LYSOGENICITY OF SALMONELLA E. GROUP ORGANISMS AND xELATIONSHIP BETWEEN BACT: RIOPHAGE AND ANTIGENIC CHANGES IN SALMCNELLAS OF GROUP E SISAO UETAKE, TAKHYA NAKAGAWA AND TORU AKIBA Department of Microbiology, Sapporo Medical College INTRODUCTION Though there had been many papers on induced changes in H antigens of Salmonella, no report had appeared on induced changes in 0 antigens until Bruner and Edwards described a short note in J. of Bacteriology, 1948, in which 0 antigens of S. anatum and Ss. meleagridis were reported to undergo a change from 3,10 to 3,15 and vice versa by antiserum. This note attracted the authors! attention as they were interested in induced antigenic changes. UVetake and Obara (1950), Obara (1950), and Nakagawa (195la; 1951b; 195348; 1953b) extended this experiment and found that 0 antigens of S. newington and S. new-brunswick were converted from 3,15 to 3,10 by cultivation in semi-solid medium containing 15 antiserun, and vice versa. Further Nakagawa (1951b; 1953b) and Uetake et al. (1952) pointed out that the antigenic changes from 3,15 to 3,10 are not analogous to transformation reactions (Austrian, 19528), Which have been observed in Pneumococcus (Avery et al., 19h); Austrian, 1952b; Hotchkiss, 1952; McLeod and Krauss, 1953), Escherichia coli (Boivin, 19).7), Hemophilus influenzae (Al exander and Leidy, 1950; 195la; 1951b; Leidy et al., 1953; Zamenhof et al., 1953), and Neisseria meningitidis (Alexander and Redman, 1953), but is rather similar to the mechanism of antigenic changes observed in Pneumococcus (Laros, 199), Paramecium aurelia (Sonneborn, 197) and/or phase variation in H antigens of Salmonella. Soon there- after it was observed that 0 antigens of S. senftenberg were changed from 1,3,19 to 3,15 by 1,19 antiserum and further spontaneously changed to 3 alone. (Usetake et al., 1952; Ise, 1953a, 1953b). Shortly thereafter Zinder and Lederberg (1952) reported a new phenomenon, transduction. In 1953 Iseki and Sakai (195398; 1954b), using S. london, S. give, S. anatum, S. newington, S. selandia, S. new-brunswick and S. senftenberg, confirmed the authors's findings and further added that the antigenic change from 3,10 to 3,15 could be induced by infection with phages which were obtained from Eo group organisms. . As this phenomenon is related to transduction and noticeable, the authors have confirmed, extended the experiments, have been and are continuing the study at present. An outline of the ra- sults was reported at the 27th General Meeting of Japan Bacterio- logical Society, in April 195), which will be described in this paper. During the above research, Iseki and Sakai (1953b) reported that the antigenic changes from 3,15 to 4,10 were induced by phage antiserum. Lederberg and Edwards (1953), and Storker, Zinder and Lederberg (1953) reported on transduction of flageller characters in Salmonella. Another related phenomenon has been observed in Corynebacterium diphtheriae. It has been reported that nontoxi- genic strain of Corynebacterium diphtheriae was changed to toxi- genic by infection with phages obtained from toxigenic strain. In this case it is pointed out that the induced variant cells were lysogenic and variation rate was much higher in transduction (Freeman, 1951; Freeman and Morse, 1952; Groman, 1953). The authors! findings on antigenic changes in El, Eo, and By organisns of Salmonella are similar to the latter findings rather than to transduction rs far as hitherto studied. MATERTALS AKD METHODS 1. Strains? S. senftenperg Aa' is a strain possessing 43 factor alone as 0 antigen, which Ise(19543r) obtained as a variant of §. senftenberg A. S. senftenberg HS-1 throug: HS-10 are strains received from Bact. Dept., Faculty of Veternary Medicine, Hokkaido University, all of which were isolated from chicken eggs. All other strains of Fy. Eo and Ez groups are international standard strains which were sent from Kauffmann to Bntercbacteria- ceae Committee of Japan, the donor. 2. Immune sera: All of H and 0 sera were preparsc according to the rovtine methods. (Kauffmann, 1951) 5. Flagellar and somatic antigens were tested by slide agglutina- tion, using suitably diluted absorbed sera. When necessary, tube agglutination and cross absorption test were carried out. he Phage suspensions were prepared by autolysis of lysogenic cells or by mixed cultivation of lysogenic and susceptible cells. Thus, about 300mg (wet weight) of 2l-hour agar culture of lysogenic strain were suspended in orl of saline and kept at 37° for 8 nours. The bacteria were remowed by centrifugation and the supernate was filtered by Cnamberland Lz. In some cases, mixed broth cultures of lysogenic and susceptible cells were centrifuged and the supernate was heated at 58°C for 30 to ho minutes and then filtered by Chamberland Lz. The filtrates were used as phage suspensions, and tested for sterility, both at the time of preparation and when used. 5. Induction of antigenic charges by phege -3- Rech phage suspension was mixed with broth in proportions of 53 to 7, Susceptible cells of =#) and E% group were cultured in phage broth mixtures, and when necessary, serially subcultured everyday up till the 10th passage. At apprcepriate intervals, the cultures were plated on nutrient agar and single colonies were subjected to serological diagnostic test. Further repeated single colony isolations were made if the results were at all ambiguous. In many cases single cell cultures technique was applied by using Peterfi's type of micromanipulator. 6. Plegue counting was conducted by agar layer method. 7. Test for lysogenicity of induced variant strains 1) After soaking tke culture in phage antiserum, single cell isoletions of induced variant strains were made with Petorfi micromanipulator. From each single cell culture autolysate was prepared and its lytic activity was tested by plaque count- ing and by nephelometric determination of the turbidity of broth culture of mixture of the autolysate and Ej, group organisms. 2) Each autolysate was also tested for its capacity of inducing antigenic changes in E, group organisms. 4) It was also tested whether each sinzsle cell culture was immune or susceptible te the phages obtained from Ho gercup organisms. &. Induction of antigenic changes by antiserum The 15 antiservm was prepared by absorbing S. newington“O" anti- scrum witn S. london. The variant strains to be tested, possessing 15 antigenic factor, were cultured in broth containing 15 anti- serum and when necessary, serially svbcultured. At appropriate -h- intervals the culture was plated on agar and single colonies were tested by monofactor serums. 9. Enzymatic treatment For enzymatic treatments, the phage suspension obtained S.cambridge . 9 from ~\, —' Was employed, which contained about x 10 phage particles per ml, calculated from plaque counting by using S. anatum as an indicator strain. For pepsin digestion test, Merck's crystallized hog gastric pepsin was used, and casein was used as a control substrate. The pH of phage suspensions was corrected to 4.8 by adding phosphate buffer solution, though the optimal PH for pepsin digestion is at 2.0 through 1.0. It is because the phages used were found to be inactivated by standing overnight in refrigerator at pH 43.0. For trypsin digestion Merck's crystallized trypsin was used and casein as a control substrate. The pH of phage suspensions was kept at 8.0, the optimal PH, during digestion. AS a desoxyribonucl ease preparation, varidase (Lederle) was used, which was a commercial preparation, containing streptokinase and streptodornase. Lytic activity and | antigenic change inducing activity of phage suspension were tested after keeping the phage-varidase mixture at 30°c for 5 hours. EXPERIMENTAL RESULTS I. Lysogenicity of E2 group strains and phage susceptibility of Ei group strains. Autolysates were prepared from S. newington, S. new-brunswick, S. selandia, S. cambridge, S. kinshase, S. canoga, S. illinois ari S. thomasville, and 0.3ml of each autolysate were mixed with 0.7ml of nutrient broth, in turn. In autolysate broth mixture, various type strains of E, group were separately cultivated in turn and, when necessary, serialy subcultured uptill loth passage. Each cul- tures were separately plated out in turn on nutrient agar and the colonies were serologically examined by monofactor serum. The results are summarized in table 1, which indicates the followings. 1) Whatever the strain may be, the organisms of Ep group, possessing 15 antigenic factor as a part of 0 antigens, were found to be lysogenic and the autolysates were found to be capable of inducing changes of 0 antigens of E, group cells from 4, 10 to 4,15. 2) The susceptibility to a definite phage of EK] group organisms varied with strains, some of which were readily changed in their O antigenic structure (for example; S. anatum, S. batantan, S. simi) while others were not so readily altered (for example; S. london, S.meleagridis). 3) The capacity cf inducing antigenic variation in a definite strain of Ej group seems to vary with phage strains. This may, however, depend upon the amount of phage, though the details have not been made clear as yet. Beside the above, it was observed during these pxperiments that culture of E] group organisms was less turbid in the autolysate- broth than in broth alone. This is due to partial bacteriolysis. II. Antigenic changes in S. lexington and S. macallen. In both §. lexington and 8. macallen, 0 antigens are 4,10, which are the same as those of the type strains described above, but they were changed to 3,10,15 by the phage obbained from S. cam- bridge. This type of change was also induced by the phage obtained from other type of organisms of Ep group (S. kinshasa, S. canoga). Since this finding is very important in that two antigenic factors, 10. and 15, are able to co-exist in a single cell, the — antigenic structures of parent strains and induced variants were reexamined by single cell culture technique with micromaniplator, and the above findings were reconfirmed. In the previous papers and the date described in the preceeding paragraph, only the changes from 4,10 to 3,15 or from 3,15 to 3,10 had been observed, but the coexistence of 10 and 15 in a single eal) had not been observed as yet. Therefore, this was the first evidence of the coexistence of 10 and 15. III. Antigenic changes in S. chittagong. The 0 antigenic structure of 5S. chittagongis shown as 1,4,10, 19. As the phages from Eo group organisms seemed to attack the organisms, possessing 10 antigenic facbor, S. chittagongwas ex- posed to the phage. By each of the phages from S, newington and from S. canoga, 0 antigens of S. chittagongwere converted from 1,3,19 to 3,15 (table 2). In this case, it is noteworthy that the antigenic changes involved the loss of 10 as well as 1,19 anti- -%- genic factors. The following experiments, however, showed the possibility of coexistence of 15 and 1,19 in a single cell. IV. Antigenic changes in S. niloese and S. senftenberg-simsbury When these two strains were serially subcultured in broth containing the autolysate of S. canoga or the mixture of auto- lysates of S. newington, S. new-brunswick and 8S. selandia and/or when serially aubcultured in broth contatning 1,19 antiserur, which was prepared by absorbing S. niloese 0 antiserum with S. london, the induction of antigenic changes was not observed in so far as examined. However, when these two strains were subcultured in broth containing both the mixture of autolysates of the above- mentioned three strains and 1,19 antiserum, antigenic variants were obtained from each of them, which were agglutinable by 15 monofactor serum. As these variants were agglutinated not only by 15 antiserum but also by 1,19 antiserum, antigenic structures were reexamined by single cell culture technique, which revealed that the 0 antigens of the variants were 1,3,15,19 (table 2). In short, it was confirmed that 0 antigens of S. niloese and S. senftenberg-simsbury were converted from 1,3,19 to 1,4,15,19. In these cases it is noteworthy that the induction of formatian of 15 antigenic factor did not result in the disappearance of 1,19 antigenic factors, when compared with antigenic variation of S. chittagong, as described above. The reason, however, has nct been made clear as yet, wheré from this difference comes. V. Antigenic variation of S. senftenberg 87Aa'. S. senftenberg 87Aa!' is a strain which was obtained from S. sengtenverg A by Ise and has only a 3 factor as a major O antigen. - 8- O antigenic structure of this strain was changed from 3 to 3,15 by being exposed to the phage from S. canoga. This change was accomplished without subculture (table 2). VI. Lysogenicity of antigenic variants. As described above all strains of Eo group, possessing 15 antigenic factor, weréeen found to be lysogenic and the phages obtained from such strains nawereeen found to be capable of inducing antigenic changes in FE, group cells, resulting in changes from 3,10 to 3,15 or to 3,10,15. Then experiments were conducted to determine whether the induced antigenic variants possessing 15 factor are lysogenic or not and, if any, whether or not phages obtained from such variants are capable of inducing the same antigenic changes as those obtained from E2 group organisms. The results were aa follows. 1) Variants possessing 15 factor, derived from E, group strains, were found to be lysogenic and the phages obtained from such variants were found to be capable of inducing antigenic changes from 3,10 to 3,15 in E, group organisms. These observations are consistent with Iseki and Sakai's report. The antigenic variants were found also to be resistant to the phages obtained from Ep group organisms. 2) Variants of S. chittagong, possessing antigenic structure of 3,15, were also found to be lysogenic and the phages obtained from them were found to be capable of inducing the same antigenic changes, 3) In variants of 8. lexington and 8. macallen the states were found to be the same as described above, as to the lysogenicity -9- and the activity of the phages. ) In variants of S. niloese and S. senftenberg-simsbut'y the states were also found to be the same as described above, as to the lysogenicity and the activity of the phages. These findings indicates the close correlation between lyso- genicity and formation of 15 antigenic factor in Salmonella organisms of E,, Ep and Es group and/or their antigenic variants. VII. Correlation between lytic acitivity and activity — inducing antigenic changes 1) Effect of pH Aliquots of phage auspensions obtained from S. cambridge were separated into several portions, in each of which pH was corrected to pH 1.0, 2.0, 3.0, .0, 5.0, 8.0, 9.0, 10.0 ana 11.0 respectively and kept in refrigerator for 20 hours. Both the lytic activity and the activity of inducing anti- genic change of the phage were found to be destroyed at pH 3.0 or below and at pH 11.0 or over, 2) Effect of temperature The results of plaque counting showed a considerable reduction in number of active phage by heating at 60°C for 30 minutes and a complete inactivation of phages by heating at 65°C for 30 minutes or at 70°C for 15 minutes. The activity .— induc@ing antigenic change was found to be retained after heating at 60°C for 30 minutes, but not at 65°C for 30 minutes or at 70°C for 15 misippes. 3) Effect of enzymes (1) Phage suspension obtained from S. cambridge was treated with pepsin at pH .8 at 37°C overnight, without any reduction -10- of lytic activity or destruction of activity inducing antigenic changes. (2) Phage suspension was treated with trypsin at pH 8.0 at 37°C for 20 hours. The results were the same as those in pepsin digestion. (3) Varidase treatment resulted in, destruction of geither lytic activity #¢or activity of inducing antigenic changes. h) Turbidity-of broth cultures The cultures of susceptible cells in phage-broth mixture, from which antigenic variants were isolated, were always less turbid than those in broth alone. Though the grade of turbidity was not exactly measured, it is clear that there is a certain correlation between bacteriolysis and antigenic changes. All of the above findings indicate. that the lytic activity is parallel to the activity @® inducing antigenic changes, and no evidence has been obtained by which to indicate the separation of the two activities. VIII. Induction or selection ? The possibility that the antigenic variants were spontaneous mutants selected by the phage minimumized by the following. Even when S. anatum or S. butantan was exposed to the phage for only one minute and plated out, many antigenic variant colonies were detected on the plates. And the number of colonies of antigenic variant were so numerous and the time of exposure vebiants to the phage was so short that “her could not have been selected but’ eould reasonably be considered as induced by phage infection. they -~ 1% .- In the course of this experiment another noticeable phenomenon was observed. Thus, it was noted for the first time that the anti - genic variantda formed translucent colonies which could be easily distinguished from the colonies of parent cells. Such translucent colonies were formed ew when the parent cells were sxposed to the phage for a short time and plated out at once, and pure cul- tures obtained from such translucent colonies did not show the aforesaid transparecy. This is a very interesting pnenomenon, though the mechanism involved has not been rade clear as yet. IX. Reversion of converted antigens. The 0 antigens 4,15 of the phage-induced variants of E] group organisms were reverted to the original structure 4,10 by cultivat- ing the variants in broth containing 15 antiserum. In the variants of S. chittagong, O antigens were reverted to 1,4,19, the structure of the parert strain, py 15 antiserum. In the variants of S. senftenberg-simsbury and S. niloese, C antigens were reverted to 1,5,19, the structure of the parent strain, by 15 antiserum. X. H antigens In each of the 0 antigenic variants, 4 antigens were fcund to be the same as those of the parent strains respectively. XI. Biochemical behaviors Biochemical benavicrs of the antigenic variants were found to be the same as those of the parent streins respectively. - le - DISCUSSION The above findings may be summarized as follows. 1) The antigenic changes hitherto observed, are summarized as below. (EyEo) (E Ez) (E2E3) 3,10,15 1,3,10,19 1,3513,19 p P| |s s| |pre p v 3,10 —— 3,15 <1, 3,19 (Ey) s (Eo) s (Ez) m| {|p 3 = Direction Direction af af of Direction of antigenic changes inducec by phage infection. of antigenic changes induced by antisdrums. of spontaneous mutation. 2) As far as tested, all of the Epo group strains were found to be lysogenic and the phages obtaired foom them were capable of inducing antigenic changes from 3,10 to 43,15 or to 3,10,15 in Ey group organisms. They were also capable of inducing anti- genic changes in Ez group organisms from 1,3,19 to 1,3,15,19. In special cases, they induced antigenic changes in S. chittagong from 1,3,10,19 to 3,15 and in S. senfterberg 87Aa! from 3 to 3,15. 3) In so far as tested, all of the ‘nduced variants, possessing 15 factor as a part of © antigens, were found to be lysogenic and the phages obtained from them were found to have the capacity of inducing antigenic changes, in comnection with the formation - 13 - -of 15 factor, in E; group organisms. This point seems to be similar to Groman's findings in Corynebacterium diphtheriae, and to be different from transductior which was observed in Salmonellas by using bacteriophage of 8. typhimz:rtum. In transduction it is pointed out that the transmission of any Givén character can be readily seperated from the actionof the phage. (Lederberg and Edwasra, 1954; Stocker, Zinder and Leder- berg, 1953) lh) In transduction, phages which were propagated on one strain, transuitted some genetic trait of this donor strain to the re- cipient strain, wnile in the autbors! experiments phages invariably induced the activity forming 15 factor in By greup orgenisms, even when they were propageted on Ei rroup cells. 5) In many of the Ey eroup organisms 0 antigenic strictures weze altered from 3,10 to 4,15, but it was suggested that they still retained their latent activity forming 10 antigenic factor, because when the variants were cultured in broth containing 15 antiserum, 0 antigenic structures were reverted to 5,10. This is also a very important point of difference which exists be- tween transduction and the authors! observations. In trans- duction, "a replacement of genetic trats" has heen pointed out (Stocker, Zinder and Lederberg, 1954), while in the authors! experiments there has been no evidence as yet that the antigenic changes from 3,10 to 3,15 are considered to involve the replace- “sment of genetic traits for formation of 10 antigenic factor or some other character ty genetic traits for formation of 15 factor. - 1h - In antigenic changes of S. chittagong, the states seems to be the same as described above. 6) In the previous papers antigenic changes were reported to occurred only from 4,10 to 4,15 or vice versa, and the formation of 15 antigenic factor seemed to be followed by the disappear- ance of 10 factor. The states were the same in many of El group strains used in the experiments herein, but the coexis- tence of 1c and 15 antigenic factors was observed for the first time in two strains. Thus in S. lexington and §. macallen, 0 antigenic structures were changed from .4,10 to 4,10,15. This fact and the findings, that 43,15 were changed to 4,10 by anti- serum, suggest that the antigenic changes from 3,10 to 4,15 do not involve such "a replacement cf genetic traits" as observed in transduction. 7) Antigenic changes from 1,4,19 to 1,4,15,19 suggest the same as mentioned above. In this experiment, the changes from 1,3,19 to 1,3,15,19 were observed only when the test strains were exposed to a mixture of phage and 1,19 antiserum, and not when exposed to the phage alone or to the antiserum singly. However, the role of anti- bodies is not clear as yet, because the variants still retain 1,3,19 factors. Besides, it can not be denied that the differ- ence among results of the above experiments might be only an apparent, depending upon the number of colonies examined in this kind of experiment that has to be limited to a certain small number. 8) In §. anatum and s, butantan, both of which are liable to -15 - undergo antigenic changes by phage infection, many antigenic variant colonies were detected even after exposing the cells to the phage suspension for only a very short time. This high variation rate is in contrast with the low frequency in trans- duction, and is rather similar to Groman's findings. This difference, however, could be quantitative rather than qualita- tive in nature, as pointed out by Groman (1953). 9) When phage was inactivated by the change of pH or heating, the activity inducing antigenic changes was lost simultaneously. 10) Desoxyribonuclease did not inactivate the activity inducing antigenic variations of phage suspension. This indicates that the said phenomenon differs from transformation which has been observed in antigenic changes of Pneumococcus, H. influenzae, E. coli and meningitidis. In transformation, the active princi- ple is inactivated by desoxyribonuclease. 11) Among the induced variants, those possessing 0 antigenic structures of 3,10,15 and 1,3,15,19 have not been isolated as ‘yet from natural sources. Furthermore, the immediate practical contributions of these variants are probably in the availability of the above for use as antigens for serum production. 12) In the authors! experiments antigenic changes have been ob- served in 0 antigens only but not in H antigens. These facts, however, do not exclude the possibility that changes in H anti- gens may occur by phage infection. It depends on the method of detecting the variants, whether or not H antigenic variants can be discovered. If H antigenic variants should be sought, it can not be denied that they might be discovered. Experiments - 16 - along these lines are now under study. 14) It has been observed that anti-15 monofactor serum induced antigenic chenges from 4,15 to 3,10, from 1,4,15,19 to 1,3,19 and from 3,15 £0.’3,10,19. Iseki and sakht Péported that phage antiserum induced a change from 3,15. to 3,10. As anti-15 mono- factor serum was prepared by absorbing S$. newington "oO" anti- serum with E, group organisms, it can not be denied thet it might have contained antiphage antibodies. This is new under investigation. Even though Iseki and Sakai's report were correct, there remains another question to be determined, as to how the ‘anti-phage antibodies induced the variation from 5,15 to 4,10. Sakai and Iseki (1953) reported that the phage anti- bodies combined with prophages within E, group celis. This explanation, however, is difficult for the authors to accept for the following reasons: The first reason is that there has been no evidence as yet that prophages can be neutralized by phage antiserum or that they possess any antigen which reacts with antiphage antibodies. (McKinley, 1925; Freeman and Morse, 1952; Lwoff, 1953) The second reason is that even when cells cf =. coli B were soaked in anti-To sertim for a 5 days period after infection with phage To, neutralization of phage (gonophage) was not observed within the cells. (Okada, personal: commhication). Now, if the prophages can not be neutralized by antiphage antibodies, what would be the appropriate explaration of the mechanism? It has been reported that lysogenic strain may a yield nonlysogenic cells in its progeny though in very small -17- number (Lwoff, 1953). This may be true in E, group strains. If so, it should be possible that such nonlysogenic cells would be selected by phage antiserum. Nonlysogenic cells are supposed to be yielded, if any, in a very small number and to be infected by free phages present unless antiphage antibodies are added. This may be one possible explanation, and further study will be required in the future. 1h) In the early stage of the research, the phages obtained from Ep group organisms seemed to be adsorbed by strains which carry the 10 somatic antigen. However, from the findings that they were also adsorbed by Ez group strains, it is doubtful whether the receptor for the bacteriophage can be the 10 antigenic factor itself, though the 10 factor has been reported to be of a complex nature (Kauffmann, 1938). 15) Taking H antigens beside 0 antigens into consideration#s, when 0 antigens of S. anatum, S. nyborg, S. meleagridis, S. give and S. uganda are altered from 3,10 to 4,15, these strains are changed to forms which are indistinguishable from S. newington, S. selandia, S. cambridge, S. new brunswick and S. kinshasa respectively. Table 3 shows a comparison of the sources from which they were isolated. Though the summarized data in table 4 are not complete as not all of the references concerned were available, they show that each pair of S. anatum and S. newington, S. nyborg and S. selandia, S. give and S. new-brunswick, and S. lexington and §. illinois has been isolated from similar sources respectively. Besides, the corresponding two types are identical or almost identical in biochemical behaviors with each other, - 18 - Though between S. meleagridis and 5. cambridge no definite similarity of sources has been found, further accumulation of data will be quired in the future, as it has not been long since S. cambridge was discovered. The state seems to be the same in the relationship between S. canoga and S. senftenberg var. newcastle, both of which are identical in biochemical behaviours. The details of S. kinshasa could not be surveyed because of the unavailability of the references. Anyhow, in the strains which have been isolated not so recently, it is noteworthy that each of the above pairs are not only identical in antigenic structure but alos are identi- cal or almost identical in biochemical behaviors and have been isolated from similar sources. These findings seem to suggest that one of the pair was yielded from the other as a result of antigenic variation, in nature. This might be supported by the fact that more types have been tsolated in BR, group than E> group and each of the Ea group types has its corresponding type in Ey group. From the above it might be assumed that E> group type strains might have been yielded as a result of infection of E, group organisms with the phages which were liberated from some unknown strain, though there has been no direct evidence as yet. - 19 - SUMMARY 1. S. newington, S. new-brunswick, S. selandia, S. cambridge, S. kinshase, S. canoga, S. illinois and §. thomasville of E2 group were found to be lysogenic, and the phages obtained from them were found to be capable of infecting Ey, or Bz group organisms, resulting in establishment of lysogenicity and anti- genic changes from 4,10 to 3,15 or 4,10,15 and/cr from 1,3,19 to 1,4,15,19. e. S. london, 8. give, S. anatum, S. amager, S. zanzibar, &. shangani, S. butantan, S. veile, S. meleagridis, S. elisgaheth- ville, &. simi, S. weltevreden, S. orion, S. lexington and Ss. macallen of Ey group were found to be receptive to the phages obtained from Eo group organisms and to be lysogenized, result- ing in the said antigenic changes. The antigens of the variants can be reverted to 3,10 by cultivating the variants in broth containing 15 antiserum. 3. S. chittagong was also lysogenized by the phage and its 0 be antigens were converted from 1,4,10,19 to 4,15, which were reverted to the original structure by cultivating the variant in broth containing 15 antiserum. S. senftenberg Aa', possessing 4 alone as a major 0 antigen, was lysogenized by the phage, resulting in antigenic changes from 3 to 4,15. D+ S. senftenberg-simsbury and §. niloese were lysogenized by the phage, resulting in antigenic changes from 1,43,19 to 1,3,15,19, weye which, reverted to the original structure by cultivating the variants in broth containing 15 antiserum. -~ 20 - 6. Thus, when the organisms possessing 15 factor, whether they may be type strains of Es group or phage-induced antigenic variants, were cultured in broth containing anti-15 monofactor serum, antigenic changes were induced from 4,15 to 5,10,from 3,15 to 1,4,10,19 and/or from 1,4,15,19 to 1,3,19. 7. The phages obtained from the antigenic variants possessing 715 factor were found to have the same activity as those from Eo group organisms, in infecting E,; group organisms and inducing antigenic changes in them. 8. Desoxyribonuclease, trypsin or pepsin treatment affected the phage in neither lytic activity nor capacity of inducing anti- genic changes. . | 9. The loss of lytic activity of the phages by heating or change of pH was accompanied by the loss of capacity inducing antigenic changes. 10. The observed phenomenon was compared with trangaduction and a few differences were focused and discussed. il. The bearing cf the present firdings on taxonomic problems was discussed, - 21 - mm 10. ll. l2. REFERENCES - Alexander, H.E. and Leidy, G. 1950 Transformation type speci- ficity of H. influenzae. Proc. Soc. Exptl. Biol. Med., 73, 485-187. . Alexander, H.E. and Leidy ¢. 195la Induction of heritable new type in type specific strains of 4. influenzae. Proc. Soc. Exptl. Biol. Med., 7&, 625-626. Alexander, F.E. end Leidy, G. 195lb Datermination of inherited traits of H. influenzae by desoxyribonucleic acid fractions isclated from typespecific cells. J. Exptl. Med., 93, 34b-359.- Alexander, H.E. and Redman, W. 1954 ‘Transformation of type specificity of meningococci. J. Exptl. Med., 97, 797-806. Austrian, R. 1952b Observation on the possible role of nucleic acid exchange reaction in pneumococcal capsular type transformation. A preliminary note. Bull. Johns Hopkins Hesp., 90, 170-171. Austrian, R. 1952a Bacterial transformation reactions. Bact. Revs., 16, 41-50. - Avery, O.T., MacLeod, C.M. and Mccarty, M. 19l4 Studies on the chemical nature of the substance inducing transformation of pneumococcal types. Induction of transformation by a desoxyribonucleic acid fraction isolated from pneumococcus Type &. J. Exptl. Med., 79, 137-158. Boivin, A. 191.7 Directed mutation in colon bacilli by an inducing principle of desoxyribonucleic nature: its meaning for the general biochemistry of heredity. Cold Spring Harbor Symposia Quant. Biol., 12, 7-17. Breed, R.S., Murray, E.G.D. and Hitchens, A.P. 1918 Bergey's manual of determinative bacteriology. 6th Ed., 92-53/). Brunner, D.W. and Edwards, P.R. 1918 Changes induced in the © antigens of Salmonella. J. Bact., 55, hh9. Freeman, V.J. 1951 Studies on the virulence of bacteriophage- infected strains of Corynebacterium diphtheriae. J. Bact., 61, 675-688. Freeman, V.J. and Morse, I.U. 1952 Further observations dn the change to virulence of bacteriophage-~-infected avirulent strains of Corynebacterium diphtheriae. J. Bact., 64, ho7-h1h. - 22- 13. 1h. 15. 16. 17. 18, 20. 21. ee. 23, 2h. 25, 26: Groman, N.B. 1953 Evidence for the induced nature of the change from nontoxigenicity to toxigenicity in Coryne- bacterium diphtheriae as a regult of exposure to specific bactericphage. J. Bact., 66, 18-191. Hotchkiss, R.D. 1952 The role of descxyribonucleates in bacterial transformations. Phosphorus metabolism, II, 126-139, Johns Hopkins Press, London. Ise, T. 1953a Antigenic variation in S. -enftenberg. Jap. J. Bact., &, 359.( In Japanese ) Ise, T. 1953b Changes induced in the 0 antigens of S. senften- berg and chemical analysis of specific 0 antigens. (Changes induced in the antigens of Salmcnella JW). Sapporo Med. J., lL, 395-loo. ( In Japanese ) Iseki, S. and Sakai, T. 1954a Immunogetic studies on micro- organisms. I]. Transforration of 0 antigens in Salmonella E group by the action of antiserum and bacterial autolysate. Med. and Biol., 28, 77-80.( In Japanese ) Iseki, S. and Sakai, T. 1954b Immunogenetic studies on micro- organisms. JIT. Antigen-transforming factor in bacilli of Salmonella E> group. Med. and Biol., 28, 193-196.(In Japanese) - Kauffrann, F. 1938 Salmonellaproblem. 2Z. Hyg., leo, 177-197. Kauffmann, F. 1951 Enterobacteriaceae. EF. Munksgaard, Copenhagen. Laros, B. 199 Induced modifications in the antigenic reac- _ tivity of the pneumococcus. J, Infectious Diseases, 85, 263-267. Lederberg, J. and Edwards, P.R. 1954 Serotypic recombination in Salmonella. J. Immunol., 71, 242-2ho0. Leidy, G., Hahn, E. and Alexander, E.E. 1954 In vitro produc- tion of new type of H. influenzae. J. Exptl. Med., 97; 67-82. Lweff, A. 1955 Lysogeny. Bact. Revs., 17. 269-447. McKinley, E.B. 1925 Serum antilytique obtenu par immunisation ecntre uni bacterie normale. Compt. redid. soc. biol., 93, 1050-1052. McLeod, C.M. and Krauss, M.R. 1953 control of factors dis- trict from specific transforming principle of the amount of capsular polysaccharide produced by the TIT pneumococei. J. Exptl. Med., 97, 707-771. - 23 - 27. 29. 31. 32. 53+ 3h. 35. 36. Nakagawa, T. 195la Changes induced in tue antigens of Salmonella (II). Reversicr. of antigenic ehanges. Record of the 4rd Meeting of the Hokkaido Branch of Japan Bact. Soc., 7+; 19548 Sapporo Med. J., h, 318-325. (In Japanese) Nakagawa, T. 1951b Changes induced in the antigens cof Salmonella (IJI). Studies on the mecsanism of changes induced in the "0" antigens of Salmonella. Acta Med. Hokkaidonensia, 28, 13-1; 1953b Sapporo Med. J., h, 366-391. (In Japanese) Obara, T. 1950 Changes induced in the antigens of Salmonella. Induced changes of 0 antigens from 3,15 to 4,110,206. J. Sapporo Univ. Med., 1, 66-7&. (In Japanese) - Sakai, T. and Iseki, S. 19543 Imrunozenetic studies on micrce- organisms. IV. Physicochemical properties of antigen-trans- forming factor in Salmonella E group. Med. and Biol., 29, 199-201.(In Japanese) Sonneborn, T.M. 1947 Beyond the gene. ° Am. Scientist, 37, "35-59. Stocker, B.A.D., Zinder, N.D. and Lederberg, J. 1953 Trans- duction of flagellar characters in Salmonella. J. Gen. Micwobicl., 9, 10-43. Uetake, H. and Obara, T. 1950 Changes induced in the antigens of Salmonella. Jap. J. Bact., 5, 16l:. (In Japanese) Uetake, H. et al. 1952 Changes induced in the 0 antigens of Salmonella. Studies on reversion and mechanism of changes and chemical analysis of monosaccharides compositicn of specific polysaccharides of 0 antigens. Jap. J. Bact., 7, i5. (In Japanese) Wilson, G.S. and Miles, A.A. 1948 fTopley end Wilsons Princi- ples of bacteriology and immunity, 3rd Ed., 702-7h9. Zamenhof, S., Alexander, H.E. and Leidy, G. 1953 Studies on the chemistry of the transforming activity. TI. Resistance to physical and chemical agents. J. Exptl. Med., 93, 393-397 - Zinder, N.D. and regener és J. 1952 Genetic exchange in Salmonella. J. Bact., 61, 679-699.