Studios concerning Chemistry and Immunological Pr-rties of Pneumococcus --- Report of DrAvery, with Drs.Heidelberger,Goebel Tillett,Julianelle,and Ddvrson. .--- __.______ ----_--- -- STUDIXi ON ANTIGENIC DISSOCLcTION: --- 1. Pneumococcus as "Complex Antigen" 2. Consideration_of the Cell as a. Two distinct and Separate antigenic systems. b. Single antigenic complex composad of carbo- hydrate (haptene) and protein. 3. Evidence for Antigenic Dissociation in Vitro? -v a. Qualitatively differont antibodies stimulatod by intact and dissolved cells. b. Relativa differonces'in the dissociation of Typos I, II, and III. 4. Evidence for fintigonic Dissociation in Vivot 8. b. 0. d. 5. Factors relating to tho knimalt a. Natural rosistonco and antibody rosponso. b. kture of bnctorial injury. c. Difforenco botwoon Natural Rosistznco and Specific Immunity. 6. Factors Relating to tho Micro-organism. a. Rolation botwoon antigonic stability and chemical structure of tho celli 7. Concept of Virulence: a. As tissue fastness. b. Relationship between Virulence and Antigenicity. Dissociation of Type III in rabbits. Antiprotoin antibodies in sarum as indox of dissociation. Ralation of `antigonic dissociation to pro- duction of antipnoumococcus sera. hntigonic potancy invorssly proportional to rat.0 end oxtont of dissociation. Investigations on the immunological relationships of the carbohydrate and protein of Pneumococcus clearly indicate that the bacterial cell functions as a complex antigen, the components of which are subject to dissociation. Moreover, the nature of the immune re- sponse of animals experimentally infected or artificially immunized with Pneumococcus is conditioned by principles which govern the rate and extent of antigenic dissociation under given conditions. Cell disintegration, whether brought about'by physical (freezing and thaw- ing), by chemical (bile and alkalis), or by enzyme action (autolysis), is always accompanied by partial or complete loss of the capacity of the dissolved material to stimulate type-specific antibodies. Al- though in cell solutions the specific polysaccharide (haptene) exists free and unaltered in specific reactivity, in this form it is com- pletely devoid of the power to provoke antibody production. Since in the natural state in which it is present in the intact cell this same carbohydrate substance ie not only highly specific but extraordinarily efficient as antigen, and since it loses its antigenic property to a greater or less extent whenever cell dissolution occurs, the conclu- aion seems inevitable, that cell disintegration Is accompanied by antigenic dissociation, in which process the soluble, specific sub- stance is split off from some Other constituent with which it forms the dominant type-specific antigen of the cell as a whole. It like- wise follows that morphological integrity is essential to the fullest expression. of the antigenic value of the cell. Although by reason of antigenic dissociation solutions of pneumococci may fail completely to elicit the type-specific antibodies, such solutions always stimu- late the formation Of antibodies which are qualitatively different in . , 222 ,I ,, `) .wr _ : . . 1 !' that they react only d th the ~~rotoin of the ccl1 :7ithOst regard to ty:3o-qeciZicity; that is, they are s-gccics but not tyrc-s?eciCic. 5 !) An irmuce semm contbining on* the anti-?roteiR irzr~e bodies ;me- / cipitatcs s01uti0m of V.-x isolated 2rOtoin regardless of ?.ts tye derivation md e,gglutirates all degri;tdcd "2" cells which bg %iLt=lral or cha~ical meazs have been stri?yed of the ty:>e-specific szbs tmce. Tixse same protein mtisera, hocover, are not reactive with enca? su- lated pneI.Xxococci or ;rith pure solutions of specific bacteriel pOb- saccharides. Further;;orc, such 6era afford 30 protection to ZiCe a- \? !i gainst infection vi th virule;lt organisms of t3e fixed types. In .: 1 other words, peumococcus protein antibodies are umolated to tke type-qecific mechanien of the cell and pertain only to the broader protein-antiprotein reactions of the species. These facts admit of at least two inter-gretztions. From observation6 tAich `have been regcatedly collhined experimentally, it ie evident that both tke carbohydrate ard protein I of pneuuoaoccus share in the ideological mechanism of the cell, either as two distinct and indmendent Et1 tigenic systexts, or as a single omplex in which carbohydrate and protein together form one antigenic unit. If these two constituents, both integral parts of the cell as a whole, be considered as forming tm separate and un- related antigens, tfien that qystm comprising the carbohjrdrate fmc- tion may be referred to as the ectcplasmic, capsular or type antigen, and the other, involving the protein of the cell *DO&?, nwr be spoken of as the endoplasmic, somatic or species antigen. Since the specific polysaccharide (haptens) is by itself non-antigenic, it follows that in the system of which this material forma a part the carbohydrate i I 223 does not exist as Such but in combination `;7ith sotle other substauce :d.th r&i& ft 2orr~ an a2tigeuic co~@ex each cqoncnt of which is separately inert. !!%at is, once this union has beer disrupted the i ) b \ antigenicity of the whole is lost and only the s;>ecific antibod;r binding-gropert; 7 of the dissociated haptme renains. This typ-spc- if ic subste-?ce of tke cell has non. ' been chaicalIg identified as a ?olyaaccharide and can be recognized in the s:rsterz after dissociatioti by its specific reactivity nith the type antibodies induced by the intact cell in which the tTo components still renain undissociated and hence antigcnic'. !Pe nature and properties of tha r"actor in this primaxy sys te3 ~hicl; enters into combination r?i th and confers anti- geuicity upon t%e qecific polysaccharide are unknown. According to the conception of @double antigen", the second and equally independ- ent antiGer& system of t3e*cell nould consist solely of the bacter- ial protein. This substazze, coiistituting tile larger ?art Of tile cell body, functions antie;enically precisely as do the isolated nucleo- s protein and the dece?sulated R s cells in stirn;lating only the a?Iti- protein antibadies which charactcriee the species. A simpler and perhaps more likely interpretation of the dual antigenic function of ths intact cell is the concept that the iacter- ial ;solysaccharide and protein exist not as individual mmbers of eeg- arato sqstezls, but as a sin&e carbohydrate-protein co@e::, the spec- ific antigcr~icity of *ich is detemined by t2.e presome of t?e t;Te- speclf ic polysaccharlde. 1:: otAer words, the bacterial cell as an anti- genie usit consists of a coxbinetion of nucleo;?rotein ar,d carbohydrate is1 which the latter deter4aos tire ty?e-specificity of the whole. %`hiS compound antigen is dissociable and the ease with which t:m lirkqe between the tno cellular constituents is disrupted varies ?Iith each of the fixed t;rpes. When dissociation occurs, as it does to a greater or less extent whenever dissolution of the cell takes place, the specific (polysaccharide) haptene is split off leaving only the protein frac& tion nhich then functions as a secondary antigen. This conception is supported by experimental evidence; although solutions of bacterial cells, in which complete dissociation has occurred, contain the total content of body substance present in the formed elements, these solu- tions are devoid of the property of stimulating the type-specific an- tibodies r;hich characterize the immune response to the intact organ- isms . . Immunization with these solutions results only in formation of the secondary antinrotein antibodies. The evidence presented thus far concerns antigenic dissocia- tion in vitro, as revealed by differences in the antibody response to intact and dissolved nneumococci. There is also considerable evidence that similar dissociation goes on in the animal body after the intro- duction of the whole cell. The most striking example of the occurrence of this phenomenon in the animal body is the character of the immune response in rabbits to immunization with Pneumococcus Type 11X. Repeat- ed injection of encapsulated Type III pneumococci fail in the great majority of instances to elicit any type-specific antibodies; the se= of immunized rabbits neither agglutinates the encapsulated'cells, nor precipitates`the homologous specific substance, nor affords protection to mice against infection vith the virulent organisms of Type III. Indeed, such sera, if tested solely for the presence of antibodies of the type-specific variety, would be considered devoid of all immune bodies. However, these sera are rich in antiprotein antibodies; they not onI7 grecili tate pro tein-contz5ni2- *L solutions of ~no1LCococci: but agglutinate t're decap$ated bacteria (VP foz?.ls) . This result Cm OLd:J be intrqretcd as evidence tkt ix t'is mimal `7ody t3c szine or a air;?ilar Process of dissociation goes OT. as that rhich occurs Mth disrxqtian 05 the cell in vitro. These observations, 20~: sxqqortod ?~y ex?erix:~t~?l evidence, furnish a i:asis for understz~ding t:ie difficulties eucountor- ed in attempts to produce an efficient zltiserum for We III ;>newo- cocccis. I;ldeed, the principle undarlJing thfs nhmoaenor? is ,)eriu>s applicable to the other tr:>es as nell. If seem 3ct mlikelg tkt t?:e relative differeilces in tl:e Totenc:; of bmun9 sera, the efZectivezess of which is greatest in !&pe I, less in Ty2e II, md least in T::;>e III, zxv be referable to difieremes in t!;e ease of dissoCiati.oi~ of t>e specific antigen complex in each type. It is interesting in t%is con- nection $0 recall the fact that, although there ie a ?ro:.ressive iu- crease in the amounts of soluble s2ecifi.c subsfance elaborated by Pr;3es I, II, III in the order ila;aed, t'kere 1s a corresAJondi;li: decrease ir- the a?Jti~nic efficiency Of tb trlroe types. The spec?.f ic polyeaccharides in each instance are `kno~~l to be chemiccLly distinct substances aid these differences in chemical constitution Iw account not only for the specificity of Back*, but may also detezine t'ne rate and canpleteneise sit'n which the antigeaic complex OP each undergoes dissociation. Accord- ing to this view, the most efficient antiG;ea is the one least eaally dlssociable: that is, the antigenic potc3:zy of a given tye is inversely proportional to the rste and extent of dissociation. Like all antibacterial processes in the Sody, the pheno:Alenon of antigenic dissociation is deterLzinedby tco indewndent but ixterre- lated zrou?s of factors; those pertaini:%; to the animal body a;id those 226 relating to the micro-organisms. E'actors relating to the animal: These comprise the antibac- terial properties which are possesled naturally by the animal organism and which together constitute the .vaguely defined quality of bodily re- sistance. Since in the case of the Type III pneumococcus the rabbit can tolerate nith impunity ten million times the dose fatal to a mouse, it is apparent that this result is determined by differences in the normal resistance of the tx animals to the same organism. Honever, to say that the rabbit possesses natural resistance to infection y:ith Pneumococcus mucoshs, and that the mouse does not possess this proper- ty, merely restates the problem and fails to explain the facts. If, on the other hand, the evidence in favor of the dissociation hypothesis be accepted, and this newer concept of the bacterial cell be applied to processes occurring xithin the body, then differences in animal resist- ance may be interpreted in terms of this dissociation process. There is immunological evidence that dissociation involves an injury to the capsular mechanism of the cell, nherebp t'ne potentially virulent organ- ism is reduced to a form resembling in antigenic and infective proper- ties those cells r;hich have been degraded W1' forms by growth in vitco, Just as degradation of the encapsulated organism by cultural methods is accompanied by loss of virulence and type-specificity, so the injury inflicted on the bacteria by the tissues of the resistant animal results in a loss of invasiveness with the liberation of the common protein anti- gen. Both results are the expression of the same phenomenon; injury to the capsular mechanism not only exposes the denuded cell to phagocytosis, thus accounting for the lack of infection, but this same injury, by dis- sociating the "St substance, discloses the undifferentiated protein of 227 ,.I, -3 -.I -, the cell, thus accounting antigenically for the presence of only anti- i'f 1 i j i protein antibodies in the blood serum of the resistant animals. Although resulting from the same process, the survival of the animal.is not neces- sarily dependent upon the concomitant development of protein immunity. The formation of the antiprotein irtxnune bodies is in all likelihood mere- ly a secondary reaction dependent upon the presence of the dissociated protein. This view seems the more likely since purely antiprotein sera do not confer passive protection on animals susceptible to infection. Although the nature of the bacterial injury which affects the invasive and antigenic function of the cell is, as yet, unknown, the re- sult is two-fold; first, the survival of the animal in the presence of a potentially virulent organism, and second, the development of a pure- ly antiprotein immunity by a cell potentially capable of stimulating the type-specific, anti-carbohydrate antibodies. !Ihe loss of virulence and the depreciation of the antigenic value of pneumccocci nithin the . body of the resistant animal are, in outcome at least, analogous to the degradation of these same tno functions crhioh is induced by growth in an unfavorable environment. Animals of different species and individual members of the same species vary in the ability to inflict the cell injury which leads to antigenic dissociation. The presence of this property in the tissues determines the animalst resistance to a particular micro-prganism, as in the case of the rabbit to Type III Pneumococcus, and the absence of this hroperty determines the animals* susceptihili ty, as in the case of the mouse to the same strain. Quantitative differences nould account for the variations observed in the susceptibility of individual animals. Natural resistance, whereby an animal disarms the invading pneumococci 228 by stripping the capsule, thus robbing the cell of infectivity and type- antigenicity, is an innate quality rrhich differs from acquired immunity in that the serum of the resistant animal is devoid of type-specific antibodies and fails to confer passive protection against infection in sus cep t ible animal s . Reference has already been made to the relative differences in the potency of immune sera of Types I, II, and III. The reason for this progressive decrease in potency of the respective antisera nas as- cribed to differences in the ease of dissociation of the homologous an- tigen. According to this vien, then, the amount of antigenic dissocia- tion occurring both before and after t'ne injection of the imunizing material into the animal body determines the proportional content of the type-specific and the antiprotein antibodies in the serum. Con- versely, the relative titre of these t?o qualitatively different anti- bodies serves as an index of the comparative ease sith which the spec- ific antigenic complex of a given type undergoes dissociation. In an- imals of the same species the factors of natural resistance which re- suit inantigenic dissociation are not $q@ly effective against all types of pneumococci ; that is, what has been described as the reaction of the normal rabbit to Type III does not obtain in the case of pneu- mococcus Type I, Factors Relating to the Micro-organism: Differences in the immune response of. the same animal to pneumococci of different types suggests the operation of factors peculiar to the micro-orgcaisms themselves in addition to those previously described as relating to the animal body. In accordance with the dissociation theory, these bacterial properties are referable to differences in th(, ease with Thich the antigenic-complex of the specific types undergoes dissocia- tion'in the presence of the resistance factors of the host. That is, dissociation, like specificity itself, is dependent upon the particu- lar chemical structure of the complex antigen of each type, Since the bacterial polysaccharides, ahich define specificity, are separate en- tities as distinctive in their chemical properties as they are specific in their serological reactions, the factors determining antigenic sta- bility of any given type of pneumococcus are inherent In the chemistry of the cell itself. For instance, in the case of Type III pneumococcus, the least stable of all type antigens, the linkage of the carbohydrate complex is more easily disrupted and complete dissociation of the type- specific polysaccharide leaves only the protein to function as antigen. This intrepretation explains the hitherto inexplicable fact that immun- ization of rabbits nith Pneumococcus Type III results only in the form- ation of antiprotein antibodies. In the case of pneumococcus Type I, on the other hand, the more stable union of the antigenic complex re- sists cleavage and the intact antigen functions in the stimulation of type-specific antibodies, The potency of antipneumococcus serum as measured by the content of type-specific antibodies is conditioned, therefore, by a balance between the factors of the animal body nhich bring about antigenic cleavage and those properties of the bacterial cell which deter&m the stability of the specific antigonic complex of each type. It is interesting to apply the same concept to the problem of bacterial vi rulence. When an animal does not possess in adequate amount the factors of natural resistance or when even in their presence the bacterial cell is so constituted as to make difficult or impossible the dissociation of the cellular components so that the Pneumococcus retains its capsular mechanism unimpaired, is it not possible that this natural or acquired resistance on the part of the micro-organism is the dominant factor in the phenomenon of bacterial virulence? And, nhen the virulence of an organism is enhanced by rapid and repeated passage through animals for which it originally possessed little virulence, is it not possible that this property consists in the acquisition of an increased resistance on the part of tne cell to the dissociation factors of the animal;- an adaptation or form of "tissue fastness" acquired by the micro-organism? According to this viea, the lack of virulence of Type III Pneumococcus for normal Rabbits, and the absence of the type-99 cific antibodies in the serum of resistant animals nould each bs referable to the same cell injury, t-hat is, to the process of cell dissociation. If this interpretation is correct, then by rabbit passage tl;e pneumo- coccus mucosti should acquire increased resistance to dissociation and the acquisition of this "tissue fastness11 should be accompanied by increased virulence and the capacity to stimulate the type-specif- ic (anti-s) antibodies which is a function of the undissociated antigen of the intact cell. In this hypothesis perhaps lies the explanation of the fact empirically discovered, that the more virulent the organ- ism used for immunization the more potent and specific is the anti- serum; a result dependent not merely upon the fact that the virulent cell elaborates more of the type-specific substance, but that with the accession of virulence, the cell becomes increasingly moro resistant to cleavage, thus conserving in its effective state the dominant type- specific antigen. 231 .,,)y 1 6; . . Application of the Principles of Antiaenic pissociation to,Productiop af Antinneumococcus Serum: (brs. Avery and Julianelle)6 In the.preceding diseussion of the theory of antigenic dissoci- ation attention has been drawn to the significance of this phenomenon in the production of potent antipneumococcus serum. Since the evidence thus far available clearly indicates that the most efficient antigen is the one least easily dissociable and that the antigenic potency of any given type of pneumococcus is inversely proportional to the rate and degree of dissociation, attempts have been made to increase the type- specific antibodies in innnune sera by special methods designed to pre- vent dissociation of the type-specific antigen. Intact, encapsulated pneumococci, bearing their full comple- ment of type-speciffc carbohydrate, have been treated with various chamical reagents with the hope of "fixing" the cell SO as to co nserve intact the effective antigenic complex by rendering it less susceptible to those factors which bring about dissociation. By reason of the them- ical nature of the type-specific carbofiydrate of `Pype I, which'contains an amino sugar group, fomol was chosen as nfixativett. The method em- ployed and the results obtained are as follows:- gethod: Young cult- of an %nimalized~ strain of Type 1 pneumococcw were grown in plain broth fbr 8-10 hours. The culture was divided into two equal parts- One portion was heated at 560 C. for one hour and the other half was killed by the addit.ion to the culture fluid of formaldewdo in final concentration of 0;2 per cent. Two series of rabbits were immunized according to the method of Cole and Moore, that is,. 1 cc- of bacterial suspension was given 1~ travenously every day for one week followed by a rest period of one week, until the'animals had received three courses of injections. One group of rabbits was given injections of heat killed culture, while a a second group was given the nonheated, formalinized culture. Before each course of injections blood was drawn from the ear and the serum tested for the presence of type-specific (anti-NV) and species spec- ific (anti-clPct) agglutinins. The accompanying protocol shows the results of the immuniza- tion. TABLE I Type Specific Antibodies. Species Specific Antibodies 1 Rab- 1 bit Heat Formal- killed inized I 0. antigen antigen No. 1 antigen antipen After 1 1 Negativ 1:8 7 # course 2 n of in- I 3 II / jections 4 II , 5 'I , 6 II I t I After 2 1 1:5 1:160 1 7 i 1:80 Negative cour8ea 2 1:lO 1:160 I 8 1:160 1:40 of in- 3 1:lO 1:180 9 * 1:80 negative jectiom '4 1:lO 1:180 10 1:40 lr20 : lr80 6 1:1&O 11 1:160 1:4Q i 6 1:lO ` 1:160 12 I 1:160 1:20 *1 *7 1:80 1:640 8 1:640 1:160 1:160 1:1280 9 1:640 1!160 l&O 1:640 10 1:640 1:80 1:ao 1:640 11 1:640 1:40 lr160 1:1280 12 1:320 1:40 * Rabbits #l and #7 were found dead Ratio of antG3' to anti-`Pi'antibodies in sera of rabbits immunized with heat-killed and formol-treated pneumococci (type I) (y 600 T 0- I I / / 0- 0' , ' Weeks 1 2 3 0 1 2 3 . Period of immunization 800 60C 3 s 5oc c .4 C 3 & 4oc 7 3oc 2oc 1oc . C 232 b Comparison of the immune response in rabbits to heat-killed and formol-treated pneumococci (type I) I I I I I I I I He4!' I , 1 2eeks 1 2 3 0 1 2 3 Period of immunization The results are sunanarixed in Table I and graphically presenb ed in Figures 1 and 2. Formalized cells give rise to type-specific an- tibodies more promptly and the titre at the end of the period of imrmrdz- ation averages over eight times the titre induced by the heat-killed , vaccines. Moreover, the striking difference in the proportional content of the type-specific and antiprotein antibodies is evident in the inverse ratio of these two qualitatively different antibodies. (See Figures land 2). If th8S8 differences are interpreted in terms Of antigenif dissoci* tion it 808me not unlikely that methods, based on the chemical require- ments of each specific type, may b8 found which will serve to inhibit the rate and degree of dissociation and thus tend to NstabilizeH these complex antigens. The increased potency of th8 formalized antigen of Type I pneumococcus, a method not applicable to the organisms of Type II and III, is at least suggestive that with appropri`ate reagents similar results may eventually be obtained with the other fixed typos. Observations on the occurrence of the becific Types of Friedlander's Bacillus in Dieease. (Dr. Julianelle). In the preceding report three epecific type8 and a heterogeneous group were described among the Friedlander bacilli. The encapeulated fOImS were type spec- ific, but by artificial methods they were rendered capsule-free and were then found to be only species specific. Recently, nine etrains of the Organism have been referred t0 us for typing. Six Of th8S8 strains were found to fall in Tspe A, one in Type B, and two in Type C. Three of these cultures were associated with%hronic human infections and all were found to be of a composite nature - that is, they included both "S" encapsulated and IQ" capsule-free, forms. In four Instances, the organisms were isolated from in- .- dividuale aufforing from lobar pneumonia due to Friedlanderla bacillus. In a fatal case of Friedlande/r pneumonia in this hospital the opportun- ity w!s offered for studying the type specific precipitin reaction in the urine. The reaction to type was prompt and ocarrred in a dilution of urine of 1:16 on the second day, and 1:64 on the third day of the disease when the patient died. Thus confirmation was gained of Blake's observations of a specific precipitin reaction in one case of pneumonia due to Friedlander's bacillus and of our own observations of a similar reaction in experimentally infested rabbits. The following table shows the distribution of the apec- lfic types of Friedlander `bacilli eummarized to date. The occurrence of Specific Types of Friedlander's Bacillus. I Total number of / TypeA Qv B `Pype C Group X I [ Strains studied 39 21 7 I ,5 t 6 j Type A - 21 a trains : 19 from lobar Pneumonia, 1 from extirpated ademids, 1 from cystitis. TypeB- 7 strains: 3 from guinea pigs (pneumonia) , 2 from lobar pneunon- la (human), 2. from horeee (genito-urinary infection) ., Type C - 5.8 train6 t 2 lobar Pneumonia, 1 from antrum infection, 2 source 1 UUkZlOWZl. 4 roup X - 6 strains: I 5 lobar Pneumonia, 1 from faecea (Pellagra) Although the total number of strain8 of Friedlander bacilli studied so far is too small to furnish conclusive data on the distri'tution and relative frequency of occurrence of the specific type8 in infections, it is, nevertheless, interesting to observe that of 28 isolated from lobar pneumonia in man 23 or 82$ belonged to one or other of the three fixed 235 types. The frequency of Type A'infection in Friedlander pneumonia in human beidgs is shorm by the fact that this type nas isolated from 19 (68%) of 28 cases studied. From three different epidemics of Fried- lander pnemonia in guinea pigs, three strains vere isolated all of which belonged to Type B. Moreover, 5 out of the 7 strains belonging to thie type were from animal soutcea, the remaining two being assoc- iated nith pneumonia in man. A Sttiv of Complex Antigens. (Dr, Julianelle) . Previous studies from this laboratory have revealed the interesting fact that organism8 of widely different species may exhibit striking reciprocal immunological reactiona. Thia was found to occur in the case of Fried- lander's bacillus (Type B) and Pneumococcua (Type II). Chemical stud- ies had shozn previously that the soluble specific substances of the two organism8 possess certain chemical properties in common, and it was thought at the time that the mutual immunological behavior of the two etrains Tas referable to the chemical similarity of the Soluble Speaific Substancea. The unique relatiOn8hip8 of the two species of bacteria afforded an opportunity of analyring the nature of complex antigens 7 in term8 of the type specific carbohydrate - or Soluble Specific Sub- stance - and the species specific protein. In this nay, light might be thrown upon the relative importance of these tno cell constitu8nta in the determination of type specificity. Accordingly observations were made on the serological relationships existing between the poly- aaccharide and Ir o tein , and between the encapsulated IBSEN cells and capsule-free W ccl18 of Pneumococcua Ty?c II and Friedlanderls bacil- lus, Type B. 236 . T:.B ' k*b The results of this etudy shorJ t&at the encaIwiiated (S) cells of both species `a,re agglutinated in anti-S sera of either organism. !l?Y;le capsule-free cells (R) on the contrary are different for each species and react only 71th innnune serum produced by the injection of pneumococci or Friedlander bacillus, respectively. It is significant in this connection that S cells differ from the correspondin g R cells only in the possession of capsules and the characters which accompany capsules, such as virulence, type specificity and elaboration of Soluble Specific Substance. In other ;:ords , these tw organisms of bilogically remote species, when encapsu- lated are endowed with similar immunologicel characters; but ahcn they are devoid of capsules, they act as difforent and unrelated antigens. Since, as has been shots previously, the type specific carbohydrates are chemically and eerologically similar, this fact constitutes direct evidence, that the inununological identity of encapsulated bacteria do- ponds upon the chemical nature of this capsular substance. It was not possible to shon any serological relationship be- tween the protein8 of the tno species by cross precipitation reactions. %%I8 fact, together nith the lack of reciprocal relations betneen the degraded R strains derived from bOth8peCie8 (Table I) thro:Ta the burden of type specificity of complex antigens on the Soluble Specific Substance. In other mrds, it appears that the proteins of encapsulated Sacteria play .little or no part in type specificity and that this property re- . sides in the capsular or carbohydrate constituent of the cell. It further appears probable that when the analogous specific polysacchar- ides of othemiae totally unrelated microorgiinisms correspond sufficient- ly in chemical constitution an immunological correspondence also results, Moreover, tho present sturQ lends supporting evidence to t'ne opinion . . - --. td F I 2 ri `C p3 D !G -5 I -.- J -_ I _- I -- I - -- i 1 _- - I I -. _ I . -._ I a_. I -- i .- i .-- $ $ -.. r .- -. I ,e `6 rH `l-l A c -- --__ -.. - ._. ._ _.- .--. - -- -- t 6 0 s C-J ii - .- t - I _ I ._. I : -. $ $ .-. $ $ -- $ -. I --.. 4 .I_. 1 -- I -.- I - _ I -. I -- I - I bf . . . --_ -. .- _ . . I . . _ -._ 1 - _. __& b i 2 In [ 1 I-. $ t + $ + $ + + -.- I -^ I I -. - I -. f + f ; .? j I I _.- I __ I - m j j I. , L 1 I-l I IJ , I .---. .-- _..- - . -..- _ .-.- . -.- I i' f $ -. + _.__ I __ _ I I - . I 237 ,..T. 1 previously expressed thati the type-specific character of the antigenic response is dependent almost entirely on the nature of the polysacchar- ido and not upon the substance to shich it is attached. Consequently since the specific carbohydrate substance of ttie Friedlander Type 3, and Pneumococcus Type XI possess in common similar chemical properties the type specificity and antlgenicity of each i`s similar even though, as the present work shoas, the proteins in each instance are quite dis- similar. Studies on Immunity to Pneumococcus t~ucosus (Type 111). (Dr. Tillett). In two publications now in press, certain phenomena with regard to the antigenicity and infectivity of Type III pneumocoo- ci for rabbits have been described. By way of recapitulation, the re- sults of those experiments will be briefly summarized;- It was found that the immunization of rabbits nith Type III pneumococcl failed in a great majority of instances to stimulate the production of type specific antibodies, but was effective in producing antibodies, react- ive with nucleo-protein common to all pneumococcl, and capable of ag- glutinating any R strain of pneumococcus. These results are, interpret- * ed as meaning that rabbits possess some mechanism which is capable of . altering the antigenic complex of the Type III pneumococcus so that the type specific component is made ineffective, Since type specific- ity resides in the soluble specific substance of the capsule, it seems evident that the injury 3nflicted on the organisms involve: the cap- SUlO. Further evidence in support of this interpretation of the im- munological response of rabbits is brought out ghen living Fne III pncumococci are injected. It was found that these organisms, al though highly pathogenic for mice and possossed of large mucoid capsules, are avirulcnt for rabbits in doses of 2 cc. to 5 CC* and some times 10 cc. Since encapsulation and virulonco arc known to be intimatoly assoCiate i od, it soomcd possiblo that tho mothod Thereby rabbits rcndcrcd the Type If1 avirulent rostod on tho same machonism which is rosponsiblo for tho desctruction of tho typo specific antigenicity of the cell, and that as a result of the injury inflicted on tho capsule of tho living organism, pirulonco is lost. An attempt to understand the method ghorobg rabbits dispose of encapsulated !Pype III pnoumococcl intravenously injected led to a study of the bacteremia prodwxd; and for purposes of comparison non- ancapsulatod V forms of pneutnococci were also omployed. Neithor of these organisms produce fatal infection in rabbits. The prom?tncss 7ith rrhich R forms aro phagocyted folloaing introduction into tho cir- culation affords adequate explanation for tho rapid disappearance of these organisms. HoTever, the encapsulated 1's" forms aro not phago- cyted and the bacteromia folloning lnjoction of organisms of this character rum an uneven course which persist8 for several days before complete disappoaranco. A cwrisoa, then, of o&nts following in- jection of @RN and sS" forms of Type III pneumococci shots striking dif- f erences. The `3s forms am readily phagocyted and disappear in a few hours. The "$s forms aro not phagocyted in their encapru.latod state and they persist in the blood stream for several days before final re- covery of the animal. It is necessary, therefore, to ascribe the nat- ural resistance of rabbits to Type XII pneumococci to some, as yot, un- . determined factor; Of the strains of Type III used in these experiments ono was made highly virulent for rabbits by means of rapid anim31 passage. . 240 LI .-I .-,:-i / -The bacteremia produced by this strain Tas oharacteristic of the sept- icemic curve, previously described by others, which consists in a sharp decrease in the number of circulating organisms follosed by a steady in- I crease until the death of the animal. This strain is identical in every other respect Rith the rabbit-avirulent strains of Type III. Therefore, avirulence must rest upon some acquired property by means of -Thick the bacteria protect themselves against the resistance factors of the host. Further evidence in support of this is brought out by the fact that im- munization of rabbits vith the virulent strain results.ln the production of type &pecific antibodies. Active Immunity against Infection 71th iiabbit-Virulent Strain of Type III Pneumococcus. Rabbits Immunized with typo specific pneumo- cocci (Types I, IX, wd III) and sR" strains derived from the three fix- ed types nere tested for active immunity against infection tith the rab- bit virulent strain of Type III. The immunized animals used in the ex- periments mey be divided into 4 groups according'to the orgzxAsms used for antigenic and the resulting antibody re9)onsa. Group I. jngnunized aith Typo fI1 pneumoco& (Rabbit Avirulent StrainIll. Sam of only 3 out of 24 of thQBsnimjle acquired both type specific antibodies &d &&ios s$ecific antibodies - usually %csigxatod as anti-P '0 which are roactivo 71th the common pneumococcus nucloo-pro- tein and with any R form of pneumococcus. The sora of the remaining animals of this group acquired only anti-P antibodice. Grow II. Immunized with Type I or Type II pneuaococci o _, Sera possessed homologous type specific antibodies and also anti-P antibodies. Group III.. Iumnizod Tith nRn forins of pnoumococci. Scra po s- sessed only anti-P antibodies. 241 . All of these immunized rabbits were infectAd intravenously with 1 cc. of the rabbit virulent strain of Type III. Appropriate controls were used in each test. Usually .OOOl cc. or.001 cc. of 1 the virulent culture was fatal for normal rabbits and -01 cc. nas always fatal. Normal rabbits receiving 1 cc, of culture succumbed in 16 to 86 hours. In many instances the bacteremia was studied by means of blood cultures taken at frequent intervals in order to de- termine the duration and intensity of the blood infection. The strik- ing fact observed in these experiments is that in rabbits active im- munity against Type III pneumococci may be induced by immunization vith pnetPnococci of a heterologous type or with WI1 strains. Of 26 rabbits so immunized, 18 (69 per cent) survived; 6 other animals lived 6 to 12 days after the controls had died. Moreover, in each instance the bacteremia was much less severe, and at autopsy they all showed a flbrino-purulent pericarditis and pleurisy. These animals were consid- ered as showing definite evidence of increased resistance. If they are included with the rabbits which survived, 24 or 92 per cent of this series showed evidence'of active\ irsnunity against infection mith Type III organisme. Of the 12 rabbits immunized with Type III, and only 3 of which possessed demonstrable type specific agglutinins, 6 (50 per cent) survived and 4 others showed evidence of increased resistance, although ultimately succumbing to a localized pericarditia and'pleurisy. In this series 10 or 91 per cent showed evidence of active immunity. The results of these experiments shorn definitely that rabbits may be actively immunized against infection by avirulent Type III pncu- mococcus although type specific antibodies are not present in the cir- culating blood. Indeed, the animals possessing type specific agglutin- 242 :,g ins shoved no.higher degree of immunity than those aithout them. An explanation of the immunity of those rabbits immunized i7ith Type III but possessing no demonstrable type specific aggl&tinins in their ser- um might possibly rest upon the fact that the specific antibodies aere sessile. However, all possibility of a type specific reaction is ex- cluded in the animals immunized with Type I and II and R forms and sub- sequently infected nith a virulent strain of !l?ype III. In these animals the active immunityis as effective as in those immunized Pith Type III9 It may be noted that all the rabbits posseseed anti-protein antibodies. All previous ?rork sive protection. these experiments has shown that this type of antibody confers no pas- Eone of the sera of the immunized rabbits used in conferred passive protection on mice except those possessing type specific antibodies. Further evidence `of the ineffect- iveness of anti-P anti-bodies in active immunity is brought out by im- munization with solutions of pneumococci, This process results in the production of anti-P antibodies but not in active immunity. Prom the experiments described it seas highly improbable tbt demonstrable antibodies are of gignificance in the active inumrnitl~ of rabbits agains infection with Type III pneumococci~ It is possible, however, that an adqquate explanation may be found in the natural re- sistanca of rabbits to Type III pneumococcI+ The experiments summaria- ed in the first part of the report on ixmnunisation and infectivity show that rabbits possess the ability to doatroy cncapsulatod Type III pnou- mococci. Since rabbit sera contain no substances capable of damaging these organisms in vitro and since rabbits 1 loucocytcs aro incapable of phagocyting encapsulated pnoumococci, the natural resistance must re- side in some factor whose nature is, as yet, undotormined but -hose 243 presence must be admitted. It is not improbable that this factor (natural resistance) is amenable to stimulation and that nhen it is exalted there results increased natural resistance and that on this reaction active immunity depends. The conception is, therefore, ad- vanced that the increased resistance to infection with Type III pneu- mococcus which may be induced in rabbits by non specific means rests not on the production of some new factor - as antibodies - but upon the heightened activity of those factors already present in the ani- mal body which endow the rabbit with natural resistance to Type III pneumococci. Chemistry o.f Soluble Specific Substance. (Dr. Heidelberger). (In the autumn Dr. Heidelberger was asked to take charge of the Chem- ical Laboratory at the Mount Sinai Hospital Rhich invitation he accpt- ed and on' February first he left us to take up his new duties. VP to the time of his departure he me busily engaged in completing work on the probleme concerning the Soluble Specific Substance Thich wareal- ready under way and the nature of which m been discussed in previous reports. The followin/g notes by br, Heidelberger indicate the nes;l ob- servations which acre made before his departure.) In the case of the soluble specific substance of !l?ype I pneu- mococcu6 it waa found that on energetic oxidation aa much as one-third of the substance could be recovered as mucic acid, indicating that ga- lactose or galacturonic acid units meke up possibly as high as 50 per cent of the specific substance. A clue to the remaining portion la furnished by the presence also of an optically active acid oxidation product rrhich has not yet beast studied in detail. It is hoped to con- tinue in the Nount Sinai chemical laboratory studies on the Type I ma- * 244 ;A; 1 tarial already on hand and publish any results that may loo obtained as. a joint communication from the t%?o hospitals. The difficulty of obtaining pneumococcus material led to a further study of the solublo specific substance dorivod from gum arabic, in nhich the results given in a previous report mere confirmed and sug- ar acids of a type found in the Type III pneumococcus substance again isolated. Prom a sample of lemon pectin in the laboratory a fraction -7as also isolated ;Thich reacted nith Type II and Type III anti-pneu- mococcus sera, but as the amount recovered nas very small, and pectin is made from lemons contaminated nith mould, the origin of the speci- f ically reacting carbohydrate remains in doubt. Chemistry of the Soluble Soecific Substance of Pneumococcus and Friedlander Eaclllus. (Dr. Goebel) . A. Methods. 1. During the course of an investigation on the chemical nature of an aldobionic acid isolated as a hydrolytic product from the specific pdlyoaccharide produced by pneumococcus Type III during growth, it became necessary / to oxidize this aldehydic sugar as a step in the elucidation of its structure. For purposes of oxidation the met'nod devised by %I1 Fischer, end :.7hich has been generally used, failed. B nen method for converting any aldose unfailingly into the corresponding sugar acid `has been devised. The method embodies the principle expressed in the equation : 2RCHC+3Sa(OH)2+212=(3COC)2Ba+Z3a12&?120, and possesses the advantages of speed, of freedom from technical difficulties, and fin- ally of bringing about the conversion aldoee ---+ sugar acid practic- alfy quantitatively. 3y this method gluconic, maltobionic, lactobion- ic and nsaccharobionic" acids havo been prepared. During the course * 245 f.. ,! D .* 3 of this research it was observed that aldoses are not quantitatively ox- idized to the corresponding sugar acid if the initial concentration of hydroxyl ions is too great. A theorectical discussion which attempts t,' explain this phenomenum has been submitted for publication. 2. The general method hitherto of separating sugar acids from true sugars In the hydrolytic products of sugar acid-containing poly- saccharides makes use of the fact that sugar acids form an insoluble lead salt. Thus, when one treats a mixture of sugar acids and sugars nith basic lead acetate, a separation may be obtained, for the su=,ar acid lead salt is precipitated and the sugars remain In solution. Thi s separation is by no'means quantitative for a relatively large part of the sugar acid lead salt remains in solution mith the sugars. It is therefore difficult to investigate the nature of the sugar fraction since it is contaminated with sugar acid. It is furthermore extremely difficult to obtain any precise knowledge of the ratio of sugar acid to sugar since a quantitative separation cannot be effected. A new technique has therefore been devised which consists in boiling t'ne hyd- 0 rolytic product of acidic polysaccharides 31th calcium carbonate. All of the sugar acids are thus converted into the calcium salts and the . non-acidic true eugars remain unaltered. On evaporating the mixture and on the subsequent addition of methyl alcohol (in which the calcium salts of sugars are 1nsoluH.e) the sugars go into solution and may be quantitatively separated from the sugar acid c.~lcium compound. This method of separation is of importance for it permits the isolation of small quantities of sugar acids which xould other4se be lost and be- cause it permits a quantitative analysis of tho ratio of sugar to su- gar acid - an important point in the elucidation of the structure of the acidic pblysaccharide molecule. B. !#ve III Pneumococcus. It has formerly been reported that the aldobionic acid CllHlgOlO COOH forms by far the major portion of the hydrolytic products of the pneumococcus !!$pe III specific polysac- charide and evidence has been presented for the belief.-that this speci- fic substance may be grossly represented by the expression (aldobimic acid) x. Since this compound is ap$arently the fundamental building stone of this carbohydrate, a detailed study of the aldobionic acid has been made. By means of the new technic,described above it has been possible to secure this material from the hydrolytic products of the polysaccharide in a high state of purity and in greater yields than hitherto. The sugar is composed of one molecule of glucose combined through glucosidic linlcage to one mole- cule of sugar acid of the glu&onic type. of the numerous possible struc- tures of this compound the two most logical may k represented by the farmulat . COOH (:H~H) S Hi-O-CHOH- (CHOH)4- CH20H (;HOH)4 1 and H;-0-CH&HOH)~-OH0 1 0 CHO OH. I II Ey means of barium hypoiodite the aldobionic acid has been oxidiaed to the corresponding *Uaccharobionic" acid and a calcium salt, having the 9 ' proper analytical constants, has been isolated. This calcium salt gives a strong naphtho resorcinol test and yields 178 of furfural on distilla- tion with 1% HCL. If formula I were the correct formula for the origin- al aldobionic acid, the oxidation product, saccharobionic acid, would ob- viously contain one molecule of glucose and one molecule of saccharic acid in glucosidic linkage, and Q 19 there nould be no substance capable of yielding furfural on distilla- tion aith hydrochloric acid, rsithin the molecule. On the other hand, in formula II, the uranic acid would remain unimpaired on oxidation to the paccharobionic acid since the glucosidic linkage is through the aldehyde group of the uranic acid and the resultant saccharobionic acid nould be a compound of gluconic acid and a uranic acid. !ChiS compound rrould, of course, yield furfural. In viea of the experiment- al observations formula II must be assigned to the aldobionic acid. This compound is unique in the field of sugar chemistry. Although a great deal of indirect evidence has been gathered which points to the assumption that the sugar acid half of the aldobi- onic acid molecule is glucuronic acid, no direct evidence (i.e. no de- rivative of glucuronic acid itself) has ever been obtained. This is due to the fact that the aldobionic acid is an extremely stable com- pound and only very prolonged boiling vith a mineral acid nil1 effect hydrolysis. Under such conditions acids of the glucuronic type are decomposed, and it is impossible to prepare any derivative, It was, however, thought possible to hydrolyze this extreme- ly stable compound by means of dry methyl alcoholic hydrochlori c acid, If this tieatment is effective one should obtain as products of `hydro- lysis the methyl glucoeide of glucose and the methyl glucoside of glu- curonic methyl &h&r. The MO compounds could be readily and quanti- tatively separated by warming with barium hydroxide and pouring the mixture into alcohol, and the end products could be readily identified as their osazonea after hydrolysis of the glucosides. This investiga- tion is at present under nay, and promises to give definite direct evi dence as to the exact nature of the uron.ic acid f rat tion of the aldo- 248 bionic acid moloculo. c. Typo II Pncumococcua. It has beon found that the Soluble Specific Substance of Type II pneumococcus contains a sugar acid of the glucuronic type and that this specific soluble substance bears a cheni- cal analogy to that of Type III Pneumococcus. D. Type A Friedlander bacillus. The soluble specific substnncc from Type A Freidlander bacillus is apparently constituted from three distinct sugars. The first is a sugar acid either identical or isomeric with that obtained from Pneumococcue Type III soluble substance, the second is glucose which has been identified by oxidation to saccharic I acid and as its osozone. The third sugar is one nhich appears together *.:ith glucose in the true sugar fraction of the polysaccharide's hydro- lflic products, Thia sugar may be separated from glucose by fermenting away the latter with yeast. It contains no glucuronic acid derivative, nor is it a pentose or pentoae derivative. It yields a precipitate on boiling nith barium hydroxide, and it forms no typical sugar has not as yet been identified. The Friedlandor specific substance itself is a white It has an acid equivalent of 460, it light lOl? to the left, it gives 60s culated as glucose). It reacts,dth tion of 1:2,000,000. amorphous ponder, rotates the plane reducing sugar on osazonc. ThiS Typd A eolublo soluble in water. of prolonged hydrol;rsis (cal- its homalogous anti-serum in dilu- `E. Type B. Friedlander bacillus. By means of the technique described under A2 the calcium salt of a sugar acid either identical or isomeric nith that obtained from the Type III soluble specific sub- stance has been isolated from the hydrolytic products of the specific polysaccharide of Type B Friedlander bacillus. This sugar vaa not found in the provioua studies on this polysaccharide. A quantitative study of the hydrolytic products reveals the fact that the polysac- charide is built up from glucose and a sugar acid of tie glucuronic acid type in a ratiol-of 3 glucose molecules to 1 uranic acid moleculo, . The Pneumococcus Type III soluble specific substance has been shown to be coratituted from glucose and a uranic acid molecule in a ratio of 1:l. This spoclfic polysaccharide may roughly be ropresented by the formula (glucose-uranic acid)x; similarly the Type B Friedlander specific polyaaccharide may be repreaentod by the formula !. rglwose, glucoee glucose - uranic acidjx. Such a compound should theoretically have an acid equivalent of 682 a,nd a carbon and hydrogen content of 42.2 and 6.2 per cent, respectively. These figures are in close analogy nith those found experimentally. Thus a close analogp betveen the polysaccharides of these widely different organisms is to be seen. It is hoped that this . analogy may, in a similar manner, be extended to the polyaaccharides of the other organisms under study. I?. Type C.Friedlander bacillus. By methods essentially the asme as those used In the case of the soluble specific substance of Pneunococcua and Type A and B Friedlander bacillus a nitrogen free poly- saccharide with specific properties was isolated frun the Type C Fried- lander bacillus. It is a uhiQe emorphous poTder having' en acid equiva- lent of 480 and la soluble in water and alkali. It reacts v;ith lta homologous antiserum in dilutions of ~:2,000,000. It rotates the plane of polarized light 1000 to the right. It yields glucose and a sugar acid on hydrolysis. The isolation of this substance concludes a sya- tematic investigation of the specific polysaccharides of the three flx- ed types of Friedlander's bacillus. It is hoped ultimateQ to show that the specific polysacchar- ides both of the fixed types of Pneumococcus and lus bear a distinct and extremely close chemical in regard to the building stones, the sugars and of Freidlander bacil- analogy to one a7other sugar acids, nhich enter into the network of the complex polysaccharide molecule, but tb.?.t the definite chemical and ixnunological differences displayed by t'ne various polysaccharides depends first on the ratio of sugar to sugar acid which go to tion of chemical molecule itself. make up the complex molecule and linkage of sugar to sugar within second on the posi- the polysaccharide Obaervationa concerning the Reversion of Ill?" Cultures of , Pneumococcus to the aSa Type. Drs. Dawson and Avery . The question of the reversion of "Rs cultures of Pneumococci - avirulent, non-type specific, having no demonstrable capsule end elaborating no specific soluble substance, - to the "Sfl type, - virulent, type-specific, cap- sulated, and producing the specific soluble derivative, - has been the subject 6f further investigation. Animal Paaaage and Cultural Methods have been employed and the following results obtained: - ,A. Animal PassaPe. By repeated rapid passage in mice "an cultures of Type If and Type III (Laboratory strains D/39/R and M/3/R have been caused to revert to the "Sa type, having all the properties characteristic of the homologous type-virulent, type specific, capsu- lated and ItSa producing. As yet it has bea found impossible to revert the Typa I R strain which has been employed (Laboratory strain 1/192/R- Neufold). Efforts are being made to revert other Type 1 R strains by this method. B. Cultural Tiethods. Inasmuch as it is well knorm that one of the best methods to effect the transformation of "Ss cultures to the Ws type is to grow the "S" organisms in media containing Anti-S anti- bodies, (i.e. homologous immune serum) , it was thought possible that the reverse process might bs initiated by groming Ws orenisms in med- ia containing anti-pneumococcus-protein (anti R) antibodies. According- ly Qves I R, II R, and III R, cultures (1/192/R, D/39/R, M/3/R) were . grown in 105 anti-protein rabbit serum. After six to eight transfers the change was effected with Types II R and III R but so far attempts to revert the I R culture have failed. Other I R strains are now be- ing used. At this time`it was alloan in connection with other York that normal human sera contained anti-antibodies. It Ivas therefore suggest- ed that "Rs cultures might revert to "S" cultures if human sera acre employed. Similar results were obtainod. ln like manner it aas also found possible to effect the change by using immune horse-sora of het- erologous types, which contain in common the pnoumococcus-protein anti- bodies. Two questions then arose: - (1) Do all the cells of a given R strain possoss the ability to ro6rt in the cultures employed or only certain onos? (2) Is reversion due to the prosencc of the specific anti- protein antibody stimulus or is it simply depondent upon nutritioml factors a-plied by the use of serum-broth? Tho first quostion has bocn ans:`lorod by omploying pure line strains dcrivod from single-co11 cultures. TV0 singlo-cell cultures of II R (D/39/R) and four of III R (M/3/R) wcro usod and rovcrsion cf- fected in each instance. Roversion by cultural methods is alwaya ac- 3 d companiod by tho acquisition of virulence so that 0.00001 cc. of tho revcrtod culture kills ;7hitc mice in 24 hours, ;yheraas the origiiml sRtl strain fails to kill in doese of 1 cc. The second question is no7 being investigated by cm~lo;Vin~ serum-broth freo of anti-protein antibodies. At present tho :?ork is incomplotc but it is possible that the cha::ge may also occur, but 1~33 I readily. in t`no absence of anti-protein antibodies. Throughout the nork the quality of the media has becn fou;ld . to bo of paramount importance. Xedia rich in growth promoting factors has given results where routine media has failed. From the foregoing it Tould appear that the great majority, if not all, of avirulent IlR" cells have the ability, under proper en- . . vironmental conditions, to revert to virulent, type-specific capsulat- ed organisms. Antigen - Antibody Balance in Pneumonia. (Dr. DaTaon). The object of the work undertaken was, generally stated, to follow the anti- gen' - antibody balance in patients suffering from lobar pneranonia. Uore especially the problem related to the demonstration of the anti- protein antibody response during the course of the disease, inasmuch as the antigenlc properties of the intact cell and the antibody re- sponse of the hm body to the intact cell and its soluble deriva- tive had already been well established. Early during the work it became apparent that the antl-pro- tein antibody response did not ahog a wide range of variation. Pre- liminary vork on immune animal sera indicated that the TIread Reaction `;7ould be the most suitable method for its detection. Serial bleedings verc taken on a series of patients with lo- 253 04 , n I * .' bar pneumonia and the sera were tested by the thread reaction. Ob- sorvations made on a series of taelvc cases suggest that little, if any, change occurs in tho content of anti-protein antibody, Those cases in :7hich tho sora wore tested by agglutination nith heat-kill- ed sRs cultures also failed to shorr an increase in this anti-body. It is noi7 proposod to mnfirm this finding by precipitin tests us- ing autolyzod `RI' `cultures and nucloo-protein as precipitinogon. . Tho olaboration of tho soluble derivative of the pncumo- COCCUS cIas folloned in those cases and the appearance of an anti- body for this substance at or about the time of crisis domonstratcd. Tho sera of a series of normal adult individuals lcro also examined for tho prcscnce of the anti-protein antibow. With the methods employod*it uas indicated that this antibody X.L.S prcsont in every instance. While this problem nas being investigated some vork ~a6 al- so done on the revorsion of the "RI' type of pneumocaccus to the rrS1' type. Following Dr. Avery! s observation that "RIB cultures, Types II and III, aould revert to sSs cultures, under cortaia circumstances, when grown in santi-Rs sera, it was found that the same transformation would occur Then type-specific horao sorum of hetorologous type vas egployed. The possibility of the phenomenon occurring in normal human scra was suggested by the demonstration of the anti-protein antibody in these sera. Rxporimontal attempts to effect the change in this way accordingly were made and have, in certain instances, also mot ni th success. It is nov proposod to demonstrate Aothcr tho transformation is simply the result of nutritional factors or whether it is duo to a specific antibbdy stimulus. QiLture Xedia - ?:iss Eolt (Report by Dr. Avery). During the past trselve months the culture medium supplied to the Bacteriological Laboratories has been very unsatisfactory. Casual attempts to improve the media by changing the quality of the meat and the other ingredients, such as peptone, salts, etc. have been wholly without success. Sug- gestiona frcm the members of the Staff and the cooperation of the Media Department have likewise failed to improve the situation. It seemed de- sirable, therefore, to make a thorough study of the factors ;ihich deter- b m%ne the growth promoting qualities of culture media, since obviously principles more subtle than mere content of nitrogenous material, the proper salt Ialancs , an d the optional hydrogen ion concentration aro involved. To this end Xiss Halt, norking in the bacteriological lab- oratories of tho Hospital, is engaged at prosont in an attempt to ana+ . lyze thoso factors which are requisite for gronth, and which satisfy the nutritional neods of the bacteria. The ovidenco available, at pre- sent indicatea that the factors requdeite for the initiation and main- tenance of groMh are of the nature of substance which moot certain p&,Mological needs of cell and which are not nocossarily furnished by the food stuffs utilized in tho procoases of cell motabolism. These factors, although prosent to a greater or 10~s extent in the original materials, are lost in tho prooea$ of filtratioa,.or are actually de- stroyed by boiling and by final aterilizati'on of the medium at an al- kaline reaction. These factors may be reprosonbod'by the oxpressfon c+ G+xj ) in which sV represents an essential though slight concen- ..- tration of a readily utilieable carbohydrate, which furnishes a roady source of energy for the initiation of grorrth; and ;P+t<) represents a physiologically active system which has previously been shorm to be * 255 .k ,' : I ossontial for tho gro;-;th of organisms of the so-called hemophilic ????? o In moat, tho C factor is present as native muscle and blood sugar, but naturally varies in concentration dopending upon the nutri- tional state ?? the animal before slaughter, and upon the amount of glycolysis and oxidation which t&s place during subsequont handling of tho meat. Consequently the IT" factor is a variable Fhich can be determined for any given lot of medium by fermentation tests and chom- t ical mo thods. This variation can bo overcome by the addition of vary small amounts of dextrose (0.02 - O.OS$) - a concentration found to bo optional for initiation of grorrth but insufficient to produce tho detrimental acidity shlch always accompanies the metabolism of larger amounts of sugar. The "V" factor" is a term used to designate these substances which are vitamineliko in nature and T,Thich correspond to sbiosfl. This factor, alnaya present in frosh plant and animal tlssuos, functions in a romarkablo way in stimulating bacterial grmth. The proaent studios indicate, honover, that this gronth-accessory substance is completely dostroyod nhen heated at the pII and for tine period rhich corresponds in alkalinity and time to the process to Thich media arc subjoctod by present mothods of preparation. By reason;rof the knorm role of V factor in the groath of B. influenza0 this organism serves as a deli- cate biological `indicator of the presence or absence of this factor in any given lot of media. BTJ methods previously norked out extracts of plant cells (tomatoes and yeast) F;hich are rich in V content, are now av,ilable and ufiilizable In this experimental work-. Used as enrich- ment fluid these extracts have an extraordinary gronth-promoting in- fluence when added to an otherwise unfavorable medium. The Yt" factor, which acts as a bio-catalyst, is of import- ance in the oxidation-reduction processes of the cell. It has the functions of a peroxidase, and to it has been attributed a role in oxygen transport. As is well known, its presence is essential in the cultivation of hmmhilfc bacteria, and, while not requisite for gro7;lth of pneumoaoccus, it apparently exerts a definite action in conserving the viability of cell. ft is known to prevent the accumulation of per- oxide which always occur8 in culture fluid nhen pneumococci are gronn in the presence of oxygen. This substance, related to the active iron salts present in plant and animal tissue, although heat stabile, is lost by filtration, being carried out of solution by adsorption to the heat and coagulable proteins during the preparation of media. Attempts are non being made to ?zonserve, as far as possible, these factors as they occur naturally in meat, and to make up their deficiency by the addition of suitable substitutes to the final media. Methods m also being tested which nil1 permit of the application of these principles to the mass production of media.