Studies on the E strain of Rickettsia prowazeki*

Bull. Org. mond. Sant6 1973, 49, 257-265Bull. Wld Hlth Org. Studies on the E strain of Rickettsia prowazeki* J. KAZAR, R. BREZINA, & J. URVOLGYI Cultivation of E strain of Rickettsia prowazeki in mouse lung resulted in a marked increase in virulence for guinea pigs following 12 passages (but not 6 passages) in both immunosuppressed and normal mice. Attempts at passaging E strain in normal guinea pigs were unsuccessful. Isolates ofE strain in the chicken embryo yolk sac either from testes and peritoneal washings or from the brain of immunosuppressed guinea pigs did not show enhanced virulence for guinea pigs. A decrease in pathogenicity has so far been de- scribed for two strains of Rickettsia prowazeki: strain E (Madrid), derived from the brain of a patient who died of typhus infection and subjected to 11 egg passages (4), and strain 5/6b after 440 passages on lice (17). The biological properties of strain E have been studied more intensively, and the strain was also used for vaccinating man (7, 8). Egg cultures of strain E were shown to be practically avirulent for guinea pigs and monkeys (5, 16, 21, 20) and only mildly pathogenic for man (7, 18, 12). Histological observation revealed that morphological alterations caused by strain E are negligible in guinea pigs in comparison with those caused by the virulent strain Breinl (14, 15). Treatment of guinea pigs with corti- sone dit not substantially increase their susceptibility to infection with strain E (10). Although strain E retained its virulence for cotton rats (16, 6) and lice (16, 18, 9), cultivation of this strain in lice did not change its virulence for mice and guinea pigs (3). Enhancement of virulence for guinea pigs and mice was obtained, however, when the E strain was passaged in mice by intranasal inoculation (1) and in guinea pigs by intraperitoneal inoculation (2). We have investigated the biological properties of strain E using a method similar to that of Balae- va (1, 2); cyclophosphamide (CP) or X-irradiation 'were also used in some experiments. The virulence of passaged E- strain for guinea pigs was compared with that of the initial egg cultures of E or Breinl strain of R. prowazeki. * From the WHO Regional Centre for Human Rickett- sioses, Institute of Virology, Slovak Academy of Sciences, Bratislava, Czechoslovakia. MATERIAL AND METHODS Strain E of R. prowazeki was obtained from the Gamaleja Institute of Epidemiology and Microbio- logy, Moscow, in 1965. This strain was lyophilized at the Gamaleja Institute on 12 March 1957 and labelled as series No. 281, the same as one of the series of strain E used by Balaeva. Two pools of 20% suspen- sions of this strain in sucrose-phosphate-glutamate (SPG) solution were prepared; the first, from the 5th yolk sac passage, was labelled E-O (E-original), and the second, after 4 additional passages of end- dilutions of E-O pool in the yolk sac of chick embryos in our laboratory, was designated as E-FD (E-final dilution). A pool of strain Breinl of R. pro- wazeki was prepared from the 25th yolk sac passage as a 20% suspension in SPG solution. Rickettsiae were grown and titrated in the yolk sacs of 6-7-day-old chick embryos, according to standard procedures. The ID50/ml were calculated by the method of Reed & Muench (19). Chick embryo yolk sacs were also used for isolation of rickettsiae from animal blood or viscera. When isolation in the first passage was unsuccessful, a blind passage was performed. Male guinea pigs weighing 220-260 g and mice of the Dobra Voda breed weighing 10-12 g were used throughout. Guinea pigs were inoculated intraperi- toneally with 1 ml of different dilutions in me- dium 199 of 20% suspensions of chick embryo yolk sacs, of coagulated blood, brain, and spleen of guinea pigs, and of coagulated blood and lungs of mice, all infected with strain E or Breinl of R. pro- wazeki. In all the experiments the guinea pigs were kept under observation for 3 days before inoculation 3110 257- J. KAZAR ET AL. and only those without febrile reactions, i.e., those with a temperature of less than 39.5°C, were em- ployed. In one experiment, guinea pigs were exposed to a whole-body dose of 200 r of X-rays 3 hours before or 20 hours after inoculation. Mice under ether anaesthesia were inoculated intranasally with 0.05 ml of dilutions of 20% suspen- sions of yolk sac or of lung of mice infected with R. prowazeki strains. In one experiment mice were infected intravenously with 0.5 ml of the same material. Serial passages of lung suspensions from mice treated with CP or phosphate-buffered saline (PBS) were carried out in parallel series of CP- and PBS-treated mice (6-10 in each group). Mice were killed 3 days after infection, the lungs separated, and the degree of pulmonary consolidation expressed as the percentage ofhepatized lung portions. Impression smears from two different lobes of each lung were made, stained by means of the method of Gime- nez (11), and examined microscopically for the pres- ence of rickettsiae (- no rickettsiae, (+) single occa- sional rickettsiae, + less than 10, +±+ tens, + + + hundreds, + + + + uncountable numbers of rickettsiae per microscopic field; the mean values of 10 different fields in each smear were recorded). From lungs that contained rickettsiae but in which no bacteria or chlamydia-like particles could be seen, 20% suspensions in SPG solution were prepared. The suspensions were tested on blood agar for bacterial contamination and the non-contaminated suspensions containing the highest number of rickett- siae were used for the following passages. Blind passages were carried out when no rickettsiae were seen microscopically. Cyclophosphamide, an immunosuppressive drug that enhances rickettsial multiplication in animal viscera (13), was used in some experiments; 3 doses of 100 mg per kg of body weight were given one day before and one and 5 days after inoculation of rickettsiae. In all experiments, control groups of animals (CP-treated or untreated) were inoculated with uninfected yolk sac or blood and viscera of animals. The antibody response of guinea pigs was deter- mined in sera collected 13 or 21 days after inocula- tion. Sera were tested by complement fixation (CF) reaction with soluble antigens or by microagglutina- tion (MA) reaction with stained corpuscular R. pro- wazeki antigen (Urv6lgyi, J., unpublished observa- tions). Passaging of E strain in mice and guinea pigs was carried out in separate rooms from which experi- ments with other rickettsiae were excluded 3 months before and during our work. All the investigated suspensions were stored at -60°C. RESULTS Virulence for guinea pigs of the initial egg cultures of the E and Breinl strains Table 1 shows that guinea pigs inoculated with the Breinl strain had a febrile reaction to all inocula and scrotal reactions at the dose of 103 ID50 or more. Guinea pigs receiving E strain showed only transient reactions with the highest doses of either pool. Mortality in the inoculated guinea pigs was variable. Marked differences were observed in the ability of the two strains to induce an antibody response. Titres between 32 and 512 were found in the sera of all groups infected with Breinl strain. Lower levels of CF antibodies between 16 and 128 were present only in the sera of the guinea pigs infected with 10 7 and 105 ID50 of either pool of strain E. No reactions were observed in the control group. This confirmed that virulence for guinea pigs of the initial Breinl strain differed markedly from that of the initial pools of strain E. Inability of material from mice and guinea pigs inocu- lated with the strain E to induce an antibody response to typhus rickettsiae in guinea pigs No antibody response to typhus rickettsiae was observed following inoculation with blood of mice (harvested 5 days after intravenous inoculation) or with blood, brain, and spleen of guinea pigs (har- vested 2, 4, or 7 days after intraperitoneal inocula- tion) previously infected with yolk sac suspension of strain E. On the contrary, guinea pigs inoculated with the same type of material from animals infected with yolk sac suspension of Breinl strain showed a clear response up to the 10-6 dilution. These findings indicate that strain E rickettsiae did not multiply in guinea pigs or in mice or multiplied less than was necessary to induce an antibody response. The same results were obtained after X-irradiation of guinea pigs. Increase in virulence for guinea pigs of the E strain after 12 passages in the lung ofPBS- or CP-treated mice. In repeated experiments, CP- or PBS-treated mice were inoculated with about 107 ID50 of the E-O or E- FD pool of the initial strain E. At the same time, PBS-treated mice were inoculated with the same dose 258 STRAIN E OF RICKETTSIA PROWAZEKI Table 1. Comparison of the virulence for guinea pigs of the initial E and Breinl strains of R. prowazeki Febrile reaction Scrotal phenomenon Antibody response as detected by dose (ID0) Mortality CFR c MAR d of R. prowazeki propor- Propor- b Propor- used tion tion a Days tion a Days b Propor- Titre b Propor- Titre b tion a tion a E 107 1/4 3/4 1-2 0/4 0 3/3 256 3/3 128 (E-0 pool) 106 0/4 0/4 0 0/4 0 4/4 128 4/4 64 103 0/4 0/4 0 0/4 0 0/4 0 2/4 4 101 0/4 0/4 0 0/4 0 0/4 0 0/4 0 E 107 0/4 2/4 1-2 1/4 1 4/4 128 4/4 64 (E-FD pool) 105 1/4 0/4 0 0/4 0 3/3 64 3/3 16 103 1/4 0/4 0 0/4 0 0/3 0 1/3 4 101 0/4 0/4 0 0/4 0 0/4 0 0/4 0 Breinl 107 1/4 4/4 7-8 4/4 5-6 3/3 512 3/3 256 105 1/4 4/4 6 3/4 3-4 3/3 128 3/3 128 103 0/4 4/4 4-5 1/4 2 4/4 256 4/4 64 101 0/4 2/4 2-3 0/4 0 3/4 64 3/4 32 a Ratio of positive to investigated animals. b Mean values of duration of febrile reaction and scrotal phenomenon or of the titres of complement fixation or agglutinating antibodies. c CFR = complement fixation reaction. d MAR = microagglutination reaction. of the Breinl strain and CP-treated mice with a 10-1 dilution of 20% suspension of uninfected yolk sac. Altogether 12 lung passages were carried out as described above. In addition to the microscopical observation, the degree of rickettsial multiplication was determined by titration in the yolk sacs of 20% suspensions of the lungs used for further mouse passages. Fig. 1 shows that according to both parameters the Breinl strain in PBS-treated mice multiplied at a stable level throughout the experiment. No rickett- siae were detected in the controls. A slight variation in the accumulation of E-O pool rickettsiae in the lungs of CPA-treated mice was accompanied by more marked differences in the yield of rickettsiae: the titre of E-O pool from the 12th mouse passage was approximately one log higher than that from the 1st mouse passage and 2-2.5 logs higher than those from the 3rd or 6th passage. The degree of multi- plication of E-O pool in PBS-treated mice decreased gradually during the first 6 passages, when deter- mined by microscopical observation as well as by titration. In the last four passages rickettsiae could scarcely be seen but could still be demonstrated by inoculation into the yolk sac, although their titre in the 12th mouse passage was about 4.5 logs lower than in the 1st passage. The multiplication pattern of E-FD pool was essentially similar in both CP- and PBS-treated mice, but only up to the 6th passage. In later passages in CP-treated mice, however, the appearance of chlamydia-like particles was observed and this passage branch was excluded from further experiments. More promising was the observation of increased multiplication ofE-FD pool in PBS-treated mice. More rickettsiae were detected microscopically in the 9th to 12th passages than in the first 4 pas- sages, and the titre of rickettsiae from the 12th mouse passage was approximately 4 to 6 logs higher than titres from the 1st or 6th passage The rickettsial multiplication was not accompanied by a similar degree of consolidation of the lungs in all cases. The increased degree of multiplication of E- O pool in CP-treated mice and of E-FD pool in PBS- treated mice in the last four passages was in agree- ment with the increased mortality of mice after 7-9 mouse passages. However, the mortality of mice 259 J. KAZAR ET AL. 10 9. 8' 7. 6' 5, 4' 3 2 (. (,. 1 3 9 O-'°-'--°---°--°---°----O-O-O 0- -O--_. _o.BREINL, PBS OA% A..A'-b--- -FD, PBS * Ms^'- '-*- 4A--S-A°..5 '-''o ;'; 3-0, CPAA. * U .0 A 0, *.. U ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~-0 P ........Z A -.- *-FD, CPA . ....... .. . ...... p 1 2 3 4 5 6 7 8 9 10 11 12 Fig. 1. Degree of multiplication of typhus rickettsiae during 12 passages on mouse lung, as determined by micro- scopical observation of the lung smears or by titration in chick embryo yolk sacs. Abscissa: number of mouse passages. Ordinate: upper part-the titre in log units of ID5o per ml; lower part-amount of rickettsiae in the lung smears. See the text for the meaning of the abbreviations. infected with Breinl strain increased after the 5th pas- sage, so that in later passages the lung suspensions had to be diluted and more mice were employed for each passage. To determine whether the biological properties of these pools had changed, the virulence for guinea pigs of E-O and E-FD pools of strain E from the 1st, 6th, and 12th mouse passages was compared with that of the Breinl strain from the same passages. As shown in Table 2, no substantial difference in the virulence of the Breinl strain from the 1st or 12th mouse passages was found, except for the absence of scrotal reaction in a group of guinea pigs infected with 101 ID50 of rickettsiae from the 12th passage and for the lower antibody response in all guinea pigs infected with rickettsiae from the 12th passage. Increased virulence for guinea pigs was observed with two pools of E strain from the later mouse passage (E-O pool from CP-treated mice and E-FD pool from PBS-treated mice), which caused febrile and scrotal reactions comparable with those observed with the virulent Breinl strain. No febrile, scrotal, or antibody response were observed in the control groups of guinea pigs, with the exception of fever in one guinea pig that died during the period of observation. The virulence for guinea pigs of the E-O or E-FD pools from the 6th mouse passage did not differ from that from the 1st mouse passages (Table 4). Comnparison of passaging ability of E and Breinl strains in PBS- or CP-treated guinea pigs. PBS- or CP-treated guinea pigs were infected intraperitoneally with 107 and 105 ID50 of the E (E-O pool) or Breinl strains. Parallel serial passages of brain suspensions were carried out in CP- and PBS- treated animals. Table 3 shows that the response to Breinl strain was more pronounced than that to the E strain and that for each strain the response was more pronounced in CP-treated animals than in those treated with PBS. When an attempt was made to re-isolate the rickettsiae in yolk sac with inocula- tions of 20% suspensions of brain taken 7-8 days after infection, the Breinl strain could be re-isolated regularly in all cases in the 1st or 2nd egg passage, while strain E could be recovered only from CP- 12 I 260 STRAIN E OF RICKETTSIA PROWAZEKI 261 Table 2. Increase in the virulence for guinea pigs of strain E rickettsiae after 12 passages in the lung of PBS- or CP-treated mice Febrile reaction Scrotal phenomenon Antibody response as detected bydstri anD50 Mortality CFR MAR of . prowazeki propor- Propor- Days a Propor- Day bR used tion tion a tion a Propor- Titre b Propor- Titre b tion a Tte tion a Tirb Breinl 107 1/4 4/4 6-7 4/4 7-8 3/3 256 3/3 128 1st PBS e 105 0/4 4/4 3-4 4/4 4-5 4/4 256 4/4 64 mouse 103 0/4 4/4 3 3/4 3 4/4 128 4/4 64 passage 10' 1/4 3/4 1-2 1/4 2 2/3 128 2/3 64 Breinl 107 1/4 4/4 5 3/4 5 3/3 128 3/3 64 12th PBS e 105 1/4 4/4 4-5 4/4 4 3/3 64 3/3 32 mouse 103 2/4 3/4 3-4 2/4 3 2/2 128 2/2 16 passage 101 1/4 3/4 3 0/4 0 2/3 64 3/3 8 E-0 pool 107 0/4 3/4 2 2/4 1 4/4 64 4/4 32 ist CPf 105 1/4 0/4 0 0/4 0 2/3 32 3/3 16 mouse 103 0/4 1/4 1 0/4 0 0/4 0 2/4 4 passage 101 0/4 0/4 0 0/4 0 0/4 0 0/4 0 E-0 pool 107 2/4 4/4 7-8 3/4 4-5 2/2 256 2/2 128 12th CPf 105 2/4 3/4 4 1/4 4 2/2 128 2/2 64 mouse 103 0/4 4/4 6-7 2/4 2-3 4/4 512 4/4 64 passage 101 1/4 3/4 3 0/4 0 3/3 128 3/3 64 E-FD pool 107 1/4 1/4 1 1/4 1 3/3 64 3/3 32 lst PBS e 105 1/4 0/4 0 0/4 0 1/3 8 2/3 8 mouse 103 1/4 0/4 0 0/4 0 0/3 0 1/3 4 passage 101 0/4 0/4 0 0/4 0 0/4 0 0/4 0 E-FD pool 107 2/4 4/4 6-7 4/4 5-6 2/2 128 2/2 64 12th PBS e 105 0/4 4/4 7 4/4 4-5 4/4 256 4/4 128 mouse 103 0/4 3/4 5 2/4 4 2/4 256 3/4 64 passage 101 1/4 2/4 2-3 0/4 0 2/4 32 3/4 8 a Ratio of positive to investigated animals. b Mean values of duration of febrile reaction and scrotal phenomenon or of the titres of complement fixation and agglutinating antibodies. c CFR = complement fixation reaction. d MAR = microagglutination reaction. e PBS = phosphate-buffered saline. f CP = cyclophosphamide. J. KAZAR ET AL. Table 3. Comparison of passaging ability of E and Breinl strains of R. prowazeki in PBS- and CP-treated guinea pigs dose (aDnd)of Guinea pigs 1st guinea pig passage a 2nd guinea pig passage bdose(ID5) treatedR.upowszeki with M FR SP AR Epi Ep2 M FR SP AR Epi Ep2used E 107 PBS 1/4 1/4 0/4 1/1 0/2 0/2 0/4 0/4 0/4 0/2 0/2 0/2 (E-0 pool) 105 0/4 0/4 0/4 2/2 0/2 0/2 0/4 0/4 0/4 0/2 0/2 0/2 107 CP 1/4 3/4 1/4 n.t. 1/2+ 1/2+ 1/4 1/4 0/4 1/1 0/2 0/2 105 0/4 1/4 0/4 n.t. 0/2 1/2 (+) 0/4 0/4 0/4 0/2 0/2 0/2 Breinl 107 PBS 1/4 4/4 3/4 1/1 1/2+ cont. 1/4 1/4 1/4 1/1 1/2 (+) 1/2+ 105 0/4 3/4 2/4 2/2 0/2 1/2 (+) 0/4 0/4 0/4 0/2 cont. n.t. 107 CP 2/4 4/4 4/4 n.t. 2/2++ 2/2+++ 1/4 4/4 2/4 1/1 1/2+ cont. 10 1/4 4/4 3/4 n.t. 1/2(+) 1/2+ 0/4 1/4 0/4 1/2 1/2(+) 1/2++ a Guinea pigs infected with a given dose of the initial yolk sac pools of typhus rickettsiae. b Guinea pigs inoculated with 10-' dilution of 20 % suspension of guinea pig brain taken 8 days post-infection in the 1st passage. Numbers indicate the ratio of positive to investigated animals. Key to symbols: n.t. = not tested cont. = bacterial contamination M = mortality within 3-14 days post-infection SP = scrotal phenomenon FR = febrile reaction (temperature higher than 39.5°C) AR = antibody response as determined by complement fixation reaction 21 days after infection Epi, Ep2 = detection of rickettsiae in the 1st or 2nd yolk sac passage (crosses indicate the amount of rickettsiae). CP = cyclophosphamide PBS = phosphate-buffered saline treated guinea pigs infected with a higher rickettsial dose. The same 20%. brain suspensions were inocu- lated into guinea pigs (the 2nd guinea pig passage). It was possible to passage Breinl strain in either PBS- or CP-treated guinea pigs, although strain E from PBS-treated animals disappeared in the 2nd passage; in CP-treated guinea pigs, 20% brain suspension from animals treated with CP and infected with a higher dose induced a febrile reaction and an anti- body response when tested 4 weeks after infection. In two successive yolk sac passages of brain material from the 2nd guinea pig passage of strain E, no rickettsiae were detected. It appears from these results that immunosuppression by CP increased the possibility of isolation of E strain from the brain of infected guinea pigs with the 1st passage. For this reason CP-treated guinea pigs were also used in further experiments aimed at the isolation ofE strain rickettsiae from the viscera of infected guinea pigs by the method of Balaeva and Nikolskaja (2). Attempts to isolate typhus rickettsiae from guinea pig viscera after infection with strain E. Characteriza- tion of the only isolate from the brain of CP-treated guinea pigs. CP- or PBS-treated guinea pigs were infected with 106-108 ID50 of E-O or E-FD pool ofE strain. At the same time, PBS-treated guinea pigs were infected with 107 ID50 of Breinl strain and CP-treated guinea pigs were inoculated with uninfected yolk sac suspen- sion. The observation was continued up to the 13th day after inoculation, when blood, brain, and spleen were harvested. A higher frequency of mor- tality and febrile and scrotal reactions were found in CPA-treated guinea pigs after infection with either pool of strain E. However, mortality and a febrile reaction occurred also in a control group ofuninfected CP-treated guinea pigs, so that it is difficult to say whether the febrile reaction in CP-treated guinea pigs infected with strain E rickettsiae was due to increased susceptibility of these animals to rickettsial infection or to the effect on intrinsic microbial flora. Never- theless, the onset and duration of febrile and scrotal reactions in CP-treated guinea pigs infected with strain E were similar to those in PBS-treated animals after infection with Breinl strain. PBS-treated guinea pigs responded to infection with strain E with fairly high antibody levels (titres of 64-256 for CF and 64-128 for MA); no antibody response was detected in the CP-treated guinea pigs. Yolk sac isolation of rickettsiae from PBS-treated guinea pigs was success- ful only after infection with Breinl strain: in both brain and spleen from one out of 3 guinea pigs. In no case of 45 investigated PBS-treated guinea pigs 262 STRAIN E OF RICKETTSIA PROWAZEKI infected with strain E could rickettsiae be isolated, either from the brain or the spleen. Out of 21 CP- treated guinea pigs the isolation in the yolk sac was positive in one case only, from the brain of a moribund guinea pig 11 days after infection with a lower dose (106 ID50) ofE-O pool of the E strain. The biological properties of this isolate were tested further on PBS-treated guinea pigs. No increase in the virulence for guinea pigs was found; there was no scrotal reaction following inoculation with 106, 104, and 102 ID50, and an antibody response occurred only with the dose 104 ID50 (Table 4). A transient febrile reaction lasting one day was observed only in the guinea pigs infected with 106 ID50. Characterization of typhus rickettsiae isolated from testes andperitoneal washings 3 days after infection with strain E. Isolation impossible in the second guinea pig passage. PBS- or CP-treated guinea pigs were infected intraperitoneally with 107 ID50 of the initial E-O or E-FD pools of strain E. At the same time, PBS- treated guinea pigs were infected with the same dose of the initial pool of Breinl strain and CP-treated guinea pigs were inoculated with 10-1 dilution of 20% suspension of uninfected yolk sac. 20% suspensions of whole testes in peritoneal washings (with 10 ml of SPG solution) were prepared 3 days after infection. Impression smears of peritoneum were stained by means of the method of Gimenez (11) and examined for rickettsiae. In some guinea pigs not killed on the 3rd day, the observation was continued until the 13th day. In guinea pigs infected with Breinl strain and in those infected with either pool of E strain, when treated with CP, rickettsiae were detected microscop- ically or by isolation in the 1st yolk sac passage. In PBS-treated guinea pigs infected with strain E, no rickettsiae were seen in the peritoneal smears and the isolation of rickettsiae was successful only in the 2nd yolk sac passage, although guinea pigs of these groups responded to infection with titres 32-128 of agglutinating antibodies. In the controls no rickett- siae were detected either microscopically or by isola- tion. The biological properties of one isolate in the 1st yolk sac passage from CP-treated guinea pigs infected with E-O pool were tested on PBS-treated guinea pigs. The complete absence of scrotal reaction, the presence of a transient febrile reaction lasting 2 days following infection with only the highest dose (106 ID50), and the capacity to induce antibody response at a dose of 104 ID50 indicate that the virulence of this isolate for guinea pigs was as low as the virulence of the initial yolk sac pools of E strain (Table 4). When groups of PBS-treated guinea pigs were inoculated with 10-1 dilution of 20% suspension of guinea pig testes in peritoneal washings, a febrile reaction, scrotal reaction, and antibody response and positive isolation in the yolk sac from brain taken 8 days after infection occurred only in the case of Breinl strain. In guinea pigs inoculated with material from strain E-infected guinea pigs, there was no scrotal reaction, a transient febrile reaction lasting 1-2 days occurred, and low levels of agglutinating antibodies (titres 4-8) were found, but no rickettsiae were isolated in two successive passages in theyolk sac. DISCUSSION AND CONCLUSIONS The characteristics of virulence for guinea pigs of strain E rickettsiae passaged on mice or isolated from guinea pigs are given in Table 4, which summarizes the main results of our experiments. Virulence for guinea pigs of strain E rickettsiae increased markedly following 12, but not 6, passages in the mouse lung; this finding is in agreement with the observation of Balaeva (1), who reported that the enhanced viru- lence of E strain upon passaging in the mouse lung appeared in the course of 7-10 passages. However, we did not observe any increase in virulence for guinea pigs of strain E re-isolated 3 days after inoculation from guinea pig testes and peritoneal washings. The same applied to the only isolate obtained from the brain of CP-treated guinea pigs 11 days after inoculation. This observation is not consistent with the finding of Balaeva and Nikols- kaja (2), who isolated typhus rickettsiae with an increased virulence from the brain and spleen of one of 5 strain E-infected guinea pigs (not treated with CPA), 13 days after infection. The significance of the difference of 2 days and the use of CP in our experiments remain to be determined. Balaeva (per- sonal communication, 1972) suggested that strain E rickettsiae change their biological properties upon passaging in non-susceptible animals. In view of the fact that treatment with CP increases the susceptibil- ity of guinea pigs and mice to infection with E strain, the biological properties of rickettsiae isolated from CP-treated animals should remain unchanged. We found, however, an increase in the virulence for guinea pigs of strain E rickettsiae when passaged in CP-treated mice as well as in PBS-treated mice. Moreover, unlike Balaeva and Nikolskaja (2), we were not able in any case to isolate by yolk sac 263 J. KAZAR ET AL. Table 4. Characteristics of virulence for guinea pigs of strain E rickettsiae passaged on mice or isolated from guinea pigs Egg culture of E Days of durationa Antibody responseb strain and its ID.FR ID.SP ID.CF ID.MA of as detected by titre (lDro) FR SP CFR MAR A 108.1 107 105 107 107 7-8 4-5 256 128 B 108.6 107 105 107 107 6-7 5-6 128 64 C 107.1 101 - 103 103 1-2 64 32 D 107.6 101 10' 103 105 1-2 1 128 128 E 107.1 101 - 103 103 2 64 32 F 106.6 10' - 101 103 1 16 8 Initial E-0 108-1 101 - 103 105 1-2 256 128 Pools E-FD 107.6 101 10' 103 105 1-2 1 128 64 a Mean values after infection with a highest dose tested. b Mean titres of antibodies to the highest infectious dose tested. A- E-0 pool from the 12th passage in the lungs of mice treated with cyclophosphamide (CP). B - E-FD pool from the 12th passage in the lungs of mice treated with phosphate-buffered saline(PBS). C- E-0 pool from the 6th passage in the lungs of CP-treated mice. D - E-FD pool from the 6th passage in the lungs of PBS-treated mice. E -culture of rickettsiae isolated from testes and peritoneal washings 3 days after infection of CP- treated guinea pigs with E-0 pool. F- culture of rickettsiae isolated from the brain of CP-treated guinea pigs 11 days after infection with E-0 pool. ID.FR, ID.SP, ID.CF, ID.MA-approximate quantity of infective doses in 20 % yolk sac suspension that caused febrile reaction (FR) or scrotal phenomenon (SP) and induced the formation of complement- fixing or agglutinating antibodies. CFR = complement fixation reaction. MAR = microagglutination reaction. inoculation rickettsiae from PBS-treated guinea pigs inoculated either with the initial pools of strain E or with suspensions of testes from PBS-treated guinea pigs 3 days after inoculation with the initial pools of strain E, although we used more guinea pigs and strain E of the same origin as in the experiments of Balaeva. In the meantime we have been informed about the inability of strain E to be passaged in guinea pigs (Vinson, personal communication, 1973). This obser- vation supports our results and our opinion that strain E of R. prowazeki does not change its biologi- cal properties so quickly and easily as was originally assumed by Balaeva and Nikolskaja (2). Strain E might therefore possibly be recommended for vac- cination in epidemic areas, where hundreds of viru- lent typhus strains are in circulation. On the other hand, the increase in the virulence of strain E upon passaging in mice indicates that this potentially virulent strain should be used for preventive vaccina- tion outside epidemic areas only after very careful consideration. ACKNOWLEDGEMENT [TheltechnicalIassistance of Mrs E. NemcovA is greatly appreciated. RItSUMI LtTUDES SUR LA SOUCHE E DE 'RICKETTSIA PRO WAZEKI La virulence pour le cobaye - mesuree par la reaction tres differente de celle de deux isolats de la souche E du febrile, la reaction scrotale et la reponse immunitaire - meme parasite ayant subi un petit nombre (souche E-O) de la souche Breinl de Rickettsia prowazeki est apparue ou un nombre plus eleve (souche E-FD) de passages 264 STRAIN E OF RICKETTSIA PROWAZEKI 265 sur l'ceuf. La souche Breinl a conserve sa virulence apr6s passages sur le poumon de souris normales, c'est-a-dire n'ayant pas subi un traitement immunosuppresseur par le cyclophosphamide (CP). La culture de la souche E-O dans le poumon de souris trait&es par CP ou de la souche E-FD dans le poumon de souris normales a entrain6 une forte augmentation de leur virulence apr6s 12, mais non apres 6 passages. Des essais de passage de la souche E sur des cobayes normaux sont restes infructueux. Alors que la souche Breinl pouvait etre reisolee regulierement sur auf a partir des testicules ou des liquides peritoneaux recueillis 3 jours apres l'infection ou a partir de suspensions de cerveaux prepar6es 8-13 jours apres l'infection de cobayes normaux ou traites par immunosuppression, la souche E n'a pu 8tre retrouvee qu'occasionnellement chez des cobayes traites. Un isolat de la souche E obtenu a partir des testicules et du peritoine 3 jours apres l'infection et l'unique isolat de la souche E obtenu 11 jours apres l'infection a partir du cerveau d'un animal moribond, chez des cobayes traites, n'ont fait preuve d'aucune aug- mentation de la virulence pour le cobaye. Chez aucun de 45 cobayes normaux infectes par l'un ou l'autre type de souche E, on n'a reussi a reisoler des rickettsies A partir du cerveau ou de la rate pr6leves 13 jours apres l'infection. REFERENCES 1. BALAEVA, N. M. Increasing virulence of the Rickett- siae prowazeki strain E by its passage in the lungs of albino mice (in Russian). Vest. Akad. Med. Nauk SSSR, 10: 51 (1969). 2. BALAEVA, N. M. & NIKOLSKAJA, V. N. Increasing virulence of the strain E Rickettsiaprowazeki with its passage through guinea pigs (in Russian). Vest. Akad. Med. Nauk SSSR, No. 2: 17 (1970). 3. BALAEVA, N. M. & NIKOLSKAJA, V. N. Enhanced virulence of the vaccine strain E of Rickettsia prowazeki on passaging in white mice and guinea pigs. Acta virol., 16: 80 (1972). 4. CLAVERO, C. G. & PEREZ-GALLARDO, P. F. Rev. Sanid. Hig. Pibl. (Madr.), 17: 1 (1943). 5. CLAVERO, C. G. & PEREZ-GALLARDO, P. F. Experi- mental study of a non-pathogenic immunizing strain of Rickettsia prowazeki. Trop. Dis. Bull., 41: 24 (1944). 6. Fox, J. P. The relative infectibility of laboratory animals and chick embryos with rickettsiae of murine or of epidemic typhus. Amer. J. Hyg., 49: 313 (1949). 7. Fox, J. P. A review of experience with an avirulent strain of R. prowazeki (strain E) as a living agent for immunizing man against epidemic typhus. Amer. J. publ. Hlth., 45: 1036 (1955). 8. Fox, J. P. ET AL. Immunization of man against epidemic typhus by infection with avirulent Rickettsia prowazeki (strain E). V. A brief review and observa- tions during a 31/2-year period as to the occurrence of typhus among vaccinated and control populations in the Peruvian Andes. Arch. Inst. Pasteur Tunis, 36: 449 (1959). 9. GAON, J. Attempts at virulence modification of " E" strain R. prowazeki by passage through body lice. Folia med. Facultatis Medic. Univ. Saraeviensis: 1, 79 (1966). 10. GENIG, V. A. The influence of cortisone on the course of rickettsial infection in guinea pigs by " Breinl " and "E" strains (in Russian). Vop. Virus., 4: 85 (1962). 1 1. GimENEz, D. F. Staining rickettsiae in yolk-sac cultures. Stain Technol., 39: 135 (1964). 12. GouNEvIcH, E. M. & YABLONSKAYA, V. A. Immu- nization of humans with live vaccine from strain E of Rickettsia prowazeki (in Russian). Vop. Infekt., Patol. Immunol.: 211 (1963). 13. KAZAR, J. Er AL. Study on the effect of cyclophospha- mide on experimental rickettsial infection. Acta virol., 15: 499 (1971). 14. KoKoRIN, I. N. Histopathology of vaccinal process and morphological assessment of immunity in guinea pigs infected with the strain E of Rickettsiaprowazeki (in Russian). Vop. Virus., 4: 272 (1959). 15. KOLESNIKOVA, N. A. ET AL. Morphological altera- tions of vascular wall in guinea pigs after inoculation of the strain " E " of Rickettsia prowazeki (in Rus- sian). Vop. Virus., 7: 222 (1962). 16. PEREZ-GALLARDO, F. P. & Fox, J. P. Infection and immunization of laboratory animals with Rickettsia prowazeki of reduced pathogenicity, strain E. Amer. J. Hyg., 48: 6 (1948). 17. PscHEcHNov, R. A. On variation of typhus rickett- siae (in Russian). In: Live vaccines, Moskva, Medgiz, 1956, p. 249. 18. PscHENIcHNov, V. A. ET AL. On the biological characterization of the vaccine strain E of Rickettsia prowazeki (observations on the immunization of humans with live epidemic typhus vaccine) (in Rus- sian). Vop. Virus., 4: 698 (1959). 19. REED, L. J. & MUENCH, H. A simple method of estimating fifty per cent endpoints. Amer. J. Hyg., 27: 493 (1938). 20. URAKOV, N. N. Susceptibility of monkey Macacus rhesus to Rickettsia prowazeki (in Russian). Vop. Virus., 7: 225 (1962). 21. ZDRODOVSKY, P. F. ET AL. Characterization of a strain " E " of Rickettsia prowazeki and its patho- genic properties (in Russian). Vop. Virus., 3: 136 (1958).