a Report of Dr. Avery (!?ith Drs, Stillman, Tillott, Jul.iancllc, Gocbcl, Dubos, Ihsson an.d Francis). Studies on Pncumococcus Infection n.nd Immunity. I. Chemo-icmnunologicnl Studies: 1. Nature of the type-specific antigen. 2. Synthesis of carbohydrate-protein antigens. II. Rcvarsion of ~1W --4 IIS forms of Pncumococcus and the Inter- convertibility of the specific types. III. Cutaneous reactions in pncumonin natients to the protrin and specific c*jrbohydrato of Pneumococcus. IV. Cutaneous infection of normal and immune rabbits with IIW and IIS11 pneumococci. v. Cutoncous vaccination of rabbits ni th Pncumococcus: the character of the antibody response, tho resistnnct? to infection ; and the hyperscnsitivencss induced in animals. VI. Rcversio;l of l!R" ----j BASIL forms of Friedl&.ndor bacilli and the intcrconvertibility of the specific types. VII. Nature and duration of tha immunity produced in rabbits by the inhalation of pneumococci, VIII. Significance of oxidation-reduction processes in bnc tarie.1 growth. IX. Publications. In the nork on Pneumococcus infection and immunity cprried on in the bnctcriologica.1 laboratories of the hospital during the pr)ast year in conjunction with the: clinical study of pneumonia in the wards, the studies of special interest qnd those sufficiently supported by cxporim~ntnl data to justif, v .thcir inclusion in the annual report to the I?o.?rd of Scic:ltific Dirxtors -.rL, the chomo-immunologic?,1 invcsti- f gation concorning tht' specific nntigrlns of Pneumococcus, the results of t7hich `xc olrz-xiy yicldin,- more prccisc !oioYlcdg(I of tho chcmic-1 nnd biologic71 propcrtios, rl:ld of th.: cnuscs .l;id the prcw:;tion of the dissociation 2nd co;;sr:qucnt inac tivntion of the sc complex nntigcns, bcforo -SC nftcr injcctio:l i:lto th;: qnim-1 Fody; the successful ?;ld origin-l synthesis of sugy.r-prot. ins bv the. couplis: of ;1 c.?rbohydr-tc r?diclc in glucosidic union ::ith serum globulin, undl cf fc:c t nnd biological specificity; which prob.qbly ;`oux. %hcir o;-i,.in in the study of the activ, priiXi?l;s of csrt?i:l n.?tur.ql dru,-s, h?vc bcconL. so nbsorbi:;c -lid so cnbr.-,ciag th't they hqv,: attracted r#iZi held the ir,tcrest of tl-c chcmis t, the phirxmn- cologist xld im~a~ologist~ T.lik. Tnc question of protc,in s2ccificity in p.p.rticulcr, .?.nd ths possibility of chnngang spccificitg by altering the pro kin molecule throu& chemic7.1 mc.xls, have ongngcd the minds of ma3y invcstigo.tors. On tht: other hsnd, the. r8lc which cnrbohydratcs ?loy in th,. phznomcn j> of immunity hns on'Lp recently been disclosed, dcspi tc th<- fact that the presence of thcsc substances in bqcteri-? ad yeast hns lo:l& been ?..own; Some ye?.rs a.go , Dochcz and Avery observed th?t filtrates of pceuaococcus cultures contained a stable sabstaxc which reacted specif ic?lly ::i th .Tn t ipxanococcus servm o f th< homolo&ous type. This observetion led to the isolzti-1: nxd idcxtific~,tioc of spccific?.lly rex ting substaces from the three specific types of Pi-mmo~o~c~s vd from the threl> t$Tc:s of T'riedl!kdtirr's bxillus. Sixc then other 208 196 investigators have isolated similar substances from various other species of bacteria. These type-specific substances fall into the class of carbohy- drates; they are unusual carbohydrates in that each contains a sugar acid as an integral part of its complex molecule, Immunologically they belong to that important group of specifically reactive but non-anti- gcnic substances which Landsteiner has named heptens. r)no of the striking characteristics of these bat terial carbo- hydrates is their failure to produce antibodies when injected into the animal organism, though in the state in which they occur in the bac- terial cell they are not only type-specific but also effectively anti- genie as well. When the nneumococcus cell is permitted to autolyse under sterile conditions it has been found that the autolytic process is accompanied by an increase in amino and non-coagulable nitrogen and in ether-soluble fatty acids. This change is due to the action of liberated intracellular enzymes, protease and lipase, upon the native substances of the cell. When extracts of pneumococci containing the active intracellular enzymes are added to heat-killed pneumococci, lysis of the cells occurs and there is an increase in the non-coagu- lable and amino nitrogen comparable to the changes accompanying spontaneous autolysls. Furthermore, when extracts containing the active intracellular enzymes are added to emulsions of the' alcohol- .. soluble lipoida extracted from pneumococci, an increase In the ether- soluble fatty acids occurs. men living pnewnococci are dissolved by means of sodium desozqcholate and such solutions are incubated at 37Ocm 9 an excess of the bile salt has been found to inhibit the action of the pneumococcus pro tease, but not the action of the lipase. Such solutions, however, no longer yield type-specific antibodies when injected into animals, It seems, therefore, that the specific immu- nizing antigen of pneumococcus is impaired by the action of its own lipase. In order that the bacterial polysaccharide may be effective as an antigen it is believed that it must be combined with another cellular constituent, possibly a protein, in ester linkage to form an easily dissociable, complex antigen. The products of hydrolysis of the specific carbohydrates of Pneumococcus and Friedlander's bacillus have been studied in detail. All except the Type I Pneumococcus carbohydrate yield glucose on hydrolysis. The complex sugars of Pneumococcus Type III and Fried- lander baCil1U.B !Fy?e A yield on hydrolysis isomeric aldobionic acids (glucose-glucuronic acid) in addition to glucose. Evidence has also been secured which indicates that the specific polysaccharides of Type II Pneumococcus and of Type B and C Friedlander bacillus also contain other aldobionic acids within their molecules, The selective specificity of encapsulated organisms, such as Pneumococcus and FriedUnder bacillus, seems to depend primarily on the hapten part of the hypothetical complex antigen. In all of the haptens studied thus far, the invariable presence of isomeric aldo- bionic acids seems to indicate that particularly the acid group (carbowl group) and its stereo-chemical relationship to other groups am . 197 the intact polysaccharide molecule, sarily be different, has a profound antibody response. I of which in each instance must neces- influence in orienting specific In order to understand more fully the r61e of carbohydrates in immune phenomena we turned td the group of mucoproteins, the naturally occurring conjoined carbohydrate-proteins, but without avpil, for we did not find the c.srbohydrates either of ovomucoid or of tendomucoid 230 198 to behave as haptins. Nor did mixtures of purified specific soluble substance and protein yield, on immunization, antibodies reactive vith the sped if ic carbohydrates. Thercforc, for the purpose of studying the role which simple sugars, :nd in particular sugar acids, of different spatial configuration, such as those trShich are found in specific poly- saccharides, might play in altering the snccificity of proteins, it was thought that it might bz possible to combine diffcrant sugars and their acid dcrivativcs with a given prot, in and to observe any specific dif- fzrcnccs in antigonic properties that might result. The problom thus bccomcs tno-fold, first to synthcsizc the desired sugar derivative, and second to devise some means of combining the sugar with the prokin. Some years ago, Psuly described a test for tyrosin and hiatidin in proteins by adding diazobcnzene sulphonic acid to an alkaline solution of protein. The diazobenzcne sulphonic acid combines with tht. protein, and on acidification the complex is pr+ cipitated. This fundamental principla was made use of in combining glucose and galactose to sew globulin. The second problem, that of preparing a sugar derivative adaptable to this principle, was solved by the following synthesis: $X,OH CH - (CHOH), - CJH L- 0 I `OH (CH3Co)20) CH,OCOCH CHOCOCH,) s - d 1 b COCH, Glucose. Pentaace tyl glucose, H H Br CH,OCOCH, - "I" - (CIIOCOCH, ) s - /- -+- _ _ _, > (-)A IX ^ _ J_gl OwLNO Te tra-acetyl-bromo-glucose. * Silver parani trophenolate. ~--j CH,OCOCH, - CH - (CHOCOCki,), - C:" m (OH), I o- OW*NO2 -> 1 Tetra-acetyl-paranitrophenol P -gl.ucoside. CH,OH - CH - (CHOH) a - ,,s `Q Parani trophenol - ,,6-glucoside. Pnraeminophenol - ,p- gl ucoside. The end product of this series of reactions may then be converted into the corresponding diazonium derivative and coupled with protein in the presence of dilute alkali: cE:,oH -ch' - ( CHOH), - Cl' ! 1 "OC,H,NH,HCl HNO, .o - I I,@ H CH,OH - CH - ( CHOH), - c' i 0-J `0 - CsH, - N = N - Cl + Protein > cH,OH - CH - (CHOII), c" "0 - CsH, - N= N- Pro te in Thus one may couple precticolly any sugar (aldooe) or its derivative with ,qny protein. Ve hnve thus fqr synthesized, in addition to intermediate glucosides, pnroaminophenol fl -glucoside, pnd par~e.aminophenol/?-g&ctoside. These hexosides have been coupled to pure seruin globulin under conditions which do not alter, by denaturation, the protein molecule. Two different synthetic sugar proteins h?.ve thus been prepared. They exhibit different chemical properties, end althou& the pro tcin substrate is in each in- stance the same, they appear to be different compounds. Animals are being immunized rith these synthetic sug7.r protein antigens. Attempts are elco being rnde to prepnre the glucosides of aldobionic qcids, the sugar acids found in the specific soluble substances of Fneumococcus, gnd to combine them with the be.cterial protein end to study the antigcnic properties of these complex substances. : I J `9 II. Studies on the Intcrc,onvcrtibility of 11Rll and IISII Forms of Pneumo- COCCUS ?nd the Interconvertibility of Fnoumococco.1 Types. (Dr. DaTson). In the January (1928) numbor of the Journ. of Hygiene thcro ngpe:)rcd an Trticlc by F. Griffith in orhich ht: clsimcd to hqve dcmonstrntcd the fol- loY/iag points: (1) That IIRil nncumococci could bc rGvcrt,tid into the homologous RS" tyncs by tht subcut-ancous inj;:ction in mice of lyrgc nmounts of live organisms; (2) th7t IIR" pneumococci could bc rcvertcd into the homologous ItSI' type b!r thi subcut.lncous injection in mice of smnll amounts of live IIR" culturcs togcthcr xxith the hc?t-killed deposits of llrgc .~mounts of the corresqonding `1~3'1 cultures; 2nd (?) that 11Rll forms could bc trnns- formod into hctarologous Il'Sll types Q: the subcut?nLous injection in mice I of smqll *Imounts of live 1~91' orgaisms togcthor vlith the hc?t-killed deponits of lqrgo "mounts of hcterologous rlS11 cultures. In vice of tht significance of these findings, ex?tzrimcnts ricrc ?t once undcrtnlan to verify Griffith's results. (1) The rcvcrsion of IIR'l forms into ths homologous IIS" type by the subcutnneous injection in mice of la:.rge Fmounts of living II3 cultures. It w?s found th,?t this was n most cffcctive method of cnusing IIlP forms to revert to the IIS' type, Dif ferat IlRll s tmi;le vqritid in ths %nounf of culture requirr:d ;lnd one strrlin (1/192/R) uniformly f:-iilcd to revert. The results obtained were cntircly in -,ccord mi th those previously obtained by other methods. In 9.11 instnnces in which rcvcrsion occurred the ItR" forms invnrisblp reverted to the s*tme specific type from which they were originally dorivcd. (2) The reversion of IIRII forms into th:. homologous I'S" types by the subcutaneous injection in mice of sm:all nmounts of live I'Rll cultures togethor ?uit'n th.:, heqt-killed dc?oeits of lagc Dmounts of the corres- ponding I~SI' cul turc. This method S,TS found to be more cffcctivc thnn .:lny 243 201 other previously employed in producing the R 4 S cha.nge. success 172s attained Mth one culture (1/192/R) which hnd resisted 711 previous efforts to ef feet the trnnsformstion. This p4rticuler culture he.d previously been pzsced intrsperi tonwlly through 105 consccut ivt: mice Ind still m?intyinc? all of its IIRll chnrac teristics. ( `Ihe qucstioncf the viability of .?ny orgnnisms in the heq.t-killed deposits is c?refully considered in the next pnrt of the report.) (3) The tr9nsformntion of I~RI' forms into hotcrologous llSI1 typos by the subcut.\neouc injectir,n in mice of small nmounts (0.25 cc.) of living IIR" organisms togcthcr aith the heat-killed deposits of large #amounts (5Q-100 cc.) of IlSfl vnccines of hcterologous types. Since the question of the viability of the organism in the vaccines is ru1 all important consideration, the procedures em?loycd in the prcpwntion of the wccincs nnd the controls used in testing for sterility . are given in dctnil. The vqccines nerc heat-killed in scaled .?mpules totally immersed in water. Fifteen minutes hen.ting q t 60' has inv.wia.bly sufficed to `kill all pneumococci but longer periods qnd higher tcmper?turcs have also been used. In Vitxb &ntrols: (1) All vaccines have been cultured on blood -- agar plntcs and in blood broth. (2) Cultures have been made aerobic- ally snd nnwrobicslly in blood broth qnd blood extract dextrose broth. (3) Vaccines hnve been cultured serirtlly in 10 per cent serum broth for 20 transfers. In Pivo Controls: (1) An eqwl number of control mice hwc -- been inoculated with the vqccinc .zlont? and l?tcr sacrificed md cultured at varying intervals. (2) Control rtnimals hnve been inoculated nith the vaccine together with other organisms, such as ,B influenzae and streptococcus. (3) ha'i co intoxicated with alcohol hnve been injected rrith hrge xnounts of v.7.ccinc. (4) In the majority of experiments there sur- vivcd mnny test -nimals in which the transform? tion did not occurm In the course of the zork additional, qnd even more convincing, controls were ?ffordcd by certain observations which vi11 be 13tcr reported. In n lirgt? series of cxpcrimcnts the folloming points have been es tablishod: - a) The subcutsneous injcc tion into mice of r\ lqrgo Lamount (50 - 100 cc.) of n vxccinc of ,7n "So culture, (heat-killed at tcmperaturcs varying from 60oC up to 7nd including 80oC for periods of from 30 minutes to 2 hours), togc thcr ai th qn IIR~~ culture derived from z hetcrologous type, results ill the formation of virulent 1ISlI pneumococci of the S,XTK type as thqt of the v,eccine omploved. (b) If the vxcine is heated to 3 higher tcmpcr?turc thgn 80oC., the rlR" culture frequently reverts to the sune type rrs th.?t from which it YS originllly derived, . not to the type of the vnccina. (c) Any ~~RII orgonism rcgrzrdless of derivation potentiqlly h-s the capxity of claborrlting the specific sol- uble substance of qny specific type of Pncumococcus, 2nd by the -3rocedure described can be transformed into 1'S'! organisms of n hetcrologous type. (d) All attempts to effect heterologous trszxformntion Ibin vitro'1 -- have failed. (a) `Phe follosing conditions arc necessvy:- 2 Large amounts of MS11 vnccino rrrc nccessrzry (50-100 cc.). 2 The vncclne must not be allowed to autolyse before being hest- killed. c The We culture must be acll ndegr>dod" else it reverts to its own type. III. Cutaneous Reactions in Pneumonia P.aticnts to Cell Constituents of the Pneumococcus. (Dr. Tillett ?nd Dr. Francis). Patients acutely ill nith and convalescent from pnoumonin are being tested by intra- cutaneous inoculntion to determine their reactivity to the protein :rnd cwbohydrntc constituents of Pneumococcus. The cell fractions used for injection are (1) soluble snecific substnnces (polysnccharides) derived from T&vpcs I, II nnd III, in purified state, qnd (2) the soc-rllcd nucleo- protein fraction, which exhibits the protein-specificity of the species. The polysnccharides and the protein are injected in mounts of O..Ol mgn. ., . On admission to the hospital e,%h pneumoniq patient receives 5 intradormnl tests: 1) Type I 1~31~ substance;; 2) `qrpo 21 IIS" substance; 3) Type III 11 SII subs tanc c ; 4) l+cl.eo-protein fraction;. 5) Salt solution control. At the swne time blood is i?ithdrann in order to determine the presence or rlbocnce in the sera of. *jntibodies reactive with the substances injected, These tests are repented every few days during the pnticntls stay in the hospital., Following injection, the patients me observed both for nn immedi-ite rewztion. and for a' delayed response which appears after 24 to 48 hours., Tno independent `qnd die tinct forms .of reactions are no ted, The first of these is prqduaed by the protein fraction. Shortly after the in- Section of the material a slight reddening occurs, wd occasionally a slight edema, This subsides rapidly, Beginning at about the eighth hour, erythema, edema and tenderness, In varying degrees, appear at the site of inoculation wrd then increase up to 20-24 hours when the maximum is reached. fn general, the appearance of this form of skin reaction to protein substance is simil!ir to the tuberculin reacticn, and its oc- currence and course are briefly as follow)- During the acute phase of the illness there is no response to the protein material, but follo,ning the return of the temperature to norms1 the reaction becomes positive. As conwlesccnce proceeds the intensity of reaction becomes more marked, and the nppe,qance more striking. Sufficient time has not elapsed since h 216 204 the beginning of thcs c experiments to dotorminc ho:i: long pnticnts con- tinue to exhibit this sltcred reactivity to Pneumoeoccus protein. The reaction of normn.1 individuals is .qlso being studied, but ns yet suf- f icicnt dnt,7. have nc t been accumulated to aarrant any conclusions. The second form of rc%tion hqs beon obtained ith the type- specific carbohydrates. During tho phase of acute illness thsre is no rent tion. However, coincident :7ith recovery a positive re?.Ction m.?y develop. The character of this type-specific response offers a striking contrast to th-lt folloaing protein injection. Fifteen to 30 minutes folloaing the intradcrmnl inoculation, thcrc -tppears n.t the site of injection a wheel-lik0 szelling with intense :7hite edemas Surrounding the wheel n zone of crythemn nppemrs which becomes incrensingly 1, rger -tnd more intense. The height of rc?ction occurs bet\:ecn 30 2nd 60 minutes, after which a grndwl regression t?kes place leaving 3. firm pale edematous are2 which may require 24 hours or longer to subside. At the site of injection of the carbohydrates of the heterclogous types uld of the salt solution, which serve as controls, no reaction appears* This : form of typical Itwhe?l mnd erythemx 11 renction is of' the immediate nnaphy- 1% tic tyoe rind is in sharp contrw t to the del.Ged protein response, which requires 24 hours for m.aximum ieaction and is allergic in character. Positive results have been uniformly type-specific, that is, the reaction has occurred only with the specific carbohydrate homologous to the type of pneumococcus causing the disease. To date it has been 8 trikingly no ted in cases of Type I pneumonia treated with Type I antipneumococcus horse serum. In these treated cases, large amounts of Type I antibodies are introduced intravenously. In serum-treated patients the occurrence of a positive reaction to the Type I I'S11 substnnce appears to be rel-Fted to recovery. It seems possible that a positive test might indicate xn excess I i 2J.r 20!5 of nntibodics in the blood :*nd thcrcby servo .?s ~11 index of the Tmoun t of serum thcr?.py rcquircd. These rcqctions .Trc of further intcrcst ?.s -Ln example cf the rc%tivi ty of the s~lublc specific substances in hum ~6. It h7s *also been found that, in p?ticnts not trcq,tcd nith immuno sorum coincident *.?ith spontlneous recovery, 3 positive rcnction may folio!-. the injection of the homolo@us !lSll substance, In addition to observations on the skin rcTctic.ns, scra obtsi;:cd at the same time hive been tested for the presence or absence of nnti- bodies reactive ri th the test mqterfn.1 - i.e. pncumococcus nucleo-protein and spccif ic .polysecchnrides. From the data so fDr obtnined, no significord; difference has been noted in the sor2 -ith refercnca tc content of pro tcin precipitins before or q.ftcr crisis. In ot?aer nerds, there scemsto be no relrltion betl7cen the unount of nntiprotcin-precipitin in the sera nnd reactivity of skin to this nubstqncc. HoTever, there does qppear to be a dof inite correlation betceen the roriction to I~Sll substances ?nd the pre- sence of specific antibodies. Althou& some cases possessing CirCUhting anti-S antibodies do not react, no case has as yet been found Thich gives n a positive rcnction in the absence of anti-S antibodies. It is too e.7.rly in the work to interpret the significvlco of these results. Hoaever, it is a striking fact that t two dis tint t typos of reaction to tno chemically different constituents of Pncumococcus, m.ay be ulicited by skin tests in pneumonia patients; the one, produced by the carbohydrate is type- specific, immediate and nnaghylac tic in character; the other, produced by I the protein, is independent of type-specificity and is ,qnnlogous to the delayed reaction characteristic of allergic phenomcnd. IV. Natural and Acquired Resistsnccs to WI pnd 1~S~~ Forms of Pnoumococcus, (Dr. Tillett). The experimental studies, reported in previous publicaticjns, on natural and acquired resistance of rabbits to pneumococcus inf. cticns have indicqted thrrt immunity m.?y exist in the ob- scncc of dcnons trnblc type-spccif ic antibodies. Ihe fact was established that normxl rabbits are c2pnblo of nithstnnding large doses of Type III pneumococci al though sntibodies rent tive P ith this orgaism could sot bti demonstrated in the serum of thtl norm.q.lly resistant minimal. It hss also bee!: shocn th? t rabbits immunized r!i th degraded ?virulent , non- type-spccif ic pneumococc i - so-called IIW strains - wquirc A considernble degree of resis trr.nco against subsequent infection *=i th virulent pncumococci of nny of the three specific tyocs. Active resis tnnce ;vas demonstrable under these conditions in spite of the fnct that the sera of the immunized rabbits contained no detectable type-specific antibodies cap,Tble of agglu- tinating type-specific organisms, or of precipitating the soluble specific substances, or of conferring passive protection on mice against pneumo- coccus infection. Furthermore, it has been shown that, althou& passive protection of mice is ineffectual, the phole blood or serum of rabbits immunized y.Tith I!RIl organisms is capable of conferring immunity on normal rabbits against any of the three specific types of Pneumococcus, The results of the ex- periments on active and passive immunity in rabbits immunized with IIRIl cells reveal characteristics which differ from typo-specific immunity; it seems possible that the underlying mechanism is dependent upon factors other than those participating in type-specific pnoumococcal reactions. Moreover, the so-called HRll immunity - stimulated in rabbits by HR" cells and effective against virulent I'S11 pneumococci - has been shown to possess certain characteristics similar to the natural resistance which normal rabbits possess against most strains of Type .X11. This form of I~Rl~ immunity appears to represent on enhancement of these fat tors which are concerned in the natural resistance of normal rabbits. During the p?st ycq,r cxperimcnts h*vc been cwricd on in en rlttcmpt to dctcrnix mori. xcurn.Lly thi f.?ctors IiiVolved in type-specific im- nuni ty, in ItW irxui~it~-, qnd in n? turxl resistxxx. For this.pUrpOSC, rabbits imuaizcd :-I tt tipc-specific org~:~isms, o thcrs immunized aith `~Rtl orgnnis;Ls, togc ther :?i th norm-61 control rlnimals hn.vo bcon infected by iatrndt;r,:?l inoculation. The chnracter of the lot-,l lcsio~s -r.nd the course of the rcsultnxt blood invnsion hq.vc beon simultnncously folloxod. Expcrixlczts of his n-turc hrrvo sfforded n muons of comp.xing certain sicilnri tics -nci difforcnces .*:hich hqvc barn found to exist in these forms of n-Lturq1 wd acquired immunity. The qnin,!ls used in thcso experiments c.~y be divided into three groups:- & Normal rabbits, 2. Rabbits immunized -xi th llR1l pncumococci. 2. Rabbits iixwnizod cl th type-specific pneumococci. , The pncumococci employed for infection hwc been as follons:- 1) ItRII 8 tr=lins. Degraded. nvirulent, xnd non- type-specific forms of pneumococci. 2) I~Sl~ streins. TTO strains of type 111 pncumococcus hnve been used: one, avirulent for rabbits (dcsigna ted SA); the o thcr , virulent for rabbits (designsted SV). 3) 11s" Pncumococci. Type I ,?nd II, virulent for rabbits. , The results of these oxperimcnts msy be summwieod as follotis:- A. Intradermal infection of normal rabbits:- 1. curls Pneumococci. The skin losion which develops is smnll, discrete, firm, slightlu red, and nodular; it appcvs within the first 24 hours .o,nd disa.ppcars in three to four daya. Repented bloo&-cultures remain sterile. Phagocytosis of "`RI' organisms by polymorpho&koar leucocytos is reodily dezonstrablo.. The infected animals invariably recover. 2. t's" Pneumococci. Type II?, , The skin lesion devclopcd is large, p Strain avirulent for rabbits. edematoue , reddish-purple, and nithin 43 hours may extend to the midline ventrylly; marked ecchymosis Commonly occurs, After a fer; dRys central necrosis and ulceration take place; several weeks is required for complete healing. Blood cultures tnkzn at frequent intervals reveal the presence of organisms varying in number from time to time, and persisting for several days but titimately disirppear ing. Phagocytosis of these IlSll organisms by polymorphantilear lcucocytos is not demons trablo. Recovery of the animal ensues. 2. Strain virulent for rabbits. The skin lesion produced by these organisms is identical with th3.t caused by the avirulent stro.in. It is large, edematous, ecchymotic and extends ventrally tc the midline. Blood cul turcs taken at frequent intervals, shon that organisms are present in the circulation a few hours after inoculation; the sopticaomia rapidly increases in severity until death ensues from ;u1 overwhelming blood infection. Xo phagocytosis is demonstrable. The infection is invariably fatal. 3. WI Pneumococci. Type I and II. The skin lesion and blood infection resulting from the inJection of the virulent strains is identical with that described for the virulent Type III, The animals die in 36.48 hours, 2 Intradermal inoculation..& rabbits immunized with `IRfi Pneumococci, 1. ~~S~~ -Pneumococci. Type III; Virulent for normal rabbitst !%e skin lesion produced is a large, edematous, reddish-purple inflammation which spreads ventrally to the midline; as in normrll rabbits inoculated with I'S" organisms ulceration and necrosis set in and several weeks are required for complete healing. Frequent blood cultures demonstrate the presence of organisms in varying.numbers for several days. They finally disappear, however, and recovery of the animal ensues, The ch;trac ter of t 1 this lesion, as llrell as the bacteremia, are similar to that occurring in normal rabbits injected with .Type III ItSI' strain, avirulent for the sllecies. 2. IIS" Pneumococci. Type I and II; Virulent for normal rabbits:- The skin lesion and course of blood infection are similnr to that described for virulent Type III. c* -, Intradermal Infection of Rabbits Immunized Mth Type-specific Pneumococci. I. IISt PneumOCOCCi;- In this group of animals the organisms inoculnted were always of a type homologous to the IIS" strain used for immunization. The skin lesion produced is small, discrete, firm, slightly red, and nodular. It appears within the first 24 hours, and disapqears in three' to four dsyss This lesion is similar to that produced in normal rabbits by the injection of 1lRll pneumococci. Blood cultures always remain sterile. Phagocytosis of the sensitized type-specific organisms by polymorphQnu&ar leucocy tes is readily demonstrable. In on analysis of'these experiments the reaction8 to skin in- fection a8 observed In normel and brnnunlzed rabbits offer certain similarities and contrasts. There appears to be, in one group of animals, a skin infection followed by recovery which may bo cited as fOllOW8: 1. Reaction and recovery of normal rabbits from infection `with IIR'l pneumococci. 2. Reaction end recovery of type-specifically immunized rabbits from infection vlith homologous type-specific organisms. The chara.ctcr of the skin lesion in th;3eo two instances is the SEliie. 8. smaJ,l, insignif icpnt nodul:: develops which disappears in a few days. Organisms are never present in the blood stream, and prompt 222 210 phagocytosis is dcmonstrjblc in each instance. Recover:- dipends pri- marily upon phagoci tic activity. The typo-specific organisms being scnsitizcd by type-specific antibodies are quickly disposed of. The IIRIl pneumococci possessing no capsule can be ingested as such. Naturnl and acquired resis tanco to pneumococcus infection, as exempli- fled in these two instances, indicates a close similar1 ty in the mechanism of recovery. In another group of animals, the reaction and recovery differ from the first group and may be cited as follows:- 1. Reaction and recovery of normal rabbits from infection pith II!9 pncumococci, Type III, avirulent for the species. 2. Rcnction find recovery of R 11 11 irrununized rabbits from infection with virulent type-specific organisms I, II or III. The character of the skin lesion in these two instances is simil7r. A lnrgc, edematous, ecchymo tic, purplish, spreading inf lam- matory reaction follow8 the intradermnl inoculation of the organisms. Ulceration ,and slough ensue end several weeks are requiked for complete disappearance of the reaction. Organisms may be present in the blood stream in varying numbers for several days before their final disappear- ance. In neither instance can phagocytosis by leucocytes be demonstrated, Natural and acquired resistance to pheumococcus infection, a8 exompli- fied in these two instances, also indicate8 n close similarity in the mechanism of recovery. The degree of the inflammation, however, and the duration of the infection differ considerably from and transient infection described in the first group. it seems not unlikely that the underlying mechanism 18 each instance. the mild lesion Consequently, different in In the first grow, prompt pha.goc:rtic activity appears to be the I * , I I initinl rc:iction involved \*!hilc in the Second group the factor2 ?rc as yet not understood. In so far as the work ha8 progrcsscd it seems unlikely th3t the primary reactions nre due solely cithcr to nntibodics, as ordinarily understood, or to phagocytosis. However, thz rcccnt work` of Griffith has Shown that IlRll pneumococci possess the c?pncity to char@ into 'any of the specific typos. Consequently, it rnny not be impossible that IlR" cells possess some increment capable of inciting a response specific for each type of pneumococcus lnd that this is responsible for the broad immunity resulting from immunization with I~R" cells. !Fhc work h.-+s not a8 yet progrcsscd sufficiently to Justify any con- clusions as to the n.?tur: of the so-cqlled non-typo-specific resistance. V. Intracutaneous Vaccination of Ribbits with Pneumococcus, Lb Antibody Response. (Dr. Julianelle). Suspensions of heat-killed pneumococci were injected into the skins of rabbit8 at intervals of seven days during a period of 10 to 14 weeks, in dose8 such that the total amount of bacterial 8UbStenCe injected was at least equivalent to that employed in the routine intravenous immunieation of rabbits. The antigens studied were Types, I and 11X and a capsule-free SR~~ strain derived from a Type II organism. The sera of the rabbit8 immunized in this way were tested for the presence of agglutinine, precipitins and protective antibodies. The sera of rabbits immunized intradermnlly with Type I failed in more than 85 per cent of the animals to show any type-specific anti- bodies. They failed to agglutinate the corresponding type of pneumo- coccus) and they failed to precipitate the homologous specific poly- saccharide. In rnre instances, pnssive prot:ction was conferred upon white mice against infection bv Type I, but in such cases the protec- tion afforded was not great, In the remaining rabbits, although tvpc- SpoCifiC n;:tibodie8 were present, the titre as mensurcd by agglutination 224 212 was low ranging from 1:l to 1:20. On th< other hand, none of the rabbits immunlzcd with bpo III by w.ted rabbits, on the other hand, the protein substances induce an inflammatory ranction at the site of injection which my persist for threl: to five deys. In testing eye sensitivity in the skin-vxcinnted rabbits, the cornea was anaesthetized ant lightly sc.:>rified, and one drop of nucleo- protein or pu=pura-producing extract wm then instilled into the con- junc tival sac. In normal rabbits this procedure causes no visible reaction. In rabbits scnsi tizcd by the intracutaneous method, a dcfin- ite reaction occurs within 24 hours which increases in intensity during the first 24 to 48 hours and disappears after 4 to 8 d3y6. The rent tion consists first of conjunctivitis with dilataticn of capillaries at the sclerocornesl junction, with clouding of ths cornea .Qnd lasz of all development of pfinms which occurs only occasionally. The ophthalmic reaction shons great vnriqtlon in severity in dif fercnt rabbit:: aild may not proceed beyond the stage of cornea1 turbidity. Of over 50 rnimals immunized by the routine intravenous method, none have given a positive eye reaction, while 90 per cent or mart: of the rabbits immunized intra- cutaneously reacted positively Fhen later tested to protein derivatives of Pn8umoceccua) The hypersensitiveness is not elicited in sensitive animals by instillation in the eye of purified type-specific cnrbohydrates. The reaction therefore, is not type- but species-specific. The hyperacnsitive state may persist for at lcn.st four months. An interesting point in this connection is the fact that after the ophthalmic reaction has definitely dispppcared, the intravenous in- jection of nucleoprotein solution often causes :' reappearance of the eye reec tion. 227 215 VI. Thi: Revcrsicn of Jti?l! Frizdlandcr bacilli to the I~SII Forms, and the Interconvertibility of Specific Types. Rumt:rouv attempts have becn pravicusly rn,qd? :o cause reversion of I~RII Friedl&nder bacilli to the 1's" v.qriety by the variaus methods Rhich hnve been effective in the case 0 f Pncumococcus. The result6 of these attempts have been uniform- ly ncgativ,:, and the reversion of an ttRI1 form of Fricdlrlnder's bacillus to its originql soecif ic tyae had never been attained by `Iin vitro11 -- methods . The recent studies of Griffith have demonstrated that, under proper conditions , IIRI' pneumococci may be induced to revert not only to the IlSll antecedent, but even to the `1st' form of n. hcterologous Qpe. The method he employed consists of the simultnncous injection subcu- taneously in white mice of the deposit of large quantities of heat- killed llSll orgqnisms end small amountsof living `~Rll cells. The con- versicn to a different type, when it does occur, is to the type of the culture employed as vaccine. Subsequent confirmntion of Griffith's results in this laboratory, suggested the poS6ibility of the reversion of 6Rl' forms of Friedl#nder bacillus by the newer technique. Up to the present time two sets of experiments have been comple- ted. On one occasion, of six mice injected nith the deposit equivalent to 40 CC* of Type A culture heated for 20 minutes ot W-60oC together . with 0.25 cc* of I'W organism6 derived from Type B or Type C, ? mice II 11 yielded cultures containing both IIR" arid ~~5'1~ colonies. The S colonies proved to bc of the Type A variety. Three mice !vhich received similar quontitiss of heat-killed bscterip alone were sacrificed after 10 days, 2nd in none of them were ,*ny bacteria cultivated from either the site of injection or the heart blood, On a second attempt, 4 mica each `r7ere injected with the deposit of bncilli from ZO cc. of I&E A culture previously haatcd (30 minutes 0 t 56-60oC), toga ther lxith 0.25 cc. of IJRI~ culture dcrivcd from Type B, `Y&PC C nnd Group X. T!vo mice of each group yielded cultures containing both IlRll and IISI' colonies; all th: IISll colonies isoln ted, proved sero logically to be of Type A. Four mice injected with the s=mc quantity of Type A vaccine and 0.25 cc. of "RI' cells derived from Type A, yieldcc' cul turcs composed purely of IlRI( colonies. As ccntrols, 4 mice received the heat-`killed Type A deuosit alone. Tncy were killed after 10 d7ys and cul turcs taken from the site of injection e.nd from the heart blood showed no growth. Since previous results from this laboratory shoscd that Type II Pneumococcus and Type B, Fricdla.nder's bacillus possess similar anti- genie properties, observations were- made on the reversibility cf !lR" Friedlander bacilli Then injected subcutaneously into white mice together with hont-killed suspensions of pneumococcus Wpe II. Three different sets of experiments were cirried out, but in no instance was it found that Type II rncumococcus exerted any influence on the rever- sion of FriodlWder's bacilli to either homolo@us or heterologous typeS. T-JO sets of experiments were performed to determine the effect of heat-killed cells of Type B, Friedlander's bacillus on the reversi- bility of 11Rll Pneumococci. In the first experiment, 3 mice each wcrc injected with the deposit equivalent to 40 cc. of T$pe B bacilli, heated for 70 minutes at !33-60oC together with 0.25 cc. of IIR" forms derived in one cast from Type II and in the other from Type III Pneumococcus. Three of the six mitt? yicldcd cultures of mixed IlRff and fiS1t pneumococcus colonies nnd in e-ch instance the )ISII colony corresponded TV the type from v!hich the IfRlt strrlin ho.d been derived. On a second occasion, 4 mice each nerc injected idith 0.25 cc. of . 217 tl culture of II&1 cells dcriwd raspcctivclbi from c~wh cf the thrcL fixed typss of Pncumococcus toe;; thcr -9th the dcnosit of hcQt;d b,:ctcria (56 - 60oC for 30 minutes) cf ?O cc. cultur;: of Type B, Fricdln:ldtmer,t n.nd r?re now living, Tjlmos t thr2;: ycnrs ( 9% .lnd 1030 dqys) q.fter -heir list exposure to spwyed pncumococci, still protect aice Tgainst infection ;yi th large doses (0.01 CC.) of a virulent cul turn of the SYXB tyne n.s that ni t.h rhich they wore originnl- ly spmycd. `Ihc scra of rabbits sprayed 10 times T:ith .r strain of 113 pncumc- coccus, dcrivi:d from Tyne II orgnnisms cont?incd no type-specific .?nti- bodies. The st`r? c.f rabbits repcntedly exuoscd to spr.ayed mousc- virulent but not r.abbi t-virulent Type II pneumococci coiltained protective nntibcdics but no demonstrable sgglutinins. Tho snm< strain of Type II ws rendered rabbit-virulent by p?.ssnge throu& rl series of rsbbits. The r?bbi ts sprayed ni th this strain s':ov.;cd the presence of both pro- tective antibodies -nd sgglutinins in their serum. Two groups of rabbits hsve beer. rcpestcdly sprayed *xi th a rabbit- avirulent and a virulent `&pe III pneurococcus. The sera of the rabbits exposed to nn .n:mosphcre of rlvirulcnt Type III orgnnism contnincd no type-specific antibodies. Although a large number of the qnimnls died of pncumococcus septicemia follo~4ng the first exposure to the rabbit- virulent strain, those which have survived repeated exposures have, so far, failed to develop any detectible type-specific nntibodies. A series of rabbits were intravenously injected l?ith n. heat-killed sus- pensicn of the same rabbit-virulent strain. The sera of R number of these a.nimals contained both agglutinins and protective antibodies. The duration of the immunity in rabbits actively immunized by the inhnlation method and by inoculation is being followed, VIII. Si@ficnncc of oxidation-reduction processes in bacterial growth. (Dr. Dubos). It has been shown by previous work from this depart- ment that Pnewnococcus cultures respond very readily to changes in the conditions of oxidation-reduction of the medium and environment in Thich they are placed. An attempt is being made to analyze the influence of such fat to rs on the growth of Pneumococcus. The folloVing points are being studied:- Initiation of groath, as measured by the minimum number of cells (used qs inoculum) rcouir;d to initintc poc?h; the grc:7t!: curves md lctg period; the density of growth, and the number of cells dcvclcping per unit volume of medium; nnd the vi?.bility of the cultures. It h?s been found that `In under-St.-ding ci tht problem could not be obtained ;rithout scmc tio.::lcdgc of the oxidnticn-reductizn propcr- ties of storile media. Vhin sterile plain broth is kept in the nbse::cc of oqgen, this medium exhibits a progressive drift tovlnrd highly reducing po ten tinls. `TAis drift has been follol:red elec trometrically and by the reduction of indicators of oxidation-reduction potentials. These indict tcrs are progressively reduced by sterile broth in the order of the electromotive series; the indophcnols being reduced moot rapidly (in a few hours) and indigo disulfonrtc more slowly (in several days or weeks). Indicators with an El, nega.tive to that of indigo di- sulfo'nate were not reduced under any circumstances, These studies also indicate that plain broth contains substances which combine rith oxygen to form toxic products qnd $hat these oxidized substa.nces may be decomposed by heat. 1) fnitiation of growth. The number of cells of Pneumococcus necessary to initiate growth may VZFJ a great deal vith the age, of :qir. Vhcn prccnuti, ns clre t.*.`:l,;: tc pr'vrnt the fcrz:;:ltiwi of th_se tcxic substnnces, or `::h~r. thrv .Qr; rezcvv,d by proper mclns ) groath of Pnamococcus *ad rclq ted o rgncisz C.T: b;: obkincd OVCII Then onu or vcr" , few cells qrc used n.s ir~culum. . 2) Bqctcriostatic .qcticn cf c:xidizcd dy;.s. Another evidacc of tilt: bnctcriostntic action c f cxidized subst?nccs hn.s bol.n obtnindd from the study of thi action of certnin dvcs on bqc L..rinl gr?rth. Whoreqs, the groath of n-7ny organisms is checked be- tki. prcszncc in the medium of tha oxidized foras of certain dyes (indonh*.nals qnd scthylcnc blue fc r ins tancc) , the reduced forms of the saw dyes - cvc:1 in lqrge coilcentrsLtions - erc 79thout qny inhibitory ncticz cn grc,Tth. Thcrt: arc also iildicntio:ls thnt the pl~cc of the revcrsiblz dyes in the oxid+ ti on-rcduc tion scnlt: boars I dcf i:li tc relatiw to their b:lc tcrios tntic gctian for difforeilt crgaisns, 3) Growth CUTVBS .nnd log period. Sc,r;l.c of C~ce~~cy~s cbsorvations on the growth curve of Pne~~~ccccus can bo partly accounted fcr by the ccndi ticns of oxidntiL,n-reduction prevsiling in thb; acdium. The course of development of the culture n'~y bo very much nccclcratcd bj- the pres- ence of reducing substances in tlic medium or the r.`.n in tcnqncc of reducing conditions. The lzg pericd probably ccrresnonds to the tine during which the cells ard reducing the toxic substa!:cas of tint: ::.cdiuI?, i;z IThich process sane of the orga-nisms arc injured n.ld succumb. 4) Viability. It, hc1.s be!:n sho.rn in .T provicus r:.p< ;,t thc1.t the processes of cellulnr oxidaticr. n??rlcudly inflqe:;cc the dca.th rntc of p;lamococci. Under cGi:ditio:;s of active atir!ltir.;i:, `~Xl;tu.:.c~coCCuS 233 221 cultlrrcs fc,iYt? l.wgo :`L-=.ounts r,f` pc?roxidz. It X-B founu t>nt the viability of tho cells is lcssmc?d Alcn the pcroxids producticn is imremod, ;:jhile the cells rcmin ~l.livc loz&ar rlhcn the conditicm ?rc such that the formtim or nccurml.~tion of poroxido is prevented. !RIC oxidntinn-rcducti in systzl.: of pncumococcus cells cmsiSt of trlc csscntinl coqoncnts. CIIC: of th `IT, CT;1 be ramved b,- mnShi35, thi: cells rcpcntedly in snlinc soluticn; this prccesfi results in 3 com~lctc inrlctiv-:ticn of the ~,xid?tinn-reduction systcn. It h?S bt::n obscrvcd that pncmococci lrhich hqv,: bmn deprived cf their :)xidn tion-reducti(,n 2ctivitics by mshi;lg in. mlinc solution rc:min vixblc longer than the origiilnl unnqS?cd ~~11s. In the course of this xrlc, it h?.s bcconc Core nud rmrc ovidont that WI nnqlysis of the oxidzticn-rcductiGn properties of Pno~~occccun cultures io rcsdercd cs?cci~lly difficult by the f,%t that plain brc.th itself is rln wtive nnd dymit-ic systw frc;-. this point of viz::. Tork is in progrcos to develop a synthetic m.cdiun the bchnvior i'f r:hicl? cm be interpretzd :.-ore n.ccurntcly. Pub1 icatf ons. I Avery, O.T., nnd Tillctt, B.S., ~mphylaxie ysitfi the typo-specific cnrbo- hydrates of ]mcumccccus. J.Exp.Med., 1929, xlix, 251. * Damon, 14.H. , Ihe intcrconvcrtibility of IlIP md ?P fcr,ls cf pnemo- COCCUS. J .Exp,Mcd. , 1928, xlvii ) 577r Dubos, R.J., Obscrvotic,ns c,n the cxidnticn-reduction propcrtios cf s torilc br?c tcriologicql mdin. J.Exp.Xod. , 1929, xlix, 507. Dub-OS, FM.) !lI~c initlatira of growth of ccrb'in fscultntivc n::?erobcs .qs rel,Ttcd IX cxid2tiowreduc tiler, processes i:l the ,-odium J.Exp.Med., 1929, xlix, 559. Dubos, R.J., The relation of tho b?,ctcriostetic `sction c f cort.q.in dyes to oxidation-reduction processes. J.Exp.Med., 1929, xlix, ,575. Goobel, IV. F., md Avery, O.T., A study of pnmmcoccus autolysis. J .Exp. Ned. , 1929, xlix, 267. 234 222 Julimelle, L.A., Bgcteriql vari,?.ticn in culturis of FricdMndcr's bacillus. J.Ejrp.kted., 1928, xlvii, 889. Julinn