Thiaixtoeettifgthatthe thesis entitled A Study d‘ the Antigenic“: of Strain: of Vibrio cholera presented by Jon-Jo mung. M. D. ha been accepted towards fulfillment . of the requirements for M_.-__ 5; degree LB «Em/057 A STUDY 0? THE ANTIGENICITY OF SIRAINS 0F VIBRIO CHOIERAE BY Jen - Jo Huang, M.D. A Thesis Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Bacteriology and Public Health Year 1946 7412.? H 376’ I. II. III. V. Contents Introduction Experimental Methods and Results a. Characterization Studies of Strains of Vibrio Cholerae b. Animal Passage and Virulence Titration of Strains of Vibrio Cholerae c. Preparation of Lyophile Cultures and vaccines d. Comparison of the Protective'Values of Vaccines 1. vaccines of Various Strains Tested Against Homologous and Reference Strains of Organisms. ii. Vaccine of Reference Strain Tested Against All Strains of Organisms. e. Toxicity of the'Vaccines. Discussion Summary Literature 12-90255 ,A Study of the Antigenicity of Strains of Vibrio cholerae Jen - Jo Huang, M.D. From the Biologics Control Division of the National Institute of Health, China* Introduction Public health workers in those countries where cholera is prevalent are convinced of the value of vaccination as a prophylactic measure against the disease. The significance of most statistical investigations concerning cholera vaccination has been Open to question because of the unequal numbers of persons in the vaccinated and control groups, the unequal time intervals between vaccination and the outbreak of epidemics, and unequal risks of exposure to cholera of the persons within the two groups. The investigations of Greenwood and Yule (1), whose subjects were members of the Sanitary Corps and other combatant groups, appear to be above these criticisms. These authors offer support for vaccination in the form of statistical data. * The work reported in this paper was carried out in the Biologics Laboratory of the Michigan Department of Health, Lansing, while the author was on study leave as a Research Fellow under the cultural COOperation program of the United States Department of State. 2. Additional support for anti-cholera vaccination is offered by immunological studies concerned with the bactericidal and agglutinative properties, and mouse protective antibodies present in the sera of persons who have been vaccinated against the disease (2, 3, 4, 5). Burrows and his co-workers (6) have pointed out that Vibrio cholerae, unlike the typhoid bacillus, remains within the lumen of the bowel and exterior to the tissues of the body in the human disease; and that artificial immunization against cholera has not been as efficacious as artificial immunization against typhoid. These investigators differentiated Vibrio cholerae into four immunolOgic types A, AB, AC, ABC according to the distribution.of major somatic antigens. They are presently engaged in studies aiming toward the elucidation of the functional mechanisms of immunity in Asiatic cholera. In general, manufacturers of cholera vaccines endeavor to employ virulent strains of Vibrio cholerae; however, little is known of the relationship between the virulence and antigenicity of this organism. Yu (7) has been the only one to use mice for determining the virulence of saline suspensions of smooth strains of Vibrio cholerae. This author demonstrated that vaccines prepared with highly virulent strains induced better protection than those prepared from strains of low virulence. 3. Tang et al (8) studied serologically 83 strains of Vibrio cholerae, isolated from the 1942 Kuming epidemic, but made no observations on either virulence or antigenicity. The National Institute of Health of China isolated 12 strains of Vibrio cholerae from patients dying with Asiatic cholera during the Chungking epidemic of 19:5. The present study was undertaken to compare the virulence and antigenicity of these newly isolated strains with strains currently in use for the production of cholera vaccines in the United States and in China, and to determine whether or not a relationship exists between the virulence and antigenicity of the organism. It was also hoped that some recommendations might be forthcoming in re- gards to the selection of strains of Vibrio cholerae for use in ' cholera vaccine production. Exnerimental Methods and Results A total of 16 strains of Vibrio cholerae were included in this study. Of these, 2 were obtained from the National Insti- tute of Health of the United States Public Health Service, 10 were recovered from 12 strains isolated during the 19¥5 Chungking epidemic, while 4 others were chosen at random from 8 strains 7 currently in use in China for cholera vaccine production. The strains bf Vibrio cholerae were received in either the dried or semi-solid state. These were transferred to veal A. infusion broth at pH 8.4 and incubated for 18 hours at 37°C., after which transfers were made onto veal infusion agar of similar pH. No organism was propagated on artificial medium beyond the third subculture before use in either lyophiling, animal injection, or vaccine preparation. Characterization Studies of the Strains of Vibrio cholerae The various strains of Vibrio cholerae were carefully studied as to their morphological, cultural, biochemical and serological characteristics. These included cellular and colonial morphology, staining properties, motility, Heiberg's sugar reactions, the cholera-red reaction, the Voges-Proskauer test, the Grieg hemolytic test and immunological typing with Specific somatic antisera. Monospecific.A, B, and C antisera were furnished by Dr. William.Burrows of the University of Chicago. Typing was carried out according to the technique recommended by Burrows and associates (6) while the biochemical tests were performed according to the methods outlined in Diagnostic Procedures and Reagents (9). The results of the characterization studies are summarized in Table I. All the strains studied showed typical characteris- tics of true, smooth Vibrio cholerae, the only atypical finding was the slight hemolytic activity exhibited by strains 5122 and 51A9. Serologically, all 1+ immunologic types A, AB, AC and ABC 5. e m< I + H = c + 2 a w: 2 a I . + H 2 2 + B 3 m: a ma I + mw . a + a a men wswwd m4 I + H = a + : deHpm . oqaooeb emcaaao mm I om< I + H = = +. = = mmmm I om< I + H a = + a : mmom I 4. I + H = = + = = deem I ma I + H = a + s g NOHm I m< I + H 2 a + = = Hmam I m< I + H a = + = = ammm I mq I + H = a + = = wmom I 4 I + H a = + = : jon I m4 .Hm + H = = + = z ijm I d .Hm + H = = + = quHPm aneeHme waflxmsdno mmHm wnwcH u< I + H = = + = = mammmH: . mmflmb asewo me I w I + H ceaeameausaae + neooam .mHz aamHz camp mean, we 9 I ooh omzp we a h a ha0Hoo we on: , caaeeaeh . as H e m > neaeaeso mcceacm a seam e.HHpoa eeaeeeeaaa c om ceases mmneHogo oHnoH> Ho mafiwnpm Ho mafipmflampowawno H capes 6. were represented. The majority, 10 out of 16 belonged to type AB. The strains isolated from.the 19u5 Chungking epidemic included 3 types A, AB, and ABC. Animal Passage and'Virulence Titration of Strains of Vibrio cholerae After the various characteristics of the strains had been determined, attempts were made to increase the virulence of the organisms through serial passages in mice. The virulence of the strains was titrated by intraperitoneal injections of mucinized suspensions of the organisms into mice. The procedure used for handling the strains for animal injection was as follows: I 1. Equal number of male and female white Swiss mice weighing from.10-2O grams were employed. The mice were bred on the Laboratory's own farm. 2. Only the third subcultures of 5-6 hours of age were used for animal injection. The parent state was either a fresh recovery from the dried state or a fresh isolate from.the heartis blood of injected and dead mice. The complete sequence of cultures was veal infusion broth '18 hours, veal infusion agar 2% hours and veal infusion agar 5-6 hours. 3. The organisms were suSpended in warm buffered physiological saline solutions from veal infusion agar. The suSpension 5. 7. of the organisms was facilitated mechanically by using sterile cotton swabs. This suspension was standardized to an arbitrary reading of 50 on a Coleman Universal Spectr0photometer. Serial ten-fold dilutions of this standardized suspension were prepared in saline and the test doses were made by ten-fold dilutions in 5% mucin from the next lower saline dilution. Groups of six mice each were given 0.5 ml amounts of the test doses ranging from.mucin dilutions 10'5 to 10’8. Injections were begun with the highest dilution in each series. All injections were finished within an hour from the time the suspensions had first been prepared. Duplicate samples of 1 ml portions of the 10"7 saline dilutions were pour-plated in veal infusion agar for colony counting. - The 5% muoin.solution was prepared according to the method recommended by the Biologic Products Division of the United States Army Medical School (10). The mice were observed for deaths for a 72 hour period after the injection of the mucinized suspensions of the organisms. The Reed and muench method (11) was used in determining the 505 end point (LD5O) of the injected mice, care being taken to include both zero or 8. near zero and 100% or near 100% end points in each experiment. This method minimizes the intrusion of statistical weighing into results by test animals failing to give an average response. The calcula- tion is based on the assumption that if one animal survives a low dilution it would certainly have survived at higher dilutions; and if it dies in a higher dilution, it would die in lower dilutions. The method is especially valuable in comparative studies, such as comparing the results of one experiment with another or comparing the data of one laboratory with that of another. A.sample protocol of an experiment with NIH35A5 strain is given to illustrate the method of calculation. Protocol of Virulence Titration (strain NIH35A3) ' Diluq o. of No. _ Percent tion organisms of' Deaths Sur- ,,Aocumu;ated ‘mor- LD50 in 0.5ml dose mice vivals Deaths Sur- tality (pour-plate) vivals 10-? 8,200 6 ll- 2 ll 2 85 10’6 820 6 4 2 87 u 64 10'7 82 6 2 n 3 8 27 =l/2,39o,ooo 10’ 8 6 1 5 1 13 7 i=543 ggggn- The 50% end point dilution lies between 10‘“6 and 10'7, and is obtained by use of the fraction: g50% — 21% (mortality next under50%) 64% (mortality next over 50%)- 27%7(mortality next under 50%) which is equal to 23/37. The product of this fraction and 9. the logarithm of the dilution factor (log 10 = 1) sub- tracted from the logarithm of the denominator of the dilution next under 50% gives the logarithm of the de- nominator of the 50% end point dilution. L08 denominator of dilution next under 50% = 7.0000 23/37 I 108,10 = .6216 Log denominator of 50% end point dilution = 6.3784 6.378”. is the logarithm of 2,390,000. The 50% end point then is l/2,390,000 dilution. The 1 ml samples of the 10"7 saline solution pour-plated averaged a colony count of 164. Since 0.5 ml amounts were used for animal in- jections, the l/2,390,000 dilution contained 82 x 10"7 2,390,000 = 343 organisms. The logarithm.of the denominator of 50% end point dilutions (log. titre) and the correSponding number of organisms in the LD50 doses for the various virulence titration experiments are summa- rized in Table II. Table III segregates the 16 strains of Vibrio cholerae according to their relative virulence into three broad groups. The virulence of an organism as titrated in experimental animals varies with the mean reactivity or sensitivity of the animal population. Thus even with organisms of established virulence, duplicate experiments do not necessarily yield identical 10. results. The virulence of the 16 strains of Vibrio cholerae was compared in groups of either two or four. When the LD5O's ex- pressed either in terms of dilution or in number of organisms oscillated in both directions during excessive experiments, the virulence of the organisms was taken as fixed if the relation of the LD50'S of the various strains remained more or less constant. No attempt was made to determine the precise order of virulence among the various strains of each group. ‘Virulence of Strains of Vibrio cholerae Table II 11. Strain of Log. titre No. of organi No. of organisms ‘Vibrio (LD5O) in 0.5 ml 10' LDfindeSO cholerae dilution (plate (—STnéIe count), determination Average 1mm 6.6600 80 175 7.1mm 100 no 11,, 6.8150 90 l20 Loooo 112 1_1_2_ NIBB5A3 6.5701 82 315 7.0000 ' 100 ' 100 212 6.6000 110 279 6.8065 ' 80 125 6. A1 0 160 5122 7.3.000 100 363 6.9000 80 100 511,9 6.2500 76 A22 193 7.5200 90 to 7.0000 $15 115 5038 5.11.710 85 2,8 5019 5.2000 90 5.6 3 5.0820 78 6,500 510A 7.0000 “(5 75 137 6.3200 100 200 6. 00 3+ 136 826 5 37 5. 5 1.350 6.0000 82 8:0 866 5 8“ 5.6300 g? 1.780 6.2000 . 0 5. 1760 - 82 6 . 133 8378 5102 E' 000 90 9.000 £700 A 10.01010 92 .2 00 10 5°“ 5.9500 90 111 £0000 120 120 5053 £0000 8 50 106 000 11 9+ A 00 100 116 5353 b' 088 g I05 627 00 0 131 12. Table II (continued) Strain of Log. titre N0. of organigms No. of organisms Vibrio (LD5O) in 0.5 ml 10' LDRO dose cholerae' dilution (plate Single ‘ count) determination Average 35 6?0000 90 900 A32 6.5300 97 ' 287 6.u200 8;, 4308 ‘ 36’ 6Jhoo ldI 161 . 121 7.0000 92 92 7.0000 110 110 pg: #5 6.1800 580 533 H07 6.3000 gt 372 6,5000 0 . 316 H8?" 6.6800 115 219 141 6.8450 80 11h _7.0000 90 90 Table III Strains of Vibrio cholerae Grouped According to Their Relative Virulence LD5O dose expressed in No. of Organisms 80-u00 Moo-2000 2000-10,000 Ninth a 5137 5038 NIH35A5 528A 5102 5122 35 5am , 15 510A 50th 5053 5353 36 H8 13. Preparation of_Lyophile Cultures and Vaccines When the virulence of the organisms reached a "steady" level, a generous supply of cultures of each strain dried from the frozen state in sterile normal horse serum was made and maintained. A 2n-h0ur growth of the second subculture of a fresh isolate from the heart's blood of injected and dead mice was used for lyophiling. The normal horse serum employed was recovered from the dried state by adding the appr0priate amount of physiological saline. Sterili- zation was effected by filtration through a Seitz filter. After thoroughly washing the growth off from the veal infusion agar, 0.2 ml amounts of the serum suspension were carefully delivered by capillary pipettes into each of 10-20‘ly0phile tubes. Freezing was accomplished either in an alcohol-dry-ice bath or in a deep freezer. The frozen cultures were dried for 48-72 hours and sealed under vacuum. Vaccines were also prepared separately for each strain after the completion of its virulence determinations. The parent state used for. vaccine production was an 18-hour growth in veal infusion broth of organisms either freshly recovered from the dried state or freshly isolated from the heart's blood of injected and dead mice. The planting was made by the loop method. Physiological saline containing 0.5 per cent phenol was used to wash off the 2n-hour growth from veal infusion agar. The washing off of the organisms was facilitated mechanically by using sterile cotton swabs. The phenolized saline suSpensions 14. were immediately stored at 5°C. Testing for non-viability of the Vibrio cholerae was generally begun on the third day. The organisms were found to have been killed by the Nth to the 7th days. The phenol-killed vaccines were standardized before use to contain 1,600 million organisms per ml. The standardization was carried out as follows: When the organisms were found to have been killed, small lots of different dilutions were prepared from the stock vaccines. The spectr0photometric reading of each of these diluted vaccines was recorded and the number of organisms per ml of the corresponding dilutions was determined by direct count. A white blood cell counting pipette and PetrofféHauser counting chamber were used for the latter purpose. Figure 1 gives a composite curve with the logarithm of the spectrophotometric readings plotted against the number of organisms in millions per ml. In practice, vaccines were first standardized to read "40" on the spectr0photometer which re- presented a count of 3,400 million organisms per ml and then diluted 1:2. The direct count method here employed with the vaccines gave higher values than the pour-plate method utilized in the case of viable organisms. The latter yielded an average of 180 organisms per ml in the 10"7 saline dilutions. The original undiluted suspension therefore contained 1,800 million organisms per ml. From the cali- bration curve a suSpension with a SpectrOphometric reading of 50 corresponds to a count of 2,400 million organisms per ml. The ratio 15. between the direct count and pour-plate count is roughly 3:2. Comparison of the Antigenicity of Various Strains of Vibrio cholerae i. Vaccines of various Strains Tested Against Homologous and Reference Strains. Three groups of 50 mice each weighing from 10-20 grams were set aside for each experiment. One of these groups was maintained as controls. The other two groups were vaccinated with vaccines prepared from.two different strains of Vibrio cholerae. Each group was given two intraperitoneal injections of 0.25 and 0.5 ml re- Spectively of a given vaccine with an interval of one week between the two injections. Two weeks after the last injection, the vaccinated mice in groups of six each were given challenge doses of mucin dilu- tions 10'1, 10'2, 10'"3 and 10-4 of the homologous and the NIHUrl strains of Vibrio cholerae. The unvaccinated controls were given test doses of the same strains ranging from.10"5 to 10-8. The HIHMI strain, being the most virulent strain, was chosen as the reference strain. The 50% end point dilution was calculated for both the control groups and the vaccinated groups with reference to both the homologous strain and the reference strain in the same way as in virulence titration experiments. The 50% end point dilution of the control group receiving test doses of one strain of Vibrio cholerae divided by the 50% end point dilution of the vaccinated group chal- lenged with the same test strain of organism gives the protective value in.M5L.D. of the given vaccine against the test strain of 16. Figure l 1 com / +82 tonne c m / L OOON .1 e w m. ./ 83 m .m / i m o / m m // +88 mm / a m / / noose. e t / pl / s w / :89. m t Ia/X e m r . w m / 4 come t / m t e m :88 mm 0 d x w/ a 288 n m / g 0 m s / Boom a S . / r n / L . .m m , / some a m x / C R n . / #000... / fl . / grooms .onom ogpflmwg 3 mefiemmn fianceepooaonpooam / / w. a _ . (Ix mg H Li! a p . a) m m .06 m w w m Millions of organi SIDS per ml. l7. organism. When the LD50 of the vaccinated group lies beyond the lowest dilution 10'1, the protection is expressed as > °°ntr°l 50% end Pt“ “1‘1“”. If it lies beyond the highest 10 dilution 10'”, the protection is given as<: control 53§033d9t°d11uti°E_ s of both vaccinated and control groups lie beyond When the LD50' the ranges under test, the protection values is of course unde- terminable. The antigenicity of each given vaccine was tested twice on different days. The protective values of vaccines pre- pared from.various strains of Vibrio cholerae are given in Table IV. ii. 'vaccine of Reference Strain Tested Against All Strains In addition to challenging mice vaccinated with vaccines prepared from the various strains with organisms of the homologous and reference strains, other groups of mdce were given vaccine of the reference strain and challenged separately with organisms of all the various strains of Vibrio cholerae. Normal unvaccinated con- trols received appropriate LD test doses of the same strains. The 50 results of these experiments are summarized in Tables V'a and b. 18. .000d000 0s» peaaaopsfl ow oaowmmoaaa pH memes 09003000000 000000030 nomspon hoomamnowflo 0003 make 0000 v 0000.0 0000.0 A 0 0000.0 A 884% .. 00 80.00 0000.0 0000.0 0 000.00 v 88.0v 0000.0 00. 000V 00 8on0 v 0000.0 0000.0 A 0 00001100 0.80.0 v .. 00 000 0 0000.0 080.0 000 000 A 0000.0 0000.0v (I00. 000.1 00 00040 v 800 .0 0000.0 A 00040 v 0000 .0 800.0 A .. 00 000.00 0000.0 0000.0 00.00 0000.0 0000.0 ms. 00..." 0H 00040 v 0000.0 0000.0 A 0000 v 0000.0 0000.0 A .. (00 000.00 0000.0 800.0 000.0 0000.0 0000.0 00 00..m 00 II 000 00 v 0000.0 0000.0A 000 «0 v 000:0..0 0000.0 A .. I: 100 000.0 0000.0 800.0 000.000 A 0000.0 008.0v 000. 0000hl 00 00010.» 0000.0 080.0 A 000 0 v 0000.0 0000.0A .. w 00 80 .0 0000 .0 0000.0. 000 .00 0000 .0 800.01 000 0000*! 00 000 . 0 0000 . 0 0000 .0 80.400 88.0 0000 .0 .. 0 000.0 0000.0 080.0 000.00 0000.0 0000.0 0. 0000 0 I 80 0000 . 0 0000 .0 00000 0 0000 .0 0000 . 0 .. 0 000. v 0000 . 0 0000.0 A 0 000mm v 0000.0 v 00 l 0000 0 000 0000 .0 0000.0 000 «000 A 0000.0 88.0 v .. 0 000 0000.0 0000.0 000.0 0000.0 0000“ 00. 0000 0 (000 «0 0000.0 0000.0 000000 A 0000.0 88.0 v .. 0 000 0000.0 000.0 000.00 A 0000.0 0000Hu0fv 00. 00W 01: 000 0000.0 0000.0 0000000040 0000.0 080.0 In I! 0 000 080 .0 0000.0 08.000 .0A 0000.0 0000 HIV 0 0000 0) II 000 v 800.0 0000.0 A 8040 0000.0 0000.0 .. J: 0 000 0000.0 800.0 000.0 0000.0 0000 JAI 00. 0000 0 II 000 «00 080.0 0000.0 08 0000.0 0000.0 0000.0 .. (0 000.0 0000.0 080.0 000.00I01 E00010 0000 H 00. 0000 _ 0 I. 08.00 0000.0 0000.0 08480 0000.0 0000.0 ( .. 0 000.0 0000.0 0000.0 000.00 0000.0 0000.0 0 0000 ms 000 «0 00 00 .0 0000 .0 80400 0000 .0 008.0 .. 0 l: 80.00 0000.0 0000.0 80.80 0000.0 080.0 00 0000002 .0! 000 «00 0000.0 0000.0 11000.00 0000.0 0000.0 .. 0 ( 000.00 0000.0 800.0 000.00 0000.0 800.0 00. 000000 0 covomponm 0Honpnoo cmpwsHoowb umpooponm, 0Hoepaoo wopwsHooe> .mno .0HHMWI 0.00 .0 .00 0.0000 020 0.500 000 0.0 .0 .2 0.2.00 000 0.0000 00.0 0o 0000800 .00 .oz nwmhpm concaommm new; vuMWQHHmmc nwwhpm moo Hoaom APHS powdeanmo cabs HSHHwMI mafiwppm oooosowcm 0S0 mnemoaoaom on» p0q00w< 0S0000m 050000> Ho 00c0000>.Ho mosawb 0>Hpoepoam >H oHan ) . If" “T' N - ‘1 ‘. fill ‘ , l . I. i ' '.. ' ' I ,- u on, '-v f‘v—vw .. v.0 Lav .. I . 19. A; Table Va 12'. ‘ I ' -. Protective Values of Vaccine of Reference Strain Against the ' if? Various Strains of Vibrio cholerae - “I.“ S . -— __ __.__. —~ 0;..— _ A w .__. ware—“- “4...“...— Mini; Prgtection in M.L.D.'_s a ainst challenges of . _ __ 1”,“,m "rum .1. T." Expts. NIH41 NIHSSAS 5122 5149 50:38 5104 5137 5284 5102 5055 5:352“ {35 . “1.8.8” 45 458__ —- :—_———__ ‘—_.' .___.,___ —— I -1H""' t J I L 2.8Jf320l 1001000 — - - 1 - - -_ ' - ' - 1 - (t - ‘ -....T.-.....I'.._-.':......0._( : " ‘ -0 ., "Norm“ ~ g _2_ 11,590“ 51,600 - - ' - - I - - I - ’ - 12;. 1 _-;_____,,, ,:.____I‘ - 211......) 1677 42,200 - 11,000 382,000 9101000 ? >lO0.000 100.000) - 1 1:" _-___ J_-— __-__ - -,_,_,.__;__.__.l II '- 7 .. I - __ ? 10,000 eagles - 01,000 - - 10 000 '- (1,000 22361400 - :_ :11...“-....-.._ w l i.- g .. .. -‘ .. .. .. , - - - - I - - .? ? >376LOQO‘ _? __ Flig— __ h 0.1. v—‘aqbwr'flu - w..- menu's-awes- - M“ ‘_ jg: __. 1.. —— v —-__ L114 .1.- 1....19..022L;.1J1-1;-.- 11:... '51. 200 __ - - 1 l - $1.1,QQQJL.-_:._-1L€ELL§3_JS.5£20L_.;«05.20193..-._( Table Vb Protection of Reference Vaccine against Various Strains Impt No LogLTiter Vaccinated Control§_ NIH41 2.5700 7.0000 2 2.2550 6.3170 16 2.3748 7.0000 1 NIH35A3 1 1.5000 6.5000 2 2.0000 6.5000 19 3.0000 7.0000 5122 16 2.7840 6.8340 1;: 8.0000 7.0000 5149 16 1.4188 7.0000 ' 17 1.0000 6.8340 . 5038 16 <.l.0000 5.0000 5104 16 > 4.0000 ) 8.0000 17 >4.0000 7.0000 19 >4.0000 7.6250 5137 16 < 1.0000 6.0000 5284 f 16 < 1.0000 6:9000 5102 17 <‘1.0000 5.0000 5044 17 > 4.0000 > 8.0000 18 > 4.0000 7.0000 5053 17 > 4.0000 7.0000 19 .>4.0000 7.1760 5553 17 < 1.0000 6.4288 35 18 > 4.0000 >8.0000 19 > 4.0000 7. 5000 36 18 > 4.0000 > 8.0000 - l9 .1 >400000 7.1760 45'_ 18 0(100000 6.5752 48 18 > 4.0000 > 8.0000 19 >4.0000 7. 3220 21. Strains NIHMl and NIH35A3 are known to possess good antigenicity (5,12,13,14) and the results here obtained confirm such findings. Strains 5122 and 5149 isolated from.the Chungking epidemic showed good protection against both the homologous and reference strains. 0f the strains that are currently in.use for vaccine production in China and strains 5053 and 5353 isolated from the Chungking epidemic, the great discrepancy between duplicate experiments precluded possible conclusions. Strains with low virulence, 5038 and 5102 yield vaccines of low antigenicity. ‘Vaccines from strains of high virulence generally produced good homologous protection but did not always accord good protection against the reference strain. The reference strain NIHHl showed good cross protection against strains 111113513, 5122, 51119, 5038, 5137, 5281+, 5102, 5353 and 115. It did not protect well against strains 5104, 5044, 5053, 35, 36 and #8. Strains 510% and 504% belong to type A, strain 5053 belongs to type ABC, but strains 35, 36 and 48 are of the same type AB as the reference strain NIHN-l. Among the strains well cross protected by the reference vaccine, NIH35A3 is type AC, 5122 is type A, 5353 type ABC, while the others are type AB. The virulence of the various strains does not seem to be related to the degree of protection accorded to them.by the reference vaccine. 22. The Toxicity of the vaccines Both the vaccinated and the control groups of mice were weighed at 0, l and 3 weeks, i.e., before each vaccination and prior to the final challenge. The number of deaths in these groups occurring during these intervals were also recorded. These were undertaken with the purpose in mind of taking the number of deaths and the average gain or loss in weight of the vaccinated groups as con- trasted with the controls as measures of the toxicity of the various vaccines. The National Institute of Health, United States Public Health Service, recommended using the average gain in weight of mice vaccinated with varying amounts of the same pertussis vaccine as an indication of the toxicity of the vaccine (15). The same principle is here applied to identical amounts of different vaccines prepared from.various strains of Vibrio cholerae. The relevant data are presented in Table VI. The toxicity of the various vaccines is difficult to analyze fran the data presented. The wide variations among the different control groups speak for non-homogenicity of the existing conditions. The initial average weight of the groups of mice was approximately the same. The temperature of the mouse room fluctuated'between 6OO - 80°F. owing to‘a defect in the thermo-regulating system. Ample feed of the same kind was provided for all groups but it had been noticed that some groups of mice ate better than others. The amount 23. of water supplied might have constituted another uncontrolled factor. Eyen the comparison between the vaccinated groups with reference to the control group in a given experiment is not without difficulty. When a greater number of deaths concurs with a lesser gain in weight, or when a fewer number of deaths concurs with a greater gain in weight in one vaccinated group as compared to another, the first vaccine may be reasonably inferred to be either more or less toxic than the second. But when a greater number of mice die leaving a resultant higher increase in weight, or when fewer animals die giving a concurrent lesser gain in weight, the inference is not at all clear. Vaccines of strains 5038, 510+, 35 and 36 appear to be highly toxic. That these strains happened to fall into two groups where both the tested vaccines yielded high toxicity might have aroused the suSpicion of a chance factor. The duplicate checks in both instances, however, did not uphold such an assumption. It is of interest to note that strain 5038 is a comparatively nonpvirulent Organism. 'Vaccine prepared from the other nonavirulent strain 5102 was tested side by side with the vaccine of a virulent strain 5044. There was no appreciable difference in the toxicity of the two vaccine$,both being rather mild in character. It would thus appear that the toxicity of a vaccine is not related to the virulence of the organism. nm 2 W N H 133 are 9m swan % n a mom m6 fin a. m, o OOH , oOH m.m m.m, Honesoo H, H o c.00H .m ~.H sewn m m s, IIIiIIwwem wwWfl FHHH NHHm m N m m 00H 00H N6 N .H Honpqoo m m I . I m.m I293 mm m, I I o.m nnom m H H I 00H I m.# I H9330 m m 0 9m We 0m WWW Hr H H. .HH mg. . mro .: m o In 00H H H.m mfi Honpdoo o e o . H.8H in 12.. «us HH m, n, m. m.m, o.m mmHm m m m 0 SH 0H r km N . m Honpsoo H H o m.moH m.m ,w.s, meHm, pl 0 o H.w an or: mmHfl H o H . H6 12.: H H 0 9H [mm .1: Hafiz m o o o 00H H m.m, ~.H, Hoasnoo m o m.mm Im.s, w.H nammmHz h m 1g b3: m.H HE H o o 0 SH uOH mt: b.m Honpqoo .Hmme .ee eemIp H .ee pMHIo .esrennuo .eexpmeo .ee wanna .es pmHuo,eoseaHoos> .npaaw mHonpnoo .Ho owepqoonom 35an A09 Mo .03 23on mo .02 me commefixo 3mm owsneaé 53 g usm 09935. 03.2 we modes HwHHom 032 no 3595 33332? Home Honpsoo on» HHH unseen Ho .3852 on» was 33.02. 3 omwoeonH omens: one Hp oHnne 5 2 m n m new I o.mHH 2m 8m Br 0 o o 803 @4er Tm Tm m: S m, o m, 00H 00H m.@ H.N Honpnoo m m o o.omH TR H. m.H me e o 0.5 How o.m .H n: sH o s, 00H 00H sum .m Hoseaoo q .n m 0.9 0.3 fin o.H mm HH OH H . 9mm oé 8H 0H mm MH m, mi : 00H 00H New N.m Honpnoo o o H m. o m.H 0 mm HH 11 o WQ 0.0m m.H m6 mm M! NH N H H 00H 00H Illrm.m :wa, Honpdoo W m m méfi flaw H H. m m.H womb P m m m.mmH o.ooH pd m.H Jimmow HH m N H 00H 00H ,mnm m.H Hoaeaoo i H o 9%. 0.8m .m. m4 RR H m m 98 95 .s as file 2 s, :1 o 00H 00H New, m.o Honpsoo H o H H.& .nM the m.m ion m H H Wmm m. we md 8%] m Inw H o H 00H 00H H.~ o.m Honpdoo H o H 98 QM Nd m.H ion m H m film” sdHH . dun H.n and] m m H H 8H 03 m6 m.m H9380 H32. as EMMEH as enHIo as some .8. enIHIIo a? some is: ano surgeons .nelenm mHonpnoo mo euopooon [dwse v Homv no .02 season Ho .02 ms commonmuo qum omuuoa< .93 sH uHsm omenohq 00H: no museum HeHhom AuoSdeuoev H> oHpuB 26. Discussion In the selection of mice, it was originally intended to follow the more strict criteria of Griffitts (5) rather than the relatively looser requirements of Ranta and Dolmen (13). The former author used equal numbers of female and male mice of the same strain weighing from 10 to 13 grams. Ranta and Dolmen employed inbred white mice of either sex of 10 to 25 grams initial weight.> Equal sex distribution.was effected in the above studies, but it had not been possible to adhere to the strict weight specifications. The non-homogenicity of the experimental conditions as evidenced by the wide discrepancies of deaths and weight increases among the different control groups emphasizes that the subsequent care of the animals is just as important, if not more so than the initial selection. Mucin enhances the virulence of Vibrio cholerae for'mice (12). By the use or the mucin vehicle, the virulence of cultures and the antigenicity of vaccines could be more accurately compared. Different lots of mucin may exhibit different enhancing capacities. While this would not have materially affected the ratio of the 50 per cent and . points between the vaccinated and unvaccinated mice in an antigenicity test, it would have made some difference in the comparison of virulence titrations between the various strains. To minimize this factor, the same lot of mucin was used for the whole series of virulence titra- tions of all strains throughout. 27. The use of living vaccine has long since been replaced by that of killed vaccines. Killed vaccine may be prepared by using either chloroform, heat, formalin or phenol. Phenol-killed vaccine has been shown by Ranta and Dolman'(14) to produce the highest and most enduring agglutinative titre in rabbits. It is also the most commonly used vaccine at the present time. Recently Jennings and Linton (16) recommended a direct cholera vaccine prepared from a fluid medium in which phenyl-mercuric acetate had been used as the killing agent. This method of vaccine preparation has not yet been widely adepted. The mouse protection test was chosen because the National Institute of Health of the'United-States has recommended it as a tentative procedure for manufacturers of cholera vaccine (1?). Griffitts (5) and Ranta and Dolmen (13) also considered it as a satis- factory antigenicity test. As no definite correlation has as yet been found to exist between the level of agglutinin and mouse pro- »tective titres, it would appear that the mouse protection test, being a more direct test, should be preferred over the agglutinative test. The two dose method of vaccination was adepted in immunizing the mice since the work of Rants and Dolmen (13) demonstrated that a single dose vaccination.does not confer high enough immunity to be clearly demonstrable. A.three dose vaccination procedure was found by the same authors not to induce significantly higher immunity than the two dose. . 28. p The fact that the strains of Vibrio cholerae isolated from the Chungking epidemic included three different types of the organism is of interest. Tang et a1 (8) in the study of strains isolated from.Kuming epidemic, and Burrows (18) in typing strains brought back by Dr. Reiman from.Chungking, encountered similar findings. These findings denote either that the particular epidemic did not originate from a single source, or that the antigenic structure of the causative organism.has undergone changes in the course of the epidemic. In any event, unless good cross protection is demon- strable among the various types of Vibrio-cholerae, which at present does not seem.to.be so demonstrated, it would be best to employ a mixed vaccine consisting of highly antigenic strains of various types for prephylaxis against the disease. The experiments so far conducted represent a preliminary survey of the problem. The results indicate that a virulent organism is required for the production of a vaccine of good antigenicity but virulence is not the sole criterion. The toxicity of a vaccine does not appear to bear any relationship to the virulence of the organism. While good cross protection exists among some of the strains, it does not exist among all strains. Reviewing the data in retrospect, »it is clear that the problem.demands more intensive study before definite conclusions can be made. It is particularly disappointing that no observations could be made concerning the strains that are currently in use for vaccine production in China. '29. The mucin suspensions used in the virulence titration experi- ments and in the mouse protection tests are virtually mixtures of living and dead organisms. The proportion between the living and the dead is apt to vary in different experiments. Biological assays using living organisms also generally yield less uniform results than those in which dead materials like toxins are employed. The endotoxin of Vibrio cholerae has been prepared and studied by Burrows and his associates and has been shown to exhibit immunologic pr0perties (19). It certainly would seem worthwhile to try to sub- stitute endotoxins of the various strains of Vibrio cholerae for the mucinized suspensions of the organism in the virulence and anti- genicity determinations. EEEEEEZ Sixteen strains of Vibrio cholerae were included in the present study. Of these, 2 came from the National Institute of Health of the United States Public Health Service, 10 were recovered from 12 strains newly isolated during the 1945 Chungking epidemic, and 4 were chosen at random from 8 strains currently in use in China for cholera vaccine production. All the strains showed typical morphological, cultural and bio- chemical characteristics of true'Vibrio cholerae. The only atypical finding was the slight hemolytic activity exhibited by strains 5122 and 5149. Serologically all 4 types.A, AB, AC, and ABC were 30. represented. The majority belonged to type AB. The strains isolated from the Chungking epidemic included types A, AB and ABC. The relative virulence of the strains of Vibrio cholerae was titrated in mice by intraperitoneal injection of mucinized suSpensions of the organisms using the Reed and.Muench method for LD50 deter- mination. Ten strains of Vibrio cholerae are highly virulent, four strains are moderately so and two strains are comparatively non- virulent. The LD5o doses are reapectively 80-400, 400-2000 and 2000-10,000 organisms. The third group is composed of strains isolated from the Chungking epidemic. The second group includes two epidemic strains and two vaccine strains from China. The first group includes 6 epidemic strains, 2 vaccine strains from China and 2 NIH strains. Phenol-killed vaccines were prepared separately from each strain. The vaccines were standardized to contain 1,600 million organisms per m1. A.two dose method of vaccination was adopted, administering 0.25 and 0.5 ml amounts of a given vaccine one week apart intraperitoneally into mice. Two weeks after the 2nd injection, the protective values of these vaccines were tested against the homologous strains and a reference strain, the NIHHl strain. The immunizing potencies of the vaccines was calculated by means of the Reed and Huench method. The protective value of the reference vaccine was also tested against the various strains. The antigenicity of vaccines prepared from.etrains of low virulence was poor. They protected against 1000 M.L.D.'s of the homologous 31. strains and 500 MeLeDe'S of the reference strain. The more virulent strains yielded vaccines of good immunizing potency against the homologous organisms, giving a protection in the order of 10,000 to 1,000,000 MtL.D.'s. They did not always protect well against the reference strain. Two of the epidemic strains yield vaccines as good as those of the two NIH strains in that they gave good protection against both homologous and reference strains. Duplicate tests on vaccines of strains 5053 and 5353 isolated from the Chungking epidemic and those of strains 35, 36, 45 and 48 currently in use for vaccine production in China gave such erratic results that no possible inter- pretation could be made. The reference vaccine gave good cross protection to nine heterologous strains but did not protect well against the Six others. The former group included representatives of all four types A, AB, AC and ABC. The latter group included members of type A, AB and ABC. There was only one type AC organism among the whole 16 strains. ‘Vaccinated and control groups of mice were weighed at O, l and 3 weeks, i.e., before each vaccination and prior to the:final challenge in an attempt to evaluate the toxicity of the vaccines from the average gain or loss of weight of the vaccinated groups as contrasted with the controls. The data obtained is difficult to analyze. Four vaccines showed high toxicity. Two of these were prepared from virulent strsins, one from a.moderate1y virulent strain and another 32. was derived from a comparatively non-virulent strain of'Vibrio ' cholerae. vaccine of another nonavirulent strain was tested side by side with the vaccine of a virulent strain. There was no appreciable difference between the toxicity of the two vaccines, both being mild in character. - The wide variations of the weight gains among the different control groups speak for nonphomogenicity of the animal pepulation. Factors which were either known to have existed or were suspected to have been present to contribute to such conditions were reviewed in the text.- It is felt that the problem.deserves more intensive study and that definite conclusions can not be made at the present time. The isolation of multiple types of Vibrio cholerae from a single epidemic, however, Justifies the recommendation of the use of a mixed vaccine, made and pooled from highly antigenic strains of various types, for prephylaxis against the disease. l. 2. 10. ll. l2. 13. 14. Literature Greenwood, M., and Yule, G. U. Proc. R. Soc. Med. Soc. Epidem. & State Med. 8: 113, 1915 Kolle, W. Deutsche Hod. 'I'I'chnschr. 23: 1+, 1897 Takano, R., Ohtsnbo, I., and Inonye, I. Studies of Cholerae in Japan, League of Nations Health Organization, Geneva, 1926 Yang,‘Y} N., Tsao, S. L., Chang, Y., and Chung, C. Y. Chin. Med. J. (Supp.) 202, Feb. 1936 Griffitts, J. J. Pub. Health Reports 59: 1374, 19w Burrows, W., et al. J. Inf. Dis. 79: 159, 168, 1946 Yu, H. Chin. Hod. J. 54: 255, 1938 Tang, Fe Fe, Chu, Ce Me and Wong, Ye We Indian J. M. Research 32: 1, 1944 Linton, R. W. Diagnostic Procedure and Reagents, Second Ed., Am. Pub. Health Assoc. The Diagnosis of Cholera, p. 513, 1945 Personal communication Reed, L. J., and.Muench, H. Am. J. Hyg., 27: A93, 1938 Griffitts, J. J. Pub. Health Reports, 57: 707, 19112 Ranta, L. E. and Dolman, c. E. Canad. J. Pub. Health, 35: 113, 1912+ Rants, L. E. and Dolmen, C. 1:3. Canad. J. Pub. Health, 34: 2o, 1943 National Institute of Health, Bethesda, Md., Jan. 5. 1946 A tentative mouse protection test for determining the anti- genicity of pertussis vaccine. 33. 16. 170 18. 19. 34. Jennings, R. K., and Linton, C. W. J. Franklin Institute, 238: 65, 1944 National Institute of Health, Bethesda, Md., 1942 Comimlnication to Manufacturers of Cholera Vaccines Personal comnunication Burrows, W., et al. Proc. Soc. Exper. Biol. 8. Med. 57: 306, 308, 1944 I < 1 '=‘\. ‘ . .1 -. E . l - 'L'. 9‘4“ 54le -.,_‘I' .~ , . st? 5 ‘57 ’ 1612.9 190255 ! 3875 Huang MICH TATE UNIVERSITY LIBRARIES 293 03082 9398 IIGIIIIINIII 3 1