A STUDY OF THE PROTEIH-HUCLEATES OF THE SPECIES OF THE GEIHJS BRUCELLA I. II. CHEMICAL CONSTITUTION OF THE PROTEIN-HUCLEATES. BIOLOGICAL PROPERTIES OF THE PR0TEI1-ÎÏÏJCLSATES. THESIS Submitted to the Faculty of Michigan State College of Agriculture and Applied Science in p artial fulfillm ent of the requirements for the degree of Doctor of Philosophy. W. H. ^ a h l July, 1939 ProQuest Number: 10008434 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest ProQuest 10008434 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 A STUDY OF THE PRGTEIN-IJUCLFATES OF THE SPECIES OF G im S BRUCELLA CONTENTS I. The Chemical Constitution of the Protein-nucleates. Introduction, ............... 1, Isolation of the P ro tein -n u cleate.................................. 2. Chemical Nature of the Protein-nucleate..................... . . . . .... Separation of the Protein-nucleate into i t s Components.... Chemical Study of the P rotein. ................. Chemical Study of the Nucleic A cid.. II. 5* 7* ........................... Discussion. Summary. If.* 10. 17 * ................................................ The Biological Activity of the Protein-nucleates Precipitation Studies, 20. ..................... ........... 22. 22. Allergic Activity 2lf.. Antigenicity........................................ 25. Toxicity............................................................................................................... 27. The Biological Activity of the Protein and Nucleic A cid.... 27. Precipitation Studies................................... Allergic A ctiv ity . 28. ........................ Antigenicity. . »• • ...................... Toxicity. 30. ............ Discussion.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary Bibliography . . . . . . . . . . . . . . ........................... Acknowledgement....................... 29* 30. . ..... 3i* . . . . . .. 32. 3338* 1. CHEMICAL CONSTITUTION OF THE PROTEIN-NUCLEATES* There have been rela tiv ely few reports in the lite ra tu re regarding the chemical composition and biological properties of b acterial ^^nucleoproteins", or protein-nucleates; or of the properties of the bacterial nucleic acids* Johnson and Brown (l) and Johnson and Coghill (2) have studied the ”nucleoprotein”, protein and nucleic acid of Mycobacterium tuberculosis. Similar studies have been carried out by Thompson and Dubos (3) on Diplococous pneumoniae, by Ferramola (i+) on Bacillus anthracis and by Mitra (3) on Vibrio comma. Studies on "nudeoprotein" and protein fractions of Strepto coccus pyogenes have been reported by Sevag, Lacktnan and Smolens (6) and by Heidelberger and Kendall ( 7 ) on Streptococcus hemolyticus. Preliminary studies on the "nucleoproteins" of the species of Brucella have been made by Topping (8) and Huddleson and associates (9*10,12). In a previous study, Huston, Hershey and Huddleson (9 ) described the preparation of an antigenic and chemically defined substance which they called a "nucleoprotein", th is noiv being termed a protein-nucleate. Further studies on th is material resulted in its production as "Brucellergen”, which is a suspensoid of the protein-nucleate and is used as a skin-testing agent in detecting undulant fever (brucellosis) in humans (12). Although clin ica l research has progressed on th is material since I 9 3 U9 very l i t t l e study has been made of i t s chemical composition and biological properties* * This study was made possible by grants in aid from the Horace A, and Mary A. Rackham Fund and by the Bureau of Animal Industry, U. S. Department of Agriculture. -2 The present papers concern the chemical nature and biological proper­ tie s of the protein-nucleate fraction of the smooth, intermediate rough and rough forms of the organisms. The smooth form w ill be designated by (S), the intermediate rough by (IR) and the rough by (R). Isolation of the Protein-nucleate The protein-nucleates were prepared from a number of smooth, in te r­ mediate rough and rough strains of Brucella abortus. Brucella melitensls and Brucella suis from the stock collection maintained in the Central Brucella Station. The one intermediate rough and tru ly rough strain studied were Brucella abortus » All gave typical reactions characteristic of th e ir species as to dye bacteriostasis, hydrogen sulfide production, and agglutinogenes i s . The dried cells were prepared in a manner similar to th a t described in a previous publication (9)* b all m ill for 7 days. The dried organisms were ground in a The resulting fine powder was suspended in suf­ fic ie n t d is tille d water (pH 6.5) to make a suspension of approximately 2.5 per cent by weight, s tirre d by means of a mechanical s tir r e r at room temperature for 3 hours and allowed to stand in a cold room at 0® to 5“C. for 2 I4. hours. After th is operation, the insoluble material was separated by passing the mixture through a steam-driven Sharpies laboratory centri­ fuge. The c e ll residue was extracted a second time in the same manner, th is time, however, adding sufficient normal sodium hydroxide to bring the pH to 8.0. The f i r s t and second water extracts which made a clear to slig htly opalescent pale yellow solution, were combined and precipitated with -3 - glacial acetic acid by bringing the pH to 3*3* was allowed to sediment over night in the cold. The resulting precipitate The following morning the precipitate was collected in the centrifuge and washed with d is tille d water. This precipitate was then stirred mechanically in cold d is tille d water, and N sodium hydroxide was added u n til the pH of the solution was 7.2, a t which point a l l of the material dissolved completely, however, s t i l l remaining opalescent. second and th ird time* This was reprecipitated and dissolved a A neutral solution of the twice reprecipitated protein-nucleate was f ilte r e d through a Seitz f i l t e r . The water clear f i l t r a t e s were acidified, the precipitated proteins washed with water and made up to a 1 per cent solution with water at pH 7*0* These were used for further study. I t is not our purpose here to argue pro or con as to the d e sira b ility of using th is method for the isolation of the protein-nucleate in it s le ast modified form. This is the method which has been satisfactory in producing an allergic skin testing agent (Brucellergen), and is taken for analysis at i t s own value. The method of Heidelberger and Kendall (7) was used for the preparation of one "nucleoprotein" fraction, and its analysis and properties have been compared with the others. consists b riefly in: This method extraction of the centrifuged whole cells with acetone and ether; centrifugation and grinding of the cells; extraction with 0.2 H acetate buffer a t pH 4*0; centrifugation and extraction of the residue with m / 15 phosphate buffer at pH 6.5; precipitation of th is extract with glacial acetic acid - th is precipitate being the "nucleoprotein" fraction used. I t was further purified as directed in the publication (7). -4 - The Chemical Nature of the Protein-nucleate As Osborne and Campbell (11) have so aptly phrased i t - "in re a lity *nucleoprotein* means rather a ’method of preparation’ than a chemical substance,” They have given evidence th at protein-nucleates, prepared from the same tissu e by s lig h tly different methods had different ratios of protein to nucleic acid. Such variations could be expected when a polyvalent base (protein) unites in different proportions with a poly­ valent acid (nucleic acid). Thus we have tr ie d to use a constant pro­ cedure for the preparation of the protein-nucleates, and even though carrying out the procedures very carefully, small differences between different preparations of protein-nucleate from the same organism were noted. These protein-nucleates, as already stated, were obtained by pre­ cip itatio n with acetic acid at th e ir isoelectric point of pH 3*5 to The white, flocculant precipitate is readily and completely soluble at pH 7,0, The solution does not coagulate on heating at neutral reaction and in the presence of electrolytes. The to ta l yield of protein-nucleate for the smooth strains of abortus, melitensis and suis was from 12,0 to 13*7 per cent of the dry weight of the c e lls ; whereas the intermediate rough and rough strains gave 18,0 to 18.2 per cent of the dry weight of the c e lls. and phosphorus content are set forth in Table I, The nitrogen “5 - Table I* Nitrogen and Phosphorus Content of Protein-nucleates K p er o e n t P per c e n t N /p r a t i o a b o r tu s 230 (S ) 1 3 .0 2 3 . c6 U .9 8 »a b o rtu s l 6 l ( s ) U t.8 2 2 .7 2 3M m e l i t e n s i s 219^4^ (S ) 14*91 2 .8 9 5 .1 6 m e l i t e n s i s 2I4.25 (S ) 1 5 .1 8 2 .7 Ü m e l i t e n s i s 9G (S ) 1 4 .8 8 2 .6 0 5 .6 8 s u is 1582 (S ) 1 4 .6 8 2 .3 1 6 .5 5 a b o r tu s 65 ( IR) 1 4 .7 6 3 .3 1 _ _ i+.ij6 a b o r tu s 805R (R) Ü1..83 3 .5 5 4 . 1^2 ^ ♦ abortus l6 l protein-nucleate prepared by method of Heidelberger and Kendall (11). A 1 per cent solution gave the following biochemical reactions* B iais’ and Molisch and Feulgens markedly positive, biuret, Rosenheims*, Milions’ and Sakaguchi positive, and labile sulfur negative. I t was found impossible to run optical rotation on most any con­ centration of protein-nucleates, due to th eir opalescence and slight yellow color. Separation of the Protein-nucleate into its Components Three methods were tr ie d for the separation of biologically active protein fractions from the protein-nucleates. The f i r s t was th a t of Johnson and Brown ( l) in which the dried material was extracted cessively with 3 per cent and 5 per cent sodium hydroxide. suc­ By th is -6 method they obtained a yield of nucleic acid that ?/as approxinmitely 1 per cent of the original weight of the tubercle bacillus# However, the alk ali metaproteinate formed from Brucella was found to be biologically inactive and not suited fo r further studies. However, the nitrogen p a rtitio n figures correspond remarkably well with those of the biologically active proteins obtained by the second method. The second method was that of Sevag, Laokman and Smolens (6), which consisted in s p littin g the protein-nucleate with sodium carbonate and separating the mixture by means of chloroform. follows: A detailed example is as 5 grams of the protein-nucleate was dissolved in 1,4 l i te r s of d is tille d water containing 7.1 grams of anhydrous sodium carbonate and heated in a water bath at for 2 hours. After the mixture was cooled and any insoluble residue removed by centrifugation, i t was neutraliz­ ed with acetic acid to pH 7*0. As Sevag states, "The protein and nucleic acid were then present in a free sta te . To remove the protein from solution a water-insoluble chloro- form-protein gel was formed. This was separated by centrifugation. Nucleic acid does not form such a combination; protein and chloroform seem to enter into a loose molecular combination, and because of the density of the chloroform th is complex settles out. On the acid side of the pH at which the protein precipitates, a s a lt-lik e recombination with or a slight adsorption of nucleic acid may take place. This can largely be prevented by adjusting the acidity with acetic acid to a point between the isoelectric point of the protein component and neutrality ." Specifically, a fte r hydrolysis and neutralization, 0.25 volume of chloro­ form and 0.1 volume of butyl alcohol were added. The mixture was shaken mechanically about 6 hours, and upon centrifugation separated into two layers< The lower layer consisted of a fa irly stable chior oform-prote in gel; the -7 - upper layer is an aqueous one. from the gel* The upper layer can easily be decanted I f an excess of chloroform is used, a th ird layer is formed which consists of pure chloroform and is found in the bottom of the b o ttle . The upper aqueous layer is further shaken and centri­ fuged u n til a gel no longer forms at the interface* Traces of non-protein component are removed by transferring the chi or of orm-prote in gel to a wet f i l t e r in a funnel* liberated by treatment with one volume of alcohol. The protein was The product was then centrifuged and washed with water adjusted to pH 5 to 6 with acetic acid. The washed protein was dissolved by adjusting to pH to 7 .5 . The supernatants from the protein-gel vhich contains the nucleic acid are combined. protein. Although they appeared clear, they s t i l l contained By concentrating in vacuo to a smaller volume and reextracting with chloroform, the la s t trace may be removed. The nucleic acid was isolated from the protein-clear f il t r a t e by treatment with ij. volumes of alcohol containing enough N hydrochloric acid to bring i t to pH 5*0, chilling in the cold room and centrifuging. The above procedure was re ­ peated to obtain a purer product* A th ird method consisted of s p littin g the protein-nucleate as above, and then fractionally precipitating out the protein a t its isoelectric point (pH 5*5 to 4*0) and the nucleic acid at pH 2.0 to 2.5. Hovrever, the composition of the purified nucleic acid was so variable on batches of the same organism that the method was abandoned. Chemical Study of the Protein The proteins studied were a l l prepared by the Sevag chloroform-gel method. The yield of protein from the protein nucleates of S strains varied from 70 to 77 per cent* From both the IR and R strain protein- nucleates, the protein yield was approximately 6o per cent* The protein gives a positive biuret te s t for the peptide linkage; a positive Mi lions ’ te s t for tyrosine; a fain t reaction in the Rosenheim te s t for tryptophane and quite a strong Sakaguchi te s t for arginine* The protein gives a positive Molisch te s t for carbohydrates; a positive reaction with Biais’ t e s t for pentoses* Nitrogen p artitio n (Hausmann numbers). The protein fractions from the S, IR and R strain preparations of protein-nucleate were analyzed for to ta l nitrogen, amide nitrogen, humin nitrogen, diamino nitrogen, mono­ amino nitrogen; the amino acids arginine, lysine, histidine and cystine and for phosphorus. The method for the nitrogen p artitio n was based on Hewitts adaptation (13) of the Thimann method (llj.)* summarized in Table 2. The resu lts are As may be seen in th is Table, the only significant differences in composition of the S, IR and R strain preparations were th eir diamino acid content* These proteins do not in any way resemble his tones or protamines, according to two main c r ite r ia . F irst, they do not possess the very high diamino nitrogen content of h i stones (1 9 to 30 per cent) and protamines (53 to 89 per cent); second, most histones and protamines are toxic in nature, but i t w ill be shown la te r that the proteins under study were non­ toxic. The proteins also f a i l to give many of the other te sts character­ is tic of histones and protamines. Carbohydrates. The Sorenson and Haugaard ( 1 5 ) procedure for car­ bohydrates was run on a l l protein fractions. Hiile no actual quantitative P cc a B C O tr -g CO t> {? O J OJ cr L T C O o o OJ LT O xO P: L fI e K F) œ ir \o -=i « r- ■ ei Ç g: CO o l C C S O- o O I § à •H C O C O •g LT, K " hC xO I-I rr ir O O C cr Q) 8 c\ pq 4-1 fo o rH K a p Q cH V Û CO to 0) 1 o o CO o> ■s C V J C O X O 8 CM Q) K > •s E h O rH C Û to •H i s rH a P .9 ^ d P. % & % xO ro- S '' 8 & xO 8 g Ü f£ L fN I— I 4â u •P “ L T \ in §) ON 3! ON OJ ON 11 0 p! 'S •H .9 M I 0 •H Q % IT N L fN o- xo! d> PÎ •H Pi •H b.0 8 § •H CD .9 73 •H p ; to •H td o & § •H p; -& to o £! i -1 0 - determination was made, nor identification of the sugar made, i t was quite evident th at polysaccharides were present in very small quantities. Re­ ducing sugars were present as such, and calculated as glucose (SchafferEartmon procedure) ran from 1*56 to 2 .1 5 per cent on the various samples. After hydrolysis with dilute hydrochloric acid, reducing sugars are found to comprise from 2.00 to 2.87 per cent. As the amino-sugar, glucosamine, has been found present in other Brucella fractions studied in th is laboratory, i t s determination was attempted on th is particular fraction. Using the technique of Palmer, Smythe and Meyer ( l6 ), no trace of glucosamine could be detected. Chemical Study of the Nucleic Acid The nucleic acids, when dried with alcohol and ether formed a vdiite powder; when precipitated with hydrochloric acid the dry powder was readily soluble in d is tille d water made slightly alkaline. After precipitation with alcohol i t was immediately soluble in water, giving a perfectly clear solution. The nitrogen and phosphorus analyses are given in Table 5. The nucleic acids a l l give negative biuret. Milions’ and Rosenheims* reactions for the detection of protein components; the Molisch te s t for carbohydrate. B iais’ te s t for pentose and Feulgens te s t for desoxypentose were positive. The yield of nucleic acid from the protein-nucleate prepared from the S forms was approximately 25 to 50 per cent, v/hereas that from the IR and R form was about ij.0 per cent. Determination of Purine Nitrogen. - Samples of nucleic acid were hydrolyzed in 5 per cent (by volume) sulphuric acid for 2-1/2 hours on a hot plate a t 100®C. To each sample a hot solution of silv er sulphate was added u n til the f i l t r a t e gave no further precipitation with th is reagent. -1 1 - Tatle 3* Nucleic Acid from Analybioal Data on Brucella Nucleic Acids Total N Total P % Purine N N;P Ratio Purine N: Pyrimidine N Ratio melitensis 219h (S) 1 3 .1 1 6 .2 5 2 .2 3 4 2 .1 0 .7 6 melitensis 21+23 (s) 1 3 .1 2 6 .9 1 1 .9 0 3U.O I.II+ suis 1382 (S) 12.73 7 .1 2 1 .7 9 4 4 .5 0 .7 h abortus 230 (S) 1 2 .1 9 6.35 1 .9 2 abortus l6 l (S) 13 .81 6 .6 3 2.08 3 2 .4 0.1+7 abortus 83 ( IE) i h .5 0 8.84 1.61^. 5 7 .5 1 .3 2 abortus 8 O3 R (R) 1 3 .5 0 7 .9 6 1 .6 9 6 3 .0 1 .6 9 thymonucleic acid^ 1 6 .7 8 9.89 1 .6 9 4 9 .9 0.97 yeast nucleic acid* 1 6 .5 2 9.64 1 .6 9 6 6 .7 2.00 1 5 .7 2 8 .4 9 1.83 7 4 .7 2 .9 5 pancreas nucleic acid 0 0 .6 0 # Calculated according to Levene and Bass ( (l?) pp. 263» 271+) ® Levene, P, A. and Jorpes, E, J, Biol. Chem. 86, 3^9» 1950» -1 2 - After cooling, the precipitate was centrifuged and washed twice with silv er sulphate solution. Nitrogen v/as determined on the purine pre­ c ip ita te and pyrimidine f i l t r a t e by the Micro-Kjeldahl method. From Table 3 one may note the amounts of purine nitrogen closest to that of thymonucleic acid. In going from the S - IR - H forms, one notices th at the amount of purine nitrogen tends to increase. Whether th is difference in the chemical composition of the nucleic acid has anything to do with the physiological characteristics of the organism is a point which should be further examined. Isolation and Identification of Purines. The purines were isolated from Br. abortus (S), Br. melitensis (S) and Hr. abortus (R) according to Jones* directions fo r the isolation of guanine and adenine (l8 ). The guanine obtained after hydrolysis of the nucleic acid was precipitated twice, including one treatment with n o rit. The crude guanine obtained from the various preparations was converted into guanine hydrochloride and recrystallized from 5 per cent hydrochloric acid. For analysis the guanine hydrochloride was dried in the a ir . CsHgNsO-HCl-PHsO. Calculated, N 31*3; found, N 31*9 The f i l t r a t e from the guanine obtained in the quantitative deter­ minations was prepared for the isolation of adenine by precipitating the purines as cuprous s a lts and decomposing with hydrogen sulfide. The adenine was then precipitated as the picrate. After recry stalliza­ tion from 23 per cent acetic acid i t melted a t 286-291**C. CiiHgNsOy. Calculated, N 30.63; found, N 31*3 Although guanine and adenine were isolated and derivatives prepared, accurate weighings of each were not made. However, from the amounts -1 3 - weighed for the nitrogen determinations and from what was le f t over, one could judge easily th a t they were present in an approximate ratio of guanine : adenine = 1:1. Isolation and Identification of Pyrimidines. 2 gm. samples of nucleic acid from Br. abortus (s), Approximately Br. melitensis (S) and Br. abortus (k) were hydrolyzed in an o il bath at 150-l60®C. for 3 hours with 20 cc. of ^ per cent sulphuric acid. After being cooled, the sulphuric acid was removed with barium hydroxide and the purines precipitated with silv er in a dilute acid solution. The pyrimidines were then removed by the addition of more silver and barium hydroxide. The resulting precipitate was decomposed with hydrogen sulfide and the f i l t r a t e from the silver sulfide concentrated. No thymine came down. From th is solution the cytosine was precipitated with picric acid. After twice recrystallizing, the picrate sintered at 256®C., and rapidly melted at 2 6 6 **C. CgHeONa CeHsOyNs. Calculated, N 21).,71; found, N 23.li+ After removal of the picric acid from the cytosine picrate mother liquors, i t was concentrated and even dried, but uracil did not cry stallize out. Portions of the above liquid before completely drying were tested by the “ Wheeler and Johnson color te s t ( 19), but in every case a negative te s t was obtained. The color reaction described by Harkins and Johnson (20) was performed at various intervals during the above described procedure, but in no case did i t give a positive reaction. In order to make certain the absence of th is pyrimidine, the following experiment was performed; i+ gms. of mixed nucleic acids was hydrolyzed with 23 per cent - 14- sulphur ic acid and the pyrimidine fraction isolated as previously described. The solution was then concentrated to a very small volume and tested for thymine by the method of Harkins and Johnson. Although the te s t is sensitive to 2 mg. of thymine in the presence of much larger quantities of cytosine, a completely negative te s t was obtained. Nature of the Sugars Pentose content - The determination of pentose was executed employing the method described by Hoffman (21). The sample was trans­ ferred into a 300 cc. d is tillin g flask made according to the directions of the author. 50 oc. of 20 per cent hydrochloric acid were added and d is tille d with a current of steam for 3 hours, always keeping the tem­ perature of the mixture constant at 103-103®C. The d is ti lla te , collected in a measuring flask, was titr a te d in th is flask with 10 per cent sodium hydroxide from a burette to neutrality to phenolphthalein, and the liquid was then diluted to an equal volume, in the present case 30 O cc. The standard solution containing furfural approxinately equal to th at expected in the d is ti lla te was placed in a flask of the same size and treated with hydrochloric acid and sodium hydroxide u n til the solution was neutral to phenolphthalein, care being taken that the sodium hydroxide added equals the amount added to the unknown. mark. This solution was also diluted to the 6 cc. of each te s t and standard solution were transferred to a te s t tube, each was treated with 0,3 oc. aniline and i+*0 cc. of glacial acetic acid, allowed to stand for 10 to 13 minutes in the dark, and then compared in a colorimeter. The amount of pentose v/as calculated on the basis of the experimental data of Hoffman, in which he can recover the furfuraldéhyde from guanine nucleotide 98.0 per cent, adenine nucleotide -1 5 - 98*0 per cent^ cytosine nucleotide If..5 per cent and uridine nucleotide 15*7 per cent. Table U. From Brucella Pentose Content of Nucleic Acids £ample mgs. Furfural recovered mgs. % Pentose (calculated as a tetranucleotide) % Pentose (Calculated as a hexanu cleotide) melitensis 219 i|. (s) 50 2 .7 7 15.3 1 2 .7 melitensis 2ij25 (S) 50 2.95 1 6 .7 1 5 .8 suis 1582 (S) 20 1 .2 2 1 7 .3 iU-3 abortus 250 (S) 50 2 .8 0 1 5 .9 13*2 l6l (s) Uo 2*kQ 1 7 .0 1 4 .0 abortus 85 (IR) 50 ' 2 .5 2 II4..5 abortus 8 O5 R (R) ho 1 .9 7 1 3 .9 abortus tetranucleotide (yeast) —' *" . - 1 1 .5 2 0 .8 hexanuc1 eotid© Thymonucleic Acid Content - 1 1 .8 2 5 .6 This determination was carried out by means of the method introduced by Widstrom (22), His method is based upon the fact th a t %hen thymonucleic acid is heated with any acid i t undergoes a disruption of i t s desoxyribose component with the formation of a genuine aldehyde, which gives a color reaction vd.th Schiffs* fuchsin sulfurons acid reagent, and the la tte r coloration is applicable to the estimation of the original nucleic acid under limited conditions. 1 cc. of the solution of the 5 different sized samples which comprised 0 ,S to -l6 - 10 mg, of thymonucleic acid was taken in a t e s t tube and made predominately acid to Orego Red by the addition of 0.1 N hydrochloric acid and then lOcc. of c itra te buffer solution of pH 2.0 were poured into i t and thoroughly mixed. Upon heating in a boiling water bath just 2-3/U minutes, the te s t tube was taken out and immediately cooled in running water. The slig h tly insoluble material was removed by centrifugation and 9 cc, of the clear solution treated with 5 oc. of the mixed reagent composed of two volumes of c itra te buffer solution as above, and one volume of Feulgens reagent. On the other hand, simultaneously a similar procedure was undertaken with The colors, a fte r standing sealed for 2h 1 cc. of the standard solution. hours, were matched in the colorimeter. The results are given in Table 5* Table 5- Thymonucleic Acid Content of the Nucleic Acids Nucleic Acid from melitensis 219U ' suis 1582 (S) 1 (S) per cent T.N.A. 4 0 .9 54a abortus l6 l (S) 46.5 abortus 250 (S) 55.5 abortus 5 6 .5 85 ( IR) abortus 805 R (R) 2 2 .4 -1 7 Gluoosamine - Glucosamine, as determined by the method of Palmer, Symthe and Meyer (l6) could not be detected in any of the nucleic acids. Comparative Test of Dissolubilit^r in Acetic Acid The well defined nucleic acids, namely yeast and thymonucleic acids, d iffer in so lu b ility in acetic acid so much that the la tte r dissolves readily in concentrated acetic acid, while the former does not. In th is point the present nucleic acids behave in a similar manner as a mixture of the two, in th a t some pentose nucleic acid precipitates out on the addition of an excess of glacial acetic acid, whereas the desoxypentose nucleic acid may be readily precipitated from the clear supernatant by the addition of ij. volumes of alcohol, DISCUSSION The protein-nucleates prepared from the S, IR and R strains of the species of Brucella contain relativ ely the same nitrogen content; the phosphorus content, however, is relatively much higher in preparations from the IR and R forms than in the 8 forms. This difference was also noted in the per cent of nucleic acid isolated. The nucleic acid in the smooth forms is approximately 23 to 30 per cent, whereas th at obtained from the variant and rough forms is approximately lj.0 per cent. By using the chloroform-gel method of separation of protein and nucleic acid, a protein which possesses specific precipitating povfers is obtained. The metaprotein obtained by hydrolysis with sodium hydro­ xide possesses very slight precipitating power. The protein obtained from the protein-nucleate is one exception to the common "test-book" statement that protein-nucleates or "nucleoproteins" consist of a protein, usually a hi stone or protamine, combined with nucleic -1 8 - acids. The Brucella proteins in no way resemble either protamines or histones, but do resemble many common proteins obtained from plant tiss u e s . In Table 6 are presented comparative nitrogen p artitio n data for some plant and bacterial protein fractions* The protein for which Johnson and Brown give analysis is that for the to ta l protein le f t in the b acterial c e ll a fte r lipide extraction and separation of nucleic acid; the tuberculin protein of Seibert is the protein present in a culture f i l t r a t e of Mycobacterium tuberculosis ; the others are a l l the protein fractions isolated from the protein-nucleate. Brucella proteins contain traces of sugars, which support the now accepted statement that a l l proteins have very small amounts of sugars present. The nitrogen to phosphorus ratios of the nucleic acids check well with the data given by Thompson and Dubos (3) on nucleic acid from Diplococcus pneumoniae, Sevag e t. al* (6) on the nucleic acid from Streptococcus pyogenes and with the accepted structure of thymonucleic and yeast nucleic acid. The per cent of purine nitrogen more closely approximates th a t of thymonucleic acid than yeast or pancreas nucleic acid* The purines guanine and adenine are found in approximately the ratio of 1:1. In the case of the pyrimidines, cytosine has been found to be present, but neither u racil nor thymine could be detected. The pentose content for the various nucleic acids ranged between iLj. and I 7 per cent, which is equal to about three fourths of the theoretical value for a tetranucleotide, and about one half when calculated for a hexanucleotide. The desoxyribonucleic acid or thymonucleic acid content. § ITS H -p u Pi cS to 6 r-i O rP ^ ret 00 0 CVJ O 43 rd — * O O ë §Ai l t n - j g k W I g % r-iW IW w O IfN KN o S O 00 rc\ LT\ ON ro\ ON -ê i C O: hC N ON o o UN r- o j: ■à to § * 43 0 U 1 o t ÎU «3^ Jh 0 to P 0 is; 0 •0 •H g 0 I M I 1 n .3 I I •H P 10 •H 0 p >> o -so- determined by the Feulgen reaction varied from 22.4 per cent for the acid prepared from the R strain to as high as li-6 .3 per cent in one of the nucleic acids prepared from a S strain . The dissolubility of the nucleic acid in glacial acetic acid tends to point out th at the nucleic acid studied is most lik ely a mixture of two separate nucleic acids. Further evidence of th is is shown by the data for the quantity of thymonucleic acid, the fact that the pentose values are about one half of what might be expected were i t a mixture of two separate nucleic rather than a single type of acid containing both pentose and desoxy­ pentose nucleotides. However, there have been several cases where there have been reported nucleic acids from bacteria that are a single acid composed of both pentose and desoxypentose nucleotides. Both Coghill ( 2 4 ) and Akasi (25) have found the presence of th is "mixed type" of nucleic acid in Mycobacterium tuberculosis and Eberthella typhi respectively. Since in the nucleic acids described in th is study neither thymine nor uracil were detected places i t neither in the so-called "animal t^rpe" (desoxypentose) nor "plant tj'pe" (pentose) of nucleic acid. However, i t is reasonable to conclude from the present data that the nucleic acids are of a special type of an unknoim nature, and should be placed in a class independent of other well defined nucleic acids. STMMARY A method is described for the preparation of the particular Brucella protein-nucleate; a method for separating i t into i t s components, protein and nucleic acid are also given. The protein-nucleate comprises approximately l4 per cent of the to ta l dry weight of the cells in the case of the preparations from S strains, and -2 1 - about 18 per cent in the case of the IR and R strain s. The protein component, which comprises about JO to 75 per cent of the protein-nucleate prepared from the S strain s and about 60 per cent of those prepared from the IR and R strain , is characterized by a nitrogen p a rtitio n (Hausmann numbers)* All proteins contain traces of sugars, but no amino-su gar. Guanine, adenine and cytosine, but no thymine or u racil, were found present in the nucleic acids. Both pentose and desoxypentose sugars were present and studied quantitatively. There is a soluble and insoluble portion when mixed with an excess of glacial acetic acid, and further data tends to show th a t there are two nucleic acids present; the other a lte r­ native, a nucleic acid composed of both pentose and desoxypentose nucleotides is possible but improbable. -2 2 - II. BIOLOGICAL ACTIVITY OF THE PROTEIK-HUCLEATES The studies of the ‘biological properties of the various proteinnuoleates were undertaken with the following objectives in mind: pré­ c ip itâ t i l ity , allerg ic activ ity ; to xicity in the guinea pig and an ti­ genicity. Protein-nuc leates prepared from the three strains of the smooth Brucella organism, on intermediate rough and one tru ly rough strain were used in these studies. Precipitation Studies The precipitation studies were made with specific serums prepared from an S s tra in of each of the three species of Brucella, one IE s tra in and one E s tra in of abortus. Normal rabbits were injected in tra­ venously with a 1 cc. suspension of living organisms, having a density comparable to 1 on the McFarland nephelometer. Wien the rabbits showed good sen sitiv ity to a skin t e s t dose of protein-nucleate, they were bled from the heart to obtain blood for serum. The collected serums were diluted 1:1 with physiological s a lt solution and s te riliz e d by passing through a Seitz f i l t e r . All precipitation te sts were made in small glass vials using 0.2 cc. serum layered with 0.2 cc. of the antigen dilution. The tubes of serum and antigen dilutions were incubated at hours and then read. for 2 I t was found that i f they were placed in the cold room for 2U hours a fte r being in the incubator, the precipitates dis­ appeared and could not be read. the cross precipitation studies. In Table 7 are recorded the results of From th is table one may readily see that a l l three of the S preparations of protein-nucleate are non-t^rpe specific. “23” Table 7* Antiserum from Rabbit Sensitized to abortus I 2 I47 (S) The Comparative Protein-nucleate Precipitin Reactions Protein-nuc1eat e from Dilutions of Protein-nucleate 1:1T* 1;2T 1:4T 1.-8T 1 j 16 t 1j 32T* abortus 230 (S) / / / / / / suis 1582 (S) / / / / / / melitensis 9 0 (S) / / / / / / _ — • — « _ abortus 230 (s) / / / / / / suis 1582 (S) / / / / / / melitensis 9 0 (S) ■/ / / / / / — — * _ — _ abortus 230 (S) / / / / / / suis 1582 (S) / / / / / / melitensis 90 (S) / / / / _ / _ abortus 8 O5 R (R) - - _ _ _ abortus 230 (S) — — — abortus 8 O5 R (R) - - — — abortus 230 (S) - — — _ suis 1582 (S) — — _ abortus 85 ( IR) abortus 805 R (R) suis 1630 (s) abortus 85 ( IR) abortus 8 O5 R (R) melitensis 2l\lU (s) ......... abortus 85 ( IR) . » suis 1582 (S) abortus 85 (IR) abortus 8 O5 R (R) melitensis 90 (S) abortus 85 ( IR) melitensis 90 (S) . _ — _ _ abortus 8 O5 R (R) abortus 85 ( IR) - - - - - *T = Thousand * = All precipitin reactions were negative at 1:6I|.T - , This re s u lt and the results with the intermediate rough strain check with those of Topping (8) and former studies in this laboratory (9j»10)* An experiment was performed to determine the effect of pH and aging on the s ta b ility of a protein nucleate prepared from a smooth stra in , Br. abortus 230# The protein-nucleates were made up and stored in b o ttles in the cold room at pH*s 3»5, 5*0, 6,0, 7.0 and 8.0, Samples were withdrawn at the end of 1, 2, 3# 5, 7 s-nd 11 months and tested fo r th e ir precipitating a b ility against an abortus goat a n ti­ serum. There was only a slig h t, i f any, deviation from the original precipitating power. Thus we find that pH and age do not appear to affect the s ta b ility of protein-nucleate solutions, a t least up to the end of eleven months# Allergic Activity of the Protein-nucleates The comparative allergic a c tiv itie s and cross reactions of Brucella protein-nucleates were determined on sensitized rabbits. Hormal rabbits were sensitized by an intravenous injection of a 1 cc. suspension of living Brucella cells suspended in s te r ile saline solution. The tu r­ bidity was made to compare with tube 3 of the McFarland nephelometer. This procedure w ill produce a high degree of skin reactiv ity in 30 days for measuring quantitatively the allergic a c tiv itie s of the various Brucella protein-nucleates* Each rabbit received on intradermal injection of 0.1 cc. of progressive dilutions of protein-nucleate ranging from 1:1,000 to 1:32,000. i+8 hour intervals. The intradermal reactions were observed at 2 Ï4. and The reactions were recorded on the forty-eighth hour, although they frequently persisted as long as five days# -2 5 - In Table 8 are recorded the collective results of several repeated experiments designed to show whether there were possible quantitative differences in the allerg ic activ ity of the various protein-nucleates against rabbits sensitized to the five cultures of Brucella. The allergic t i t e r s of the protein nucleates from the three S strains show approximately the same degree of response in the animals sensitized to the S strain s. The protein-nucleates used on i t s homo­ logous sensitized rabbit e lic its a slightly higher t i t e r with the ex­ ception of Br. suis, The suis preparation was found to have less bio­ logical a ctiv ity in a l l studies made on i t than the Br. abortus or Br, melitensis protein-nucleates. The IR strain provokes a reaction to a small degree in a l l the sensitized rabbits with the exception of the one sensitized to the very rough strain . The protein-nucleate prepared from the R strain induces no allergic response in any of the rabbits sensitized to the S or IR strains, but e lic its a feeble reaction in the rabbit sensitized to the R stra in . Antigenicity of the Protein-nucleates In th is experiment guinea pigs were used to study the a b ility of the protein-nucleates to provoke antibodies. The rapid agglutination te s t was used to measure agglutinin production; the phagocytic system previously described (12) was used to measure opsonin production. The pigs were given an intraperitoneal dose and bled on the tenth day after injection and examined for antibodies. I t has been found that antibody production was moderately stimulated in 19 out of 20 guinea pigs with 10 mgs, of antigen injected intraperitoneally. — 26— Table 8. Rabbit Sensitized to The Comparative Allergic Activity of the Protein-nucleates. Prote in-nu cl eate abortus 230 (s) abortus 230 (s) 12hnn« 12mm* 10mm, lOmm, 7mm, 5mm# suis 1582 (S) 8mm, 8mm. 6mm, 5mm. 5mm, 3mm* melitensis 98 (S) 8mm. 5mm, 5mm, 3mm, 2mm, 2mm, abortus 85 ( IR) 5mm, 5mm, Ipnm. 2mm, abortus 8 O5 R (R) suis 1582 (s) — — — abortps 230 (S) 20mm. 15 mm, 10mm, 7mm, 3mm, suis 1582 (S) 12mm, 10mm, 5mm, 5mm, 3mm, melitensis 98 (s) 12mm, 10mm, 7mm, 5mm. 3mm, 5mm. Ipom. 3mm, abortus 85 ( IR) abortus 85 (IE) — 10mm, abortus 230 (S) 10mm, 9mm. 9mm. 7mm, 5mm* 8mm, 6mm, 5mm, 5mm, 3mm, 12mm. 12mm, 10mm. 10mm, 5mm, abortus 85 ( IRl) 6mm. 5mm, Ipnm. 3mm, abortus 805 R (R) 10mm, 9mm, melitensis 98 (S) abortus 230 (S) ijmm. suis 1582 (S) l|mm. melitensis 98 (S) 3mm, abortus 85 ( IR) 5mm, abortus 8 O5 R (R) Ipmm. abortus 230 (S) — — - - - — _ _ _ - - — — - — _ _ — — 2mm* Ifinm, — iunm. _ _ melitensis 98 (S) _ abortus 85 ( IR) _ abortus 805 R (R) — - suis 1582 (S) abortus 8 O5 R (R) — abortus 8 O5 R (R) suis 1582 (S) melitensis 2ij.ll4. (S) Skin Reactions of Dilutions of Protein- nucleate 1-52T 1-lT 1-2T 1-16T 1-4T 1-8T 5mm, _ . .. - - - - - —27— I t is interesting to note that the protein-nucleate prepared from the R strain very mildly stimulated opsonins and agglutinins and cor­ respondingly, as we shall see, 3 out of 5 guinea pigs showed traces of antibody a fte r injection with the protein obtained from th e homologous protein-nuc1eat e• Toxicity of the Protein-nucleates The toxic property of each preparation of protein-nucleate was arrived a t by injecting guinea pigs, weighing from 300 to l\5 0 grams, in t r ape r i t onea lly with one to four cc, of the preparations, containing definite amounts measured on a dry weight basis. The c r ite r ia of toxi­ city are based on those observed by Pennell and Huddle son (23) when using preparations of endo-antigen; "The temperature of the animal begins to drop afte r the second hour, and may continue to do so u n til death occurs, or the temperature may rise again to normal i f the injection is not fa ta l. perature drops, the animal becomes nervous and ir r ita b le . As the tem­ The hair on the back becomes roughened, and the abdomen may become very tense and distended," Hoivever, of 15 pigs inoculated with varying amounts of proteinnucleate, only one showed a drop of more than 2®C. in temperature. The results would indicate th at the protein-nucleate preparations are non­ toxic • Biological Activity of the Protein and Nucleic Acid Each of the constituents obtained from the protein-nucleate, namely, the protein and nucleic acid has been examined for its a b ility to pre- —28— c ip ita te specific serum, allergic activ ity , for tox icity and for anti­ genicity# The comparative precipitating power of the original protein-nucleate and i t s isolated components is shown in Table 9* at the same time using a Br. abortus (s) These were a l l performed goat antiserum. The resu lts point to the fact th a t practically a ll the precipitating power lie s in the protein portion of the protein-nucleates, and th at the nucleic acid shows very low precipitating power and may be attributed to traces of Table 9» Comparative Precipitin Titers of Protein-nucleate, Protein and Nucleic Acid* Precipitin Titer of Antigen protein-nucle at e protein nucleic acid abortus 230 (S) 1 :3 2 ,0 0 0 1 : 3 2 ,0 0 0 1 :I f 000 abortus l 6 lX (S) 1:61:, 000 1:61:, 000 1 : 2 ,0 0 0 suis 1582 (S) 1 : 1 6 ,0 0 0 1 : 1 6 ,0 0 0 1 : 1 ,0 0 0 melitensis 2L9ii (S) 1 : 1 6 ,0 0 0 1 : 1 6 ,0 0 0 1 : 1 ,0 0 0 melitensis 2/.]25 (S) 1 : 3 2 ,0 0 0 1 : 8 ,0 0 0 1 : 1 ,0 0 0 abortus 85 ( IR) abortus 805 R (R) _ « k » TL..___ _____ *Prepared by method of Heidelberger and Kendall ( 11) fi( abortus goat antiserum was used in a ll titr a tio n s . protein ii^ich have not been completely separated from the nucleic acid, —29“ Allergie Activity of the Protein and Nucleic Acid Allergic a c tiv ity was determined in the same manner as previously described in th is paper. Table 10* The results are set forth in Table 10. Comparison of Allergic Activity of Protein-nucleate, Protein and Nucleic acid Skin Reaction of Dilution of 1-8T 1-16T l-l:T 1-lT* 1-2T Rabbit Sensitized to 1-52T abortus 250 (s) protein-nucleate 12 mm. 10 mm. 10 mm* 8mm. 5mm* 5mm* abortus 250 (S) protein 10 mm* 9 mm* 6 mm* 5ima* 5 mm* 5mm* 8mm* 8mm. 8mm* 5mm. 5mm* abortus 250 (S) nucleic acid melitensis 2l|25 (S) protein abortus 2^0 (s)i melitensis 21:25 (S) nucleic acid melitensis 2191: (S) protein - - — - - 10 mm* 9 mm* 8mm* 8mm. 5mm, 12 mm. 8mm* 6mm* Ipnm* 2 mm. melitensis 2 1^ (S) nucleic acid suis 1582 (S) protein suis 1582 (S) nucleic acid abortus 85 ( IR) protein 14mm. abortus 85 ( IR) nucleic acid abortus 8 O5 R (R) protein aborti.is 805R (R) nucleic acid — T = thousand — - 5mm* -5 0 - I t is quit© evident from the results th at practically a l l the allerg ic activ ity lie s in the protein, and th a t the nucleic acid portion of the protein-nucleate plays no part in provoking an allergic reaction. Antigenicity of the Protein and Nucleic Acid In the case of the protein-nucleates, a l l but those prepared from the R s tra in showed highly antigenic powers, the R stra in only very feebly e lic itin g antibodies. Thirty guinea pigs were injected int rape r i t onea 1ly with varjring doses (5 to 25 mgs,) of the protein fraction. Of these, IJ4. showed marked antigenic powers, 11 showed a moderate stimulation. The protein from the R stra in produced a very feeble response in the case of three pigs and none whatsoever in the other two. Thus the power of the protein to e l i c i t antibodies closely parallels that of the protein-nucleates. Seventeen guinea pigs were used to study the antigenicity of the n nucleic acids. Thirteen pigs sliovred no opsonins or agglutinins while four produced a very feeble opsonin response, but no agglutinins could be detected. From the above study we have concluded that a ll the antibody stimulating power is present in the protein portion of the proteinnucleate, and that the nucleic acid plays no part. Toxicity of the Protein and Nucleic Acid Toxicity was again determined by the method already mentioned. In the case of the protein fraction, 22 guinea pigs were subjected to varying doses of a l l five of the different proteins. only 5 showed any symptoms of toxicity. Of the 22, The three showing toxic -3 1 symptoms were each out of three different groups, each of which was injected with a different preparation. The isolated instances of toxicity could not he explained. In the case of the nucleic acid fraction 15 guinea pigs were used. All injected pigs remained normal. Thus we may conclude that neither the protein nor nucleic acid possesses toxic properties, DISCUSSION Precipitation te s ts on the protein-nucleates indicate th at the 8 strains are non-type specific in nature, and that those from IR and R strains do not have precipitating powers. The same type of results was obtained on the protein portion of the protein-nucleates. The nucleic acid component is inactive, thus the to ta l biological a ctiv ity resides in the protein portion of the protein-nucleate. are much in accord with prepared These results those of Sevag e t. a l . , in that a nucleicacid from streptococcal "nucleoproteins" by hydrolysiswith sodium carbonate were 'serologically in activ e;' that the protein portion 'reacted with streptococcal antisera in precipitation and complement fixation t e s t s '. However, they state that 'further work is in progress to clarify the serological specificity of these fra c tio n s.' Similar results were obtained by Ferramola (4) in his study of _B. anthracis. In comparing the allergic a c tiv itie s of the protein-nucleates there was found to be very l i t t l e , i f any type specificity. The IR and R stra in sensitized rabbits reacted feebly to any of the proteinnucleate preparations. Again, the protein component shovrs a l l of the allergic activ ity while the nucleic acid shows none. —52— In comparing the allerg ic a c tiv itie s of the protein-nucleates there was found to be very l i t t l e , i f any type specificity. The IR and R strain sensitized rabbits reacted feebly to any of the protein-nucleate prepara­ tio n s. Again, the protein component shows a l l of the allesrgic activ ity while the nucleic acid shows none. Both the protein-nucleate and its protein component moderately stimulate the production of antibodies, whereas the nucleic acid has no such power. The protein-nucleates, protein and nucleic acids were found to be non-toxic. SIB/1M ARY The results of precipitation studies with the protein-nucleates and protein components from S s tra in preparations show that they are not type-specific, and that the nucleic acid possesses no precipitating power. The protein-nucleate and protein component from the IR and R strain reacted only slig h tly with homologous and heterologous antiserum. Their nucleic acids also possessed no precipitating power. The protein-nucleate and protein component e l ic it non type-specific reactions in Brucella sensitized rabbits. The IR and R strain prepara­ tions e l i c i t only a slight allergic reaction. The protein-nucleate, protein and nucleic acid were found to be nontoxic • -3 3 - BIBLIOGRAHîY 1# Johnson, T.B, and Brown, E.B. "The Preparation of Nncleic Acid from the Nueleoproteins of Tubercle B acilli". Journal of Biological Chemistry 5 I;., 721, 1922, 2. Johnson, T, B. and Coghill, R. D. "The Distribution of Nitrogen in the Protein Fraction After Removal of Tuberculinic Acid." 3* Journal of Biological Chemistry 6 3 , 225, 1925* Thompson, R. H. S. and Dubos, R. J. "The Isolation of Nucleic Acid and Nuoleoprotein Fractions from Pneumococci." Journal of Biological Chemistry 125, 6 5 , 1938, I4.. Ferramola, R, "Nudeoproteins of Bacillus Anthracis." Ann. Assoc. Quim. Argentina 23, 93, 1933* 5, Mitra, B. M, "Nucleic Acid of Proteins in Vibrio Cholera." Indian Journal for Medical Research 2lj., 1, 1936* 6, Sevag, M. G,, Lackman, D. B. and Smolens, J. "The Isolation of the Components of Streptococcal Nuo1eopr ote in in Sero­ logically Active Form." Journal of Biological Chemistry 12k. k 2 5 . 1938. 7. Heidelberger, M. and Kendall, F. K. "Protein Fractions of a Scarlatinal Strain of Streptococcus Hemolyticus Journal of Experimental Medicine 54, 515, 1931* 8. Topping, L. "Carbohydrate and Nude oprotein Fractions Isolated from the Brucella Group." Journal of Pathology and Bacteriology XXXIX, 6 6 5 , 1934* - 9* 34- Huston, R, C*, Huddle son, I . F* and Hershey, A.D* "The Chemical Separation of Some Cellular Constituents of the Brucella Group of Microorganisms." Technical Bulletin No.137, Michigan Agricultural Experiment Station, 1934* 10. Hershey, A. D. and Huddle son, I . F. "The Chemical Examination of an Avirulent Strain of Brucella Abortus." Technical Bulletin No* l49, Michigan Agricultural Experiment Station, 1 936 . 11. Osborne, T. B. and Campbell, G, F. "The Nucleic Acid of the Embryo of Wheat and its Protein Compounds." Journal of the American Chemical Society 22, 379, 1900, 12. Huddle son, I . F. "Brucella Infections in Animal and Man." Commonwealth Fund, New York. 13. Hewitt, L. P. 1939* "Chemistry of Antibodies and Serum Proteins." The Biochemical Journal XXVII, 2080, 1934* 14. Thimann, K. V. "A Micro-method for the Determination of the Hausmann Numbers of Proteins,” Biochemical Journal 20 , 1190, 1926, 15 . Sorenson, M. and Haugaard, 0, "Tiber die Anwendbarkeit der Orcin reaktion zur Bestiramung der Art und Menge von Kohlenhydratgruppen in Eiweiss- stoffen." Zeitschrift 16. 26o, 247» 1933* Palmer, J. W., Smythe, E. M. and Meyer, K. Hexosamine." Biochemische "The Estimation of Journal of Biological Chemistry, I I 9 , 491* -3 5 - 17* Levene, P. A, and Bass, L, W * "Nucleic Acids," Catalogue Company, Inc., New York. 18. Jones, Walter. "Nucleic Acids." London. 19* The Chemical 1931* Longmans, Green and Co., 1920. TWieeler, H. L. and Johnson, T. B. and Cytosine." "On a Color Test for Uracil Journal of Biological Chemistry 3, 183, 1907* 20. Harkins, H. H. and Johnson, T. B. "A Test for Thymine and 5- methyl cytosine in the Presence of Uracil and Cytosine." Journal of the American Chemical Society 5 I* 1237* 1929* 21. Hoffman, W. S. "&Iicro-method for the Determination of Pentose in Yeast." 22. Widstrom, G. Journal of Biological Chemistry 73* 15* 1927* "Uber die Verwendbarkeit der Schiffschen Fuchsin- schwefligsauereaktion zur quantitative Bestimmung von Thymonucleinsaure." Biochemische Zeitschrift 119* 298 , 1928. 23. Pennell, R, B. and Huddle son, I. F, "The Chemical Constitution and Biological Properties of the Endo-ant igen of the Brucella Group of Microorganisms." Technical Bulletin No, 156 , Michigan Agricultural Experiment Station. 24. Coghill, R. D. "Nucleic Acid of the Timothy Bacillus." 1937* Journal of Biological Chemistry XC, 57* 1931* 25. Akasi, S. "Nucleic Acid of Typhoid Bacillus of Mice." Biochemistry of Japan 28, 355* 1938. Journal of -3 6 - 26, Kiesel, A, and Belozersky, A. "Uber die Nude ins anre und die Nude opr ote ide der Erbsenkeime. " Physiologische Chemie 228, 27, Seibert, F« B. and Munday, B* l6o, Zeitschrift fur 1934* "The Chemical Composition of the Active Principle of Tuberculin." Journal of Biological Chemistry 101, 7 6 3 * 1933* 28, Hoffstadt, R. E. and Clark, W. M, "The Chemical Composition and Antigenic Properties of Fractions of the Smooth and Rough Strains of Staphylococcus Aureus. Diseases 62, JO, 1958* Journal of Infectious -3 8 - ACKITOYJLEDGEIÎENT The author wishes to express his grateful appreciation to Mrs. Myrtle Munger of the Central Brucella Station for conducting the blood examinations mentioned in the study and to thank Dr. I . F. Huddle8on and Dr. B, B. Pennell, also of the Central Brucella Station, for th e ir helpful suggestions and technical assistance.