[Reprinted from Taz Journat or EXPERIMENTAL Menpictne, February 1, 1933, Vol. 57, No. 2, pp. 265-278] FURTHER OBSERVATIONS ON THE USE OF PNEUMO- COCCUS EXTRACTS IN EFFECTING TRANSFORMATION OF TYPE IN VITRO By J. LIONEL ALLOWAY, M.D. (From the Hospital of The Rockefeller Institute for Medical Research) (Received for publication, October 7, 1932) The fact has been conclusively proved by Griffith (1), and subse- quently confirmed by Neufeld and Levinthal (2) and Dawson (3) that R pneumococci derived from S organisms of one specific type may be transformed into S forms of a different specific type. The change was first accomplished in vivo by Griffith, who injected into mice living R organisms derived from one type of Pneumococcus together with large quantities of heat-killed S pneumococci of a different type. Living S forms of the same type as that of the killed S organisms injected, were recovered from the animals. Dawson and Sia (4, 5) succeeded later in bringing about transforma- tion of type im vitro. They inoculated with minute quantities of liv- ing R organisms derived from one type of Pneumococcus a medium of broth containing anti-R serum and the whole heat-killed S organisms of a different type. From the cultures originally inoculated with R pneumococci, S organisms were recovered which were identical in type with that of the killed S pneumococci present in the medium. In a previous paper from this laboratory (7), experiments were de- scribed in which transformation of type had been effected in vitro through the use of cell-free extracts of S pneumococci. These ex- tracts, although filtrable, were crude and contained large amounts of cellular debris now known to be unessential for the transformation. Moreover, the earlier method of preparation resulted in the loss of a considerable amount of the active substance. The present paper describes a more efficient method of extracting pneumococci, and records the results of attempts to purify further the active substance. 265 266 TRANSFORMATION OF PNEUMOCOCCUS TYPES Methods Cultures—The cultures of R pneumococci used in the experiments were stock strains originally derived from type-specific S pneumococci by growth in broth containing 10 per cent antipneumococcus serum of the homologous type. The pneumococci from which the extracts were prepared were type-specific S strains, the virulence of which had been maintained by frequent animal passages. Preparation of Pneumococcus Extract.—The S pneumococci used in preparing extracts were grown in meat infusion broth, pH 7.8, containing dextrose 0.01 per cent. Large inocula were used to bring about rapid and dense growth. At least 100 cc, of a moderately heavy culture was used to inoculate 5 liters of media, The cultures were grown at 37°C. for 8 to 10 hours only. The organisms from 5 liters of culture were thrown down and then taken up in 50 cc. of sterile distilled water. To this suspension was added 3.5 cc. of a sterile 10 per cent solution of sodium desoxycholate. The mixture was kept in an ice bath for 10 minutes, then brought slowly to 60°C. ina water bath. The organisms treated in this manner are quickly dissolved, forming a thick, extremely viscous gel. The preparation was kept in the water bath at 60°C. for 15 minutes to kill any surviving organisms, and to in- hibit or destroy the autolytic enzymes released by dissolution of the cells. The bacterial solution was added slowly to 500 cc. of absolute alcohol previously chilled in an ice bath. A thick, stringy precipitate formed which slowly settled out on standing. ‘The sodium desoxycholate, being soluble in alcohol, remained in the supernatant fluid. The precipitate was thrown down after 30 minutes, by cen- trifugation, the supernatant fluid being discarded. The sodium desoxycholate solution was thus eliminated. The precipitate was washed with alcohol to remove the last traces of the bile salt and was dried ix vacuo, or by exposure for 12 to 14 hours to cold dry air, The dried material was then extracted in 100 cc. of sterile 0.85 per cent sodium chloride solution, made slightly alkaline by the addition of 0.5 cc. of N/10 sodium hydroxide, sealed in glass ampoules, and immersed for 15 minutes in a water bath at 60°C. The preparation was then centrifuged for 30 minutes and the insoluble residue, which contained only a small amount of the active material, was discarded. The supernatant extract was a slightly turbid, thin, opalescent fluid. The extracts were subjected to rigid tests for sterility. They were cultured in plain broth, in blood broth, on blood agar, and in media en- riched by the addition of ascitic or chest fluid. In no instance were pheumococci grown from the extracts. Injected in amounts as large as 1 cc. into mice, they caused no untoward effects other than slight lethargy for a few hours. All the test animals survived. Culture Medium.—Dawson and Sia found in their in vitro experi- ments (4, 5), that the transformation of pneumococci from one specific type to another was facilitated by the addition to the culture medium of anti-R serum. Their observations have been repeatedly confirmed J. LIONEL ALLOWAY 267 in this laboratory. In fact, no transformation has been possible in the author’s experience without the use of serum or a serous fluid in the medium. Hog serum and anti-R rabbit serum were used in earlier experiments. It has been found recently, however, that sterile ascitic or chest fluid is more effective than blood serum, when added to the nutrient broth in proportion of 1 to3. The chest fluid used in the present experiments was obtained from a patient with cardiac failure and generalanasarca. It wasa straw-colored, clear transudate, and proved sterile on culture. Prior to use it was filtered through a Berkefeld V candle, Titrations of its anti-R properties revealed that it would agglutinate R pneumococci only in dilutions of 1-20, or, rarely, 1-40. Cultural Technic.—The cultural technic employed in effecting the transforma- tions was similar, save for minor changes, to that described in an earlier paper (7). Inocula, consisting of 1 drop of a 10~ dilution of an 8 hour culture of R pneumo- cocci, were added to a series of tubes containing 1.5 cc. of broth, 0.5 cc. of chest fluid, and varying quantities, 0.05 to 1.0 cc., of the specific pneumococcus extract. The cultures were grown aerobically at 37°C. for 24 hours. Transfers to fresh medium were made serially every 24 hours, 1 drop of culture being carried forward to a corresponding tube in the second series. At the time of each transfer, sub- cultures were made on blood agar plates for the subsequent study of the colony characteristics of the organisms. No experiment was considered negative until five serial transfers were made. Whenever smooth colonies were noted on sub- culture, one typical colony was transferred to blood broth and the specific type of the organisms was subsequently identified serologically. Activity of Extracts —-Extracts made from S pneumococci by the action of sodium desoxycholate proved much more effective in induc- ing transformations in type than did the extracts which were formerly prepared from frozen and thawed organisms. They were effective in high dilution, and even failed to induce a change if present in too great concentration. Conversion of R pneumococci derived from Type II S organisms into Type III pneumococci could be accomplished in al- most all instances by means of an extract of Type III organisms. Table I shows the results of a typical experiment. Further Purification of Extracts Adsorption on Charcoal.—Attempts were made to separate the ac- tive transforming material from the inert cellular debris by the use of various adsorbents. Powdered wood charcoal was found to give 268 TRANSFORMATION OF PNEUMOCOCCUS TYPES satisfactory and consistent results. Purification with charcoal was carried out as follows: 100 ce. of the extract prepared in the manner described was diluted with an equal volume of sterile 0.85 per cent solution of sodium chloride. 8 gm. of sterile pow- dered wood charcoal was added to the diluted extract and the mixture was shaken for 5 minutes. The preparation was then centrifuged at high speed for 30 min- utes and the supernatant fluid, still containing considerable amounts of finely TABLE I Activity of Extract of S Pneumococct before Adsorption and Purification Specifi Serum R extract . Specific. Tube Broth (chest organism Type mt Colonies agglutination fuid) cocci® First culture ce. cb. ce. fa 1.5 0.5 D-39-RT 0.1 RandS Type III 2a 1.5 0.5 D-39-R 0.1 R only _ 3a 1.5 0.5 D-39-R 0.05 R only ~ 4a 1.5 0.5 D-39-R 0.05 R only _ 5a 1.5 0.5 _ 0.1 Sterile _ First subculture 1b 1.5 0.5 From Tube 1a 0.1 Rand & Type III 2b 1.5 0.5 From Tube 2a 0.1 RandS Type [II 3b 1.5 0.5 From Tube 3a 0.05 Rand 5 Type Ill 4b 1.5 0.5 From Tube 4a 0.05 R only _ * Pneumococcus extract prepared by dissolving Type IIIS organisms in a solu- tion of sodium desoxycholate with removal of bile salt by precipitating the ex- tract in alcohol. } Strain of R Pneumococcus derived from Type II S organisms. suspended charcoal, was removed. This solution was filtered through sterile filter paper and finally through a Berkefeld V candle. The filtered extract was generally water-clear, colorless, and quite limpid. Extracts purified by the use of charcoal were more active in effect- ing transformation than were the crude extracts. Moreover, it was much easier to determine when transformation had occurred in cul- tures containing these clear extracts. The medium was entirely clear and transparent on inoculation. Growth was granular and settled J. LIONEL ALLOWAY 269 out at the bottom of the tube as long as the organisms retained their R characteristics. When S cells developed, however, growth became diffuse throughout the medium. The presence of diffuse growth in the culture medium was presumptive evidence that S colonies would be found later on plating. TABLE It Activity of Pneumococcus Extract after Charcoal Adsorption Serum R strain of Specific extract Specific Tube | Broth toe Pneumococcus derived | | pneumococci* Colonies agglutination nes from Type: — S colonies Type | Amount First culture ce. ce. cc. la 1.5 0.5 I I 0.75; RandS Type I 2a 1.5 0.5 I Il 0.75} Ronly —_ 3a 1.5 0.5 I II 0.75 | Ronly — 4a 1.5 0.5 Ii I 0.75 | Ronly _— Sa 1.5 0.5 Il II 0.75 | RandS | Type II 6a 1.5 0.5 II Til 0.75 | RandS | Type III 7a 1.5 0.5 sank I 0.75 | RandS | Type III 8a 1.5 0.5 III II 0.75 | Ronly — 9a 1.5 0.5 III II 0.75 | RandS | Type IIL 10a 1.5 0.5 _— I 1.0t | Sterile — lla 1.5 0.5 _ I 1.0 Sterile — 12a 1.5 0.5 _— Tit 1.0 Sterile — First subculture 2b 1.5 0.5 From Tube 2a II 0.75 | Ronly _— 3b 1.5 0.5 From Tube 3a Tit 0.75 | RandS | Type IIT 4b | 1.5 0.5 From Tube 4a I 0.75 | RandS | Typel 8b 1.5 0.5 From Tube 8a II 0.75 | RandS | Type II Second subculture 2 | 1.5 | 0.5 | FromTube2 | m | 0.75| RandS | Typell * Extracts prepared by dissolving S pneumococci in a solution of sodium des- oxycholate, precipitating in alcohol, redissolving in salt solution, adsorbing on char- coal, and filtering through Berkefeld V candle; Type I extract prepared from Type I pneumococci, Type II from Type I pneumococci, and Type III from Type III pneumococci. Tt Controls were prepared using 1.0 and 0.5 cc. quantities of the extract in the original experiment which is shown here much abridged. All controls were sterile. 270 TRANSFORMATION OF PNEUMOCOCCUS TYPES Charcoal-adsorbed extracts of Types I, IJ, and III S pneumococci were prepared and found to be highly active in all instances. In Table II are recorded the transformations brought about through the use of extracts after removal of considerable inactive material by char- coal adsorption. As the results presented in Table II show, it was possible to cause R pneumococci derived from each of the three specific types to revert to their original S forms through the use of specific extracts of the homol- ogous type. Likewise, it was possible, with one exception, to effect a selective transformation in type, whereby the R cells derived from each type of Pneumococcus were changed into the S forms of each of the other two specific types in the presence of the appropriate ex- tract. It was impossible, in two attempts made, to effect transforma- tion of R pneumococci derived from Type III S organisms into Type I pneumococci. Instead, these particular R pneumococci reverted to S forms of the original specific type from which they were derived; namely, Type III. The tendency of R pneumococci to revert to S organisms of the original type, even in the presence of a suspension of heat-killed S organisms of a heterologous type, was encountered and commented upon both by Griffith and by Dawson. As stated in the previous paper (7), it was quite easy to convert R pneumococci derived from Type IIS organisms into Type III S forms. In most instances the change occurred in the first culture within 15 to 20 hours. Other R strains changed less readily and often required two, three, and even four transfers. Thus, in the experiment shown in Table II, the R organisms derived from Type I S pneumococci were changed into Type II forms only on the third successive cultivation in the extract-containing medium. Reprecipitation in Acetone or Alcohol.—Still further purification of the charcoal-adsorbed extracts was accomplished by precipitation in acetone. The filtered extract, after charcoal adsorption, was precipitated by 10 volumes of acetone in the cold. The preparation, after standing 1 hour, was centrifuged at high speed for 30 minutes and the supernatant acetone discarded. The sedi- ment was dried in vacuo and taken up in the original volume of sterile distilled water. Much of the precipitate remained insoluble. The insoluble residue was thrown down by centrifugation and discarded. The supernatant contained the active material without appreciable loss. J. LIONEL ALLOWAY 271 Alcohol was substituted for acetone with equal success. Most of the active material was precipitated by 70 per cent alcohol: all by 100 percent. The active substance after precipitation in acetone or alco- hol was soluble in water. No demonstrable loss in potency resulted from treatment with these reagents as is evidenced by the data pre- sented in Table III. Properties of the Extract Resistance to Heat-——The transforming substance was more resist- ant to heat in extracts freed from much of the extraneous material present in the whole pneumococcus cell and in the crude preparations. Griffith and Dawson both found (1, 3, 6) that heating the intact cells TABLE III Activity of Pneumococcus Extract after Precipitation by Acetone R Pneumo- Specific . Serum factor | coccus derived | ¢xtract of : Specific Tube Broth (chest fluid) from Type I Dacumo- Colonies agglutination ce. oc. ce. 1 1.5 0.5 D-39-R 1.0 Rand S$ Type III 2 1.5 0.5 D-39-R 1.0 Rand S$ Type IIT 3 1.5 0.5 D-39-R 0.5 RandS Type III 4 1.5 0.5 _ 1.0 Sterile — * Pneumococcus extract prepared by precipitating the charcoal-adsorbed puri- fied extract of S organisms in acetone and extracting the precipitate in distilled water. to temperatures above 80°C. rendered the bacterial suspensions used in effecting transformations in vivo, incapable of inducing changes in type, whereas heating them at lower temperatures, if the exposure was not too long, caused very little decrease in potency. Experi- ments with purified extracts demonstrated that although these prepa- rations showed a progressive drop in potency on heating above 80°C., nevertheless they were occasionally active after 10 minutes’ exposure in the water bath to a temperature as high as 90°C. The effect on the activity of an extract heated for 10 minutes at 70°, 80°, and 90°C., respectively, is evident from the experimental results given in Table IV. 272 TRANSFORMATION OF PNEUMOCOCCUS TYPES TABLE IV Effect of Heat on the Activity of Purified Pneumococcus Extract Type III Serum 7 specife e extract* Specific fact _— . CHIC. Tube | Broth (ch est organism Heated Colonies agglutination uid) for 10 | Amount min. at First culture ce. ce. °C. ce. la 1.5 0.5 D-39-Rf 70 1.0 Rand S$ Type III Za 1.5 0.5 D-39-R 70 1.0 R only _ 3a 1.5 0.5 D-39-R 70 0.5 R only —_ 4a 1.5 0.5 D-39-R 80 1.0 R only _— Sa 1.5 0.5 D-39-R 80 1.0 R only — 6a 1.5 0.5 D-39-R 80 0.5 RandS | Type II 7a 1.5 0.5 D-39-R 90 1.0 R only —_ 8a 1.5 0.5 D-39-R 90 1.0 R only _ 9a 1.5 0.5 D-39-R 90 0.5 R only _ First subculture 1b 1.5 0.5 From Tube ta 70 1.0 Rand S Type IIT 2b 1.5 0.5 From Tube 2a 70 1.0 R only _ 3b 1.5 0.5 From Tube 3a 70 0.5 RandS Type IIt 4b 1.5 0.5 From Tube 4a 80 1.0 Rand 5S Type It 5b 1.5 0.5 From Tube 5a 80 1.0 Rand $ Type III 6b 1.5 0.5 From Tube 6a 80 0.5 RandS | Type Ill 7b 1.5 0.5 From Tube 7a 90 1.0 R only _ 8b 1.5 0.5 From Tube 8a 90 1.0 RandS | Type III 9b 1.5 0.5 From Tube 9a 90 0.5 R only _ Second subculture 2c 1.5 0.5 From Tube 2b 70 1.0 RandS Type III 7c 1.5 0.5 From Tube 7b 90 1.0 R only _ &c 1.5 0.5 From Tube 8b 90 1.0 RandS Type III 9c 1.5 0.5 From Tube 9b 90 0.5 R only _ Third subculture 7d 1.5 0.5 From Tube 7c 90 1.0 R only _- 9d 1.5 0.5 From Tube 9c 90 0.5 R only ~~ * Extract of Type III S pneumococci prepared by dissolving the culture in a solution of sodium desoxycholate, precipitating in alcohol, and extracting the precipitate in saline solution. { Strain of R Pneumococcus derived originally from Type II S organisms. J. LIONEL ALLOWAY 273 From the data presented in Table IV, it is apparent that pneumo- coccus extracts became progressively less active after exposure to temperatures above 80°C., and that the loss incurred was approxi- mately in direct proportion to the degree of heating. However, it is also evident that in the present state of purity, an extract occasionally retained sufficient activity to bring about changes even after heating to 90°C. for 10 minutes. Boiling invariably destroyed the activity of the purified extracts. TABLE V Activity of Purified Pneumococcus Extract after Filtration through Berkefeld W Candle Specific Serum tract Specifi Tube Broth (chest orgatism prepared Colonies agglutination Huid) Type III S* First culture ce. ce. oe. la 1.5 0.5 D-39-Rt 1.0 R only _ 2a 1.5 0.5 D-39-R 1.0 R only _ 3a 1.5 0.5 D-39-R 0.5 R only _— 4a 1.5 0.5 D-39-R 0.5 R only _ Sa 1.5 0.5 D-39-R 0.25 R only a 6a 1.5 0.5 _ 0.5 Sterile _ First subculture 1b 1.5 0.5 From Tube 1a 1.0 RandS Type III 2b 1.5 0.5 From Tube 2a 1.0 Rand S$ Type III 3b 1.5 0.5 From Tube 3a 0.5 Rand S$ Type IIT 4b 1.5 0.5 From Tube 4a 0.5 Rand S$ Type III Sb 1.5 0. From Tube 5a 0.25 R only _ * Extract of Type III pneumococci prepared by filtration of purified, charcoal- adsorbed extract through Berkefeld W candle. { Strain of R Pneumococcus derived from Type II S. Filirability —It was stated in a previous paper (7) that pneumo- coccus extracts prepared by freezing and thawing the bacterial cells, could be filtered through Berkefeld N candles and still retain their capacity for effecting changes in type. Recent investigations showed that when the purified extract was alkaline in reaction, filtration could be carried out without demonstrable loss of potency, not only tt i Foe ; oe EOI a 274 TRANSFORMATION OF PNEUMOCOCCUS TYPES through the N but also through the W type of Berkefeld filter. When the extract was acid, however, filtration resulted in complete loss of _activity. The purified extracts passed through W filters were crystal- clear and colorless. They still exhibited full activity as demonstrated by the results of the filtration experiment recorded in Table V. TABLE VI Active Immunity Induced in Mice by Injection of Purified Pneumococcus Extract (Type I) Type I extract* injectedt Virulent culture of Mouse pe I pneumococci Result Apr. 27 Apr. 30 | May 4 injectedt May 11 ce. ce. a. ce. 1 0.2 0.2 0.2 0.001 S 2 0.2 0.2 0.2 0.001 D 18 3 0.2 0.2 0.2 0.0001 D 18 4 0.2 0.2 0.2 0.0001 s 5 0.2 0.2 0.2 0.00001 S 6 0.2 0.2 0.2 0.00001 S 7 0.2 0.2 0.2 0.000001 S 8 0.2 0.2 0.2 0.000001 D 60 Virulence Control Mouse Virulent Type I culturet Result ce. 1 0.001 D 18 2 0.0001 D 18 3 0.00001 D 24 4 0.000001 D 40 5 0.0000001 D 40 6 0.00000001 D 40 * Extract prepared by dissolving Type I pneumococci in a solution of sodium desoxycholate, precipitating in alcohol, extracting the precipitate in saline solu- tion, adsorbing with charcoal, and filtering through Berkefeld V candle. } All injections made intraperitoneally. S = survived; period of observation 21 days. D = died; the numeral indicates the number of hours before death of the animal. Soluble Specific Substance.—The purified extracts contained vary- ing amounts of the soluble specific substance of the pneumococci from which they were prepared. However, the concentration of the type- coy J. LIONEL ALLOWAY 275 specific substance was relatively low, since in most instances the puri- fied extracts reacted specifically in antipneumococcus serum of the homologous type only in dilutions of 1-40, or occasionally 1-80. This indicates a probable concentration of the specific capsular polysac- charide of approximately 0.01 mg. per cc. of extract. Antigenicity.—It was of interest to learn whether these extracts possessed antigenic as well as transforming properties. It was found that the extracts as prepared were not only active in effecting trans- formation in type, but also were capable of inducing active immunity in mice treated with the purified preparations. The protocol shown in Table VI illustrates the degree of active immunity induced in mice by repeated intraperitoneal injections of an active extract. From the results recorded in Table VI, it can be seen that consider- able active immunity was induced in mice by three injections, at 3 day intervals, of 0.2 cc. of purified extract of Type I pneumococci. Of 8 mice so treated and inoculated 7 days later with a strain of viru- lent Type I Pneumococcus, 5 survived an infecting inoculum 100 to 100,000 times greater than that which invariably proved fatal to the normal control animals. DISCUSSION Extracts of S pneumococci prepared by the dissolving action of sodium desoxycholate were as active as were the intact cells themselves in causing R forms to assume type-specific characters. Their action was specific, the change in type being selectively determined by the specificity of the extract employed. The active substance or sub- stances in the crude extract inducing the changes could be consider- ably freed from accompanying impurities by precipitation in alcohol, by charcoal adsorption, and reprecipitation in alcohol or acetone. These procedures seemed not to decrease the potency of the active preparations. In fact, on further purification, the extracts exhibited in most instances increased activity, inducing more prompt trans- formation. Despite the fact that the capsular polysaccharide of the Pneu- mococcus determines its type specificity, it was not possible to corre- late the activity of the extracts which stimulate the development of type-specific characteristics in R pneumococci, with the presence of paces E Pm pn inner nc i mR 276 TRANSFORMATION OF PNEUMOCOCCUS TYPES the soluble specific substance. Extracts which were apparently equally potent in causing transformation varied considerably in their content of soluble specific substance. It has been proved (6) that the specific capsular polysaccharide in chemically purified form, as such, is ineffective in inducing transformation in type. It seems probable therefore, that if the soluble specific substance in these extracts is concerned at all in the reaction, it is present there in a different physical state, or in combination with some other substance which confers upon it properties not found in the chemically isolated and highly purified substance. Certain strains of R pneumococci were found to be more resistant to transformation than were others, but none were encountered which were completely refractory. Transformation of R pneumococci de- rived from Type II strains to the Type III S forms in the presence of Type III extract seemed to take place almost abruptly. Very rarely were transitional colonies noted. Likewise, under the influence of Type II extract, R cells derived from Type III organisms changed abruptly to Type II pneumococci. However, the change of R forms derived from Type II pneumococci to Type I S organisms was a more gradual one, and required, at times, a series of transfers in the specific extract medium. In this instance, all stages in transformation were noted in colonies plated from an individual culture. Once the change was complete, however, the newly acquired type-specific characters persisted. The factor which is presumably common to blood serum, ascitic and chest fluid, and which isessential in the reaction, remains unknown. No experiments were successfully completed without the addition to the medium of one or another of these three related substances. The present experiments afford additional evidence that the trans- formation in type is not apparent, but real, and that the changes are brought about in the presence of the extract through the specific ac- tion of a soluble constituent present in S forms of pneumococci. It is almost inconceivable that any living element in the pneumococcus cell could survive the drastic procedures employed in the preparation of the extracts. Through the action of sodium desoxycholate, pneu- mococci were completely dissolved so that no recognizable cellular forms remained. The extracts were heated to 60°C. for a total of 30 J. LIONEL ALLOWAY 277 minutes during the course of preparation, and in sealed glass tubes were completely immersed in the water bath during part of the heat- ing. They were exposed to the action of absolute alcohol for 30 minutes, and were saturated with alcohol of varying strength for several hours. They were treated with charcoal which removed much of the particulate matter, and were finally filtered through Berkefeld V filters. They could even be heated to 90°C. for short periods, or passed through Berkefeld W filters and still remain active. Controls of sterility were exceedingly rigid. The extracts were injected in large amounts into mice without any untoward effects. All cultures of the extracts and of animals sacrificed at various intervals after the injection of active material, were sterile. The exact nature of the active material in these extracts still re- mains to be determined. That it acts asa specific stimulus to the R cells which have potentially the capacity of elaborating the capsular polysaccharides of any one of the several types of pneumococci seems clear. SUMMARY Pneumococcus extracts highly active in inducing the i» viiro trans- formation of the specific types of Pneumococcus have been prepared by dissolving 5 cells with sodium desoxycholate, precipitating the dis- solved material in alcohol in which the bile salt remains soluble, and extracting the precipitate in salt solution. Further purification of these active extracts has been attained by the removal of considerable inactive material by charcoal adsorption and by reprecipitation of the adsorbed extract in alcohol or acetone. The importance of using young cultures for extraction, and of preventing autolysis during the preparation of the extracts, is emphasized. Extracts prepared by the method described have been filtered through Berkefeld Candles (V, N, and W) without appreciable loss in activity, provided the reaction of the extract was slightly alkaline at the time of filtration. The purified and filtered extracts are water- clear, and sterile by rigid cultural and animal tests. They have been heated to temperatures of 60°C. for 30 minutes without appreciable loss in their capacity to induce specific changes in type. And al- though they have generally shown definite decrease in potency after 278 TRANSFORMATION OF PNEUMOCOCCUS TYPES heating to temperatures above 80°C., some extracts have been found active even after an exposure of 10 minutes to a temperature of 90°C. They have been completely inactivated by boiling. Relatively small amounts of extract have been effective when added to a broth medium containing normal serum or serous fluid. In this medium, R pneumococci, irrespective of their type derivation, have developed and thereafter retained all the type-specific characteristics of the encapsulated S cells from which the extract was prepared. The specific action of the extracts is discussed with reference to their transforming and antigenic properties. BIBLIOGRAPHY 1. Griffith, F., J. Hyg., 1928, 27, 113. 2, Neufeld, F., and Levinthal, W., Z. Immunitatsforsch., 1928, 65, 324. 3. Dawson, M. H., J. Exp. Med., 1930, 51, 123. 4. Dawson, M. H., and Sia, H. P., Proc. Soc. Exp. Biol. and Med., 1929-30, 27, 989. 5. Dawson, M. H., and Sia, R. H. P., J. Exp. Med., 1931, 54, 681. 6. Sia, R. H. P., and Dawson, M. H., J. Exp. Med., 1931, 64, 701. 7. Alloway, J. L., J. Exp. Med., 1932, 55, 91.