1935) . 2 0 . Sanger, P., Biochem. J., ^ 9, 507 (19k5). 21 . Porter, R.R., and Sanger, P., Biochem. J. kj2, 287 (19i|-8). 22. Abderhalden, E., and Biumberg, P., Z. physiol. Chem., 6>|, 318 (1910). -23. Abderhalden, E., and Stix, W., Z. physiol. Chem., 129., Ik3 (1923). 2k. Blackburn, S., and Lowther, A.C., Biochem. J. k 8 , 126 ( 1 9 5 D . 2 5 . Biserte, 6«, and Osteux, R., Bull. Soc. chim. b i o l . , i l , 50 ( 1 9 5 D . 26. Sanger, P., Biochem. J.-, k^, 563 (19k9). 2 7 . Sanger, P., and Tuppy, H., Biochem. J., IL9, k63 ( 1 9 5 D . 2 8 . Sanger, P., and Tuppy, H., Biochem. J., 1L9, k 8 l (1951) . 2 9 . Sanger, P.-, and Thompson, E.O.P., Biochem. J., 53 , 3 5 3 . ( 1 9 5 3 ) . 3 0 . Sanger, P., and Thompson, E.O.P., Biochem. J . , 5 3 , 366.(1953)., 31> Edman, P., Acta chem. Scad., k_* 2 ^ 3 (1950) . 3 2 . Praenkel-Conrat, H., J. Am. Chem. Soc, J_6, 3 6 O 6 ( 1 9 5 k ) . 33* Schlack, P., and Kumpf, W., Z. physiol. Chem., 15k, - : 125 (1926). . 3 k . Abderhalden, E., and Brockmann, H., Biochem. Z., 225, 386 (1930). 3 5 . Bergmann, M., and Zervas, L., J. B i o l . Chem., 113, 3 k l (1936) . 36. Khorana, H.G., J. Chem. Soc, 2 0 8 l ( 1952) . 37* Promageot, C , Jutlsz, M., Meyer, D., and Penasse, L., Biochim. et biohpys. acta 6 , 283 (1950) . 3 8 . Chibnall, A.C., and Rees, M., Biochem. J., k 8 , x l v i i (1951) . 3 9 . Akabori, S., Ohno, K., and Narita, K., Bull. Chem. Soc. Japan, ££, 21k (1952) . 4 0 . Akabori, S., Ohno, K., and Narita, K., Bull . Chem. Soc. Japan., 2£, 214 (1952). 4 l « Chambers, R.W., and Carpenter, P.H., J. Am. Chem. Soc. 21, 1527 (1955) . 4 2 . Neurath, H., and Schwert, G.W., Chem. Rev. 1L6, 69 ( 1 9 5 0 ) . 4 3 . Smith, E.L., Advances i n Enzymol., 12, 191 (195l). 4 4 . Waldschmidt-Leitz, E., Physiol. Rev., 11, 358 (1934) . 4 5 . Anson, M.L., J. Gen. Physiol. 20, 663 (1937). 4 6 . Hofmann, K., and Bergmann, M., J. Bio l . Chem. 134. 225 (1940) . 47* Stahmann, M.A., Pruton, J.S., and Bergmann, M., J. Bio l . Chem., 164, 753.(1946).-4 8 . Bergmann, M., and Pruton, J.S., J. Biol. Chem. 145, 247 ( 1942) . 4 9 . Smith, E.L. in Wolstenholme, G.E.W., and Cameron, M.P., The Chemical Structure of protein, London ( 1953) . 5 0 . Grassmann, W., Oycherhoff, H., and Eibeler, H., Z. physiol. Chem. 189, 112 (1930). 51 . Lens, J., Biochem. et blophys. acta, 367 (1949). 5 2 . Martin, A.J.P., B r i t . Med. Bull., 10, 161 ( 1954) . 5 3 . Hughes, W.L., in Neurath, H., and Bailey, K., The Protein, II, B, New *ork ( 1954) . 54* Smith, E.L. and Vager, B.V., Ann. Rev. Microbial., 6, 207 (1952).. . 5 5 . Oncley, J.L., Cordon, P.H., Gurd., P.R.N., and Lontie, R.A.Abstract Of .paper presented at the Division of.Biological Chemistry, 119th meeting of-the American Chemical Society, Boston, Mass., Apr i l 1 - 5 , 1951, P. 250* 56. Cann.,J.R.,, and Kirkwood, J.G., Cold Spring Harbor Symposia Quant. Biol., l i ^ , 9 ( 1 9 4 9 ) . 57 . Tiselius, A., and Kabat, E.A., J. Exptl. Med., 69, 119 (1939). . . . . 5 8 . Oncley, J.L., and Scatchard, G., and Brown. A«, J. phys.. and Colloid Chem., j>l, 184- (1947) . 5 9 . Porter, R.R., Biochem. J., 1^ 6, 473 (1950) . 60. Van.Vunakis, H., Microfilm Abstracts 11, 822 (1951). 61. McFadden, M.L., and Smith, E.L., J. Am. Chem. Soc*, j£, 278k (1953). 62. Putnam, P.W., J. Am. Chem. Soc. 7J>, 2785 (1953). 63. Breinl, P., and Haurowitz, P., Z. physiol. Chem., 192, k5 (1930). 6k. Burnet, P.M., and Penner, P., The Production of Anti-bodies, 2nd ed., MacMillah, Melbourne (19k9). 65. Pauling, L., J. An. Chem. Soc* 62, k57 (I9k0). 66. McPadden, M.L. and Smith, E.L., J. B i o l . Chem., 21k, . 185 (1955). - — * 67. Boyd, W.C., i n Neurath, H., and Bailey, K., Tie B?oteln, I3Ey ByrNewlYpH . (1954). 68* Green, P.C, and Kay, L*M*, Anal. Chem., 2k, 726 (1952). 69. Rao, K.R., and Saber, H.A., J. Am. Chem. Soc* 76, 1326 (195k). . 70* Porter, R.R., i n Gerard, R.W., Methods i n medical research, Chicago \t.256 (1950). 71. M i l l s , G.L., Biochem, J., £0, 707 (1952)* 72. Blackburn, S., and Lowther, A.G., Biochem. J., k8, 126 (195D* 73. Mills, G.L., Nature 165, k03 (1950). 7k. Perrone, J.C., Nature, 167, 513 (195D* 75* Jansen, E.P., Nutting, M.D., Jang, R., and Balls, A*K«, J. B i o l • Chem. 179, 189 (1949). 76. Locker, R.H., Biochim. et.biophys. acta, ly., 533 (195k). 77. Ohno, K., J. Biochem. Japan, k€>, 6k (1953). 78. Woolley, D.W., J. Bi o l . Chem. 1J9, 593 (I9k9). 80. Boulanger, P., and Biserte, G«, i n Wolstenholme, G.E.W., and Cameron, M.P., The Chemical Structure of Protein, London (1953)'. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items