The protective effect in bank voles of some strains of BCG*

Bull. Org. mond. Sante 1 1970 43, 71-90Bull. Wld Hlth Org. The Protective Effect in Bank Voles of Some Strains of BCG* A. LADEFOGED,1 K. BUNCH-CHRISTENSEN' & 3. GULD' The bank vole, like the guinea-pig, can be i}nwnzed against tuberculosis with very small doses ofBCG, presumably because BCG is sufficiently virulent for both these species to multiply freely until the immunization of the animal has reached a maximum. Large doses of BCG induce an earlier, but not an ultimatdy stronger, immunity. However, the smallest immunizing dose is not the same for all BCG strains and this is afeature that may be used to differentiate strains. In the studies reported by the authors, the smallest immu- nizing dose for the bank vole was estimatedfor 11 BCG strains, some ofthem widely used in man. The results indicate potency ratios of20: 1 between the strongest and the weakest strains. The ranking of strains in terms of the smallest immunizing dose was found to be correlated in most, but not in all, cases with a previously reported ranking of the lethal effects of the various strains in golden hamsters. The implications of these findings for production requirements and for reference preparations ofBCG are discussed. It has been suggested (WHO Expert Committee on Biological Standardization, 1966, p. 38) that a BCG strain (seed-lot) used for production of vac- cine ". when injected into guinea-pigs and mice in a range of different doses under reproducible con- ditions, should afford a measure of protection against a range of challenge doses of a fully virulent human strain of Mycobacterium tuberculosis, not less than that specified by the national control authority in terns of a national reference preparation" cali- brated in terms of the international reference preparation. Traditionally, many manufacturers of BCG vac- cine and some national control authorities have satisfied themselves about the protective effect of a BCG vaccine by showing that a large dose (of a magnitude corresponding to that used in man) induces a certain degree of resistance to experi- mental challenge in vaccinated guinea-pigs, when From the WHO International Reference Centre for BCG Seed-Lots and Control ofBCG Products, Copenhagen, Denmark. 'BCG Department, Statens Seruminstitut, Copenhagen, Denmark. Chief, BCG Department, Statens Seruminstitut, Copen- hagen, Denmark. IMedical Officer, Tuberculosis, Division of Communi- cable Diseases, World Health Organization, Geneva, Switzrland. compared with non-vaccinated control animals. Dur- ing recent years, it has been demonstrated that the ultimate effect of BCG vaccination is independent of vaccine dose over a wide range, both in the guinea-pig (Jespersen, 1956; Tolderlund et al., 1960) and in the bank vole (Jespersen, 1954; Jespersen & Weis Bentzon, 1964). Large doses ofBCG induce an earlier, but not an ultimately stronger, response than do small doses, presumably because BCG is suffliciently virulent for both species of animal to multiply freely until the immunization has reached a maximum. In very low dosages, i.e., those con- taining about 10 viable units, there is, however, a dis- continuity in the dose-reponse function; such dos- ages, or those still smaller, may fail to immunize. On the basis of these findings, Jespersen (1954) sug- gested that differences between BCG strains might be more easily demonstrated at this low dosage level. Strains shown to differ in activity in a particular animal model need not necessarily differ in their effect in man and, indeed, they sometimes fail to show any difference in another animal model. In the absence of precise knowledge about the relative effect of different strains in man, it would never- theless seem prudent to compare strains in widely different animal models and perhaps avoid the use of strains that tend to rank low, even if only in some models, either in terms of vaccination-induced 254 - 71 A. IADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD resistance to experital challenge or in terms of ability to multiply in the mammalian body (Dubos & Pierce, 1956). The WHO International Reference Centre for BCG Seed-Lots and Control of BCG Products is carrying out a number ofstudies, in different species of animal, aiming at characterization and ranking ofBCG strains, especially strains widely used in man. The present report covers 6 studies of the minimum protective dose ofBCG in bank voles (Clethrionomys glareolus), comparing altogether 11 strains. Myco- bacterium bovis was used for challenge because Myco. tuberculosis is of low virulence in this species. Length of survival after challenge was taken as an indication of immunity. MATERIAL AND METHODS Preliminary studies Several preliminary studies were carried out, using various seed-lots and orial BCG products. The interpretation of these stuidies was difficult since it was not always clear whether potency differences were due to differences in production methods or to genetic differences between strains. These prelimi- nary experiments, with one relevant exception (Fig.7), are not reported here. The six main experiments, described below, were all done with liquid vaccines specially prepared for the purpose by a uniform technique. Design Altogether, 11 BCG strains were examined during the course of 6 uniformly designed experiments. Each experiment involved 5 strains, and each strain was examined in 2 or 3 experiments. For each BCG strain, doses ranging in fourfold steps from about 1 culturable particle to about 1000 culturable particles (as estimated by counting colonies growing on solid medium) were inoculated (0.1 ml intra- peritoneally) in bank voles; 6 males and 6 females being used for each dosage. Eight weeks after vac- cination, the vaccinated animals and a non-vac- cinated group given placebo only (about 50 animals in each study) were challenged intravenously with virulent Myco. bovis. The animals were observed until spontaneous death occurred, the duration of survival after chal- lenge being taken as an index of the relative protec- tion conferred by the BCG. No autopsy was per- formed but a control group of non-vaccinated, non- challenged voles (about 40 per experiment) was included in order to reveal any sudden, 1ethal, non-tuberculous epidemic among the experimental animals. BCG strains Strain 1331, used in Copenhagen and Madras since 1966, was included in all six experiments. Other strains were those routinely used in Paris (1173P2), Moscow (3522), Tokyo (172), Rio de Janeiro (Moreau), Gothenburg, London (FIO) and Prague (725). Strain 809, used in Madras until 1966, and two experimental strains (designated Y and R) were also included. The strains were included in the six experiments as shown in Table 1. TABLE 1 STRAINS OF BCG USED IN THE SIX EXPERIMENTS Experiment Strain of BCG 1 2 3 4 5 6 Copenhagen 1331 Paris 1173P2 Moscow 3522 Tokyo 172 y London FIO R Madras 809 Prague 725 Rio de Janeiro (Moreau) Gothenburg x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Vaccine preparations All strains were grown as surface cultures on Sauton medium and harvested after 10 days. Liquid vaccines were prepared by a uniform technique, according to the routine used in the BCG Depart- ment of the Statens Seruminstitut, Copenhagen. The strains from Paris and Madras grew faster than the other strains and showed pronounced pigmentation. For the Paris strain, the first prepara- tion contained a lower proportion of live organisms, as indicated by lower oxygen uptake (see Table 2), but the suspension prepared for the second experi- ment was satisfactory. The strain from Tokyo yielded vaccines with a very high colony count 72 73PROTECFIVE EPFECT IN BANK VOLES OF SOME STRAIN OF DBM TABLE 2 IN VITRO CHARACTERISTICS OF BCG SUSPENSIONS Opacity Oxygen uptake Germination Cufturabie Strain Experiment (suspension xgIper 120 mg rate particlesNo. of slper 120um) (percentag at (millions0.75 mg/ml) per ou 24 hours) per mg) 1331 1 0.34 136 50-75 12.3 2 0.40 167 75-90 21.6 3 0.38 154 75 17.9 4 0.40 173 75 21.1 5 0.30 179 75 189 6 0.36 162 75 15.5 Paris 1173 P2 1 0.41 86 75 15.5 2 0.29 149 75-90 17.3 Moscow 3522 1 0.37 171 75-(90) 22 2 0.39 208 75-90 34.7 Tokyo 172 1 0.38 170 75 8B7 2 0.38 195 75 79.5 Y 1 0.44 181 75 32.3 2 0.42 143 75 183 London FIO 3 0.38 73 75 5.9 4 0.36 84 50 7.2 6 0.36 77 50-75 6 R 3 0.33 171 75-90 533 4 0.34 174 75 44.3 6 0.40 162 75 25.3 Madras 809 3 0.39 174 75-0 29.6 5 0.38 193 75-90 22.1 Prague 725 3 0.34 146 75 14.9 5 0.33 121 75 15.7 Rio de Janeiro 4 0.41 218 75-0 28.8 5 0.38 155 (50)-75 23.5 6 0.41 174 75 22.9 Gothenburg 4 0.40 166 75 480 5 0.33 186 75 363 6 0.32 192 75 29.9 4 The parentheses indicate that the result was considered to be closer to the percentage not in parentheses. A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD which is a well-known feature of this strain. The strain from Gothenburg and the experimental strain designated R also yielded vaccines with a rather. high colony count. The London strain grew more slowly than the other strains, and the morphology of the surface culture was distinctly different from the other strains. The proportion of live organisms in all three vaccine suspensions was slightly reduced as shown by oxygen uptake and germination rate. Otherwise, the vaccines from different strains were much alike in a number of in vitro tests, all details of which are given in Table 2. Test animals The bank voles were 7-8 weeks old at the time of vaccination. For each experiment the animals were randomly allocated (though stratified for sex) to the different strains and doses of BCG, to the placebo group and to the non-vaccinated, non- challenged control group. Each animal was housed in a stone jar fitted with a metal lid. The jars were placed in the animal house in a formally randomized order. Challenge Cultures (10-12 days old) of a virulent bovine strain (E6884), grown on Dubos liquid medium, were injected intravenously into the tail vein in doses of 0.2 ml of 1/1000 dilution. The challenge doses were estimated from colony counts performed on solid medium, to contain the following numbers of culturable particles. Experiment No. 1 2 3 4 5 6 No. of culturabk particles 18 000 27 000 23 000 11 000 14 000 11 000 Follow-up and analysis Experiments 1, 2, 3, 4 and 6 continued for 225, 177, 207, 184 and 227 days, respectively, after chal- lenge, at which time more than 99% of the challenged controls, and more than 97% of the challenged, vaccinated animals had died. Experiment 5 was prematurely interrupted on the 94th day when a breakdown in the ventilation system in the animal house caused a large number of deaths among all groups of animals; out of 39 non-challenged con- trols still alive on the 93rd day 28 died within the next 48 hours. A special method has therefore been used for the analysis of this experiment. Of the 1848 BCG-vaccinated voles, 137 have been excluded from the analysis: between vaccination and challenge, 101 died, and 4 escaped from the cages and could not afterwards be identified. After chal- lenge, 6 died within 1 day, and another 26 within 12 days; the latter have also been excluded since it seems most unlikely that they could have died from tuberculosis at this early stage of challenge. Of the 300 challenged control animals, 20 have been excluded: 1 escaped from the cage, 17 died before challenge and 2 died less than 12 days after challenge. Of the 229 non-challenged controls, 19 died before the rest of the animals were chal- lenged, 3 died within 12 days of challenge, and another 39 died before the end of the respective studies. The survival times for all animals still alive at the date of challenge are listed in Appendix Tables 1-6. The distributions mostly tend to be skewed to the right and the analysis has therefore been made in terms of geometric means, i.e., by use of a loga- rithmic transformation. In the computations, for animals still alive at the end of the experiment, the duration of the experiment has been taken as the survival time. For experiment 5, the results are given in terms of the proportions of animals still surviving 93 days after challenge, at which time this experiment was prematurely terminated. RESULTS The results are illustrated in Fig. 1-6, in terms of survival as a function of vaccine dose in millionths of a milligram (semi-dry weight). The strains tend to differ at quite low dosages, i.e., around 0.000001 mg or about 5-50 culturable particles. For high dosages and dosages even lower than 0.000001 mg, the curves tend to level off; this trend was anticipated in the design of the experiments. Assuming that the curves are of sigmoid shape, it is reasonable, from a statistical point of view, to associate the concept of minimum protective dose with the point of steepest ascent for each curve. In practice, not all curves conform smoothly to the sigmoid model; this is true especially of those for the best-protected groups, and judgement of significance must ulti- mately be based on consistency from experiment to experiment. The following comments are largely based on inspection of the curves. A comparison of the first two experiments (Fig. 1 and 2), both dealing with the same 5 strains, indicates 74 PROTECTIVE EFFCr IN BANK VOLES OF SOME STRAINS OF BCO FIG. I EXPERIMENT 1: MEAN SURVIVAL OF CHALLENGED BANK VOLES,a AS A FUNCTION OF DOSE OF LIQUID BCG 130 720 J 770~~~~~~~~~~~~~~~~~~~~~~~ 7000 Q2: Lu 90 Iaj o Z tu ~~~~~~~~~CONTROL ANIMALS60 50 40 1331 v') : e o o o eoPARIS 1173 P2 a 30 MOSCOW 3522 HHHH- TOKYO 172 b U Ub y DOSE Of BCG a Challenge dose: 18 000 culturable particles of a bovine strain. 75. A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD FIG. 2 EXPERIMENT 2: MEAN SURVIVAL OF CHALLENGED BANK VOLES,a AS A FUNCTION OF DOSE OF LIQUID.BCG. 130 120 2 110 'U,, u 100" z '8U Laj Lu 60 50 U) CONTROL ANIMALS 00- ~~~~~~~~~~~~~~~~1331 a 30 PARIS 1173 P2 MOSCOW 3522 *I TTOKYO 172 Y 012 0.48 1.9 7.5 DOSE OF B CG (10J6mg) 30 a Challenge dose: 27 000 culturable particles of a bovine strain. 76 PROTECTIVE EFFECr IN BANK VOLES OF. SOME STRAINS OF BCO FIG. 3 ffIPIMENT 8: MEAN SURVIVAL OF CHALLENGED BANK VOLES,5 AS A FUNCTION OF DOSE OF LIQUID BCG 130 , 1206 ,,,, 1001 LU2 90 X M 05 a: .*3 us ti. X40 LONDONF I -10 CA 30 -MARAS80R 77 *- PRAGUE 725 ' Challenge dose: 23 000 culturable particles of a bovine strain. E 78 A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. OULD FIG. 4 EXPERIMENT 4: MEAN SURVIVAL OF CHALLENGED BANK VOLES,a AS A FUNCTION OF DOSE OF LIQUID BC6 ItDtJAYS 1300 120- % 170 _I X 141 o 100- LU90 0090 Z 70 - g60 / .*CONTROL ANIMALS 50 V) 40- U, . -~~~~~~~~~~~~~~~ ~~1331 030F A4A-~' LONDON FIO o. O R 4a a a 0-oa RIO DE JANEIRO ----. GOTHENBURG DOSE OF BCG It Challenge dose: 11 000 culturable particles of a bovine strain. PROTECrIVE EFFECIr IN BANK VOLES OF SOME SRNS OF BO7 FIG. 5 EXPERIMENT 5: PERCENTAGE OF CHALLENGED BANK VOLES,a SURVIVING FOR THE TOTAL DURATION OF THE EXPERIMENT,b AS A FUNCTION OF DOSE OF LIQUID BCG 9! so - ! CONTROL ANIMALS CONTROL ANIMALS - *Z1331 E3--- -- GOTHENBURG QS0 ------0 MADRAS 809 *-.-.--.-. PRAGUE 725 .S-ein--ne- RIO DE JANEIRO A MO 0OORO 0.03 0.12 0.48 1.9 7.5 30 120 DOSE OF BCG(10-6mg) ° Challenge dose: 14 000 culturable particles of a bovine strain. b 93 days. a rather consistent ranking, the Paris strain being the strongest and the Tokyo strain the weakest of the 5. The Moscow strain and strains 1331 and Y are intermediate. If the curves are compared at the level of the steepest ascent, the horizontal distance between the extreme curves will be seen to correspond to a potency ratio in terms of bacillary weight of almost 16 times. The difference would have been even more striking (almost 80 times) if the compari- son had been made on the basis of culturable par- ticles, the number ofculturable particles being 5 times higher for the Tokyo than for the Paris strain in the present experiment (see Table 2). However, the number of culturable particles may be less relevant since the bacilli of the Tokyo strain are much smaller than those of the Paris strain (Conge & Dubos, 1966) and also because the vaccination skin lesions in children for these two strains are rather more equal in size for doses equal by weight than for doses equal by count (Bunch-Christensen et al., unpublish- ed data). The experimental strain R is included in experi- ments 3, 4 and 6. It is very weak in the two former experiments but quite potent in the latter, an incon- sistency that was confirmed in studies of the smallest allergenic dose in guinea-pigs (Bunch-Christensen et al., unpublished data). This strain originates from a single, selected colony of the Tokyo strain; its erratic behaviour in these experiments is difficult to explain. Otherwise, the results of experiments 3-6 are rather consistent. The most potent strains are the former Madras strain, the Rio de Janeiro and Gothenburg strains and strain 1331, perhaps in that order. The strains from London and Prague are consistently less potent than the rest, having to be given in a dose some 5-20 times larger than that of any other strain to confer immunity. For the London strain, the significance of this finding is not entirely unambiguous, as discussed elsewhere (Bunch- Christensen et al., 1970). In preliminary experi- ments, attempts were made to compare original, 79 80 A. LADEFOGED, K. BUNCH-CRISTENSEN & J. GULD FIG. 6 EXPERIMENT 6: MEAN SURVIVAL OF CHALLENGED BANK VOLES,5 AS A FUNCTION OF DOSE OF LIQUID BCG 150 140 _ 130- 120- 'u 110 "90 tI1 ~80/ (3 &- 70- ?~60 ,~~~~~50~ ~ ~ 50 Q 40 *- 1331 b- a * A *LONDON FlO O.........O R a O O O O- RIO DE JANEIRO -- GOTHENBURG a03 DOSE OF ECG (1/0mg) a Challenge dose: 11 000 culturable particles'of a bovine strain. PROTECTIVE EFFECT IN BANK VOLES OF SOME STRAINS OF BCG routinely made, freeze-dried vaccines from the respective production laboratories and the results of one such experiment, which included an original Glaxo vaccine batch F10, are illustrated in Fig. 7. At the time, several explanations for the low potency of the Glaxo vaccine were considered. The low number of culturable particles per ml of vaccine (which in turn could be due to a smaller content of bacterial mass, to low viability, or to clumping) was still not quite low enough to explain the difference in the bank voles. It was assumed at the time, on the basis of information from the manufacturer, FIG. 7 MEAN SURVIVAL OF CHALLENGED BANK VOLES,a AS A FUNCTION OF DOSE OF FREEZE-DRIED BCG b 1201 1101- w I.. LIZ 'ii -JQ: -J IdI a o 0 U)- Qu 100- 901- 80 70 60 SO 40 301- 20 - 10 A J* 2 8 DOSE OF DCG (/O m{) a Challenge dose: 10000 culturable particles of a bovine strain. b Production laboratories in Dakar (French strain), London and Gothenburg. 81 CONTROL ANIMALS 0 o o o- O DAkAR (1173) 11.SX /0 mI '- - -* LONDON FlO 3.3X/O'ml - - - GOTHENBURG I1.6X/O' mI 32 128 wmo oom A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD that the vaccine had been prepared from the Copen- hagen strain, and the remaining possibilities were either a lower in vivo potency of the Copenhagen strain as such or a phenotypic reduction of potency when the strain is grown in deep culture as practised by the manufacturer. The next obvious step was to transfer the vaccine to the surface of Sauton medium, the medium on which the Copenhagen strain is tradi- tionally grown, and this revealed that the change was not simply phenotypic but profoundly genotypic. In other words, the London strain is genetically different from the Copenhagen strain. DISCUSSION These studies have confirmed that the minimum immunizing dose of BCG in bank-voles is very low for certain strains; e.g., for the Paris strain 5-10 cul- turable particles. It is an obvious inference that the BCG must have multiplied extensively in vivo, and presumably what is really measured with this model is the minimum number of culturable particles of a certain strain needed to induce a BCG infection in the species concerned. In other words, the model measures an aspect of residual virulence. Dubos & Pierce (1956) suggested that a not-too-low residual virulence is an important attribute of BCG: '. . . there is no doubt that immunity is an expres- sion of the ability of the BCG organisms to multiply in the body of the vaccinated individual. The immunity is dependable and lasting if the BCG culture is endowed with great invasiveness, whereas it is at best weak and transient if the BCG culture has become very attenuated ". The residual virulence in the bank vole, or in any other species of rodent, does not necessarily predict the residual virulence in man but it should not be forgotten that one particular strain, characterized by Dubos et al. (1953) as having exceptionally low virulence in white mice, was obtained from a labora- tory that at about the same time supplied vaccine to the one controlled human trial in which BCG failed to give protection (Comstock & Palmer, 1966; Comstock & Webster, 1969). The low virulence and low protective power of this strain were later con- firmed by Willis & Vandiviere (1961) in guinea-pigs, and by Jespersen & Weis Bentzon (1964) in hamsters and bank voles. Jespersen & Weis Bentzon (op. cit.) found a clear relationship between the lethal effect of a BCG strain for hamsters and its ability to induce acquired resistance in bank voles. This is mostly true also for the strains examined in this study, for which the lethal effect in hamsters has been reported elsewhere (Bunch-Christensen et al., 1968, 1970). A striking exception is the former Madras strain which is of low virulence in hamsters but highly potent in bank voles. On the other hand, the strains most lethal for hamsters are all protective in low dosages in bank voles. The relatively low activity of the Tokyo strain, both in hamsters and in bank voles, is in agreement with findings in mice reported by Conge & Dubos (1966). This strain is the same as that of the Inter- national Reference Preparation of BCG, which was established with the proviso that " since the protec- tive value of BCG vaccines prepared from different strains has not yet been adequately compared . . . for the present the international reference preparation should be reserved only for laboratory control tests and tests for allergenic activity and vaccination lesions in children" (WHO Expert Committee on Biological Standardization, 1966, p. 14). The use of the preparation as a reference for protection in mice and guinea-pigs is nevertheless implied in the manufacturing requirements for dried BCG vaccine (WHO Expert Committee on Biological Standard- ization, 1966, p. 38). The present results do not encourage the use of the existing reference for this latter purpose; if animal tests are to be used at all as a criterion for selection of BCG strains, a reasonable reference would be a strain that (among strains proved to be acceptable and safe in man) was not inferior, in terms of activity in animal tests, to any existing or potential national reference preparation. "UM8 EFFET PROTECTEUR DE CERTAINES SOUCHES DE BCG CHEZ LE CAMPAGNOL GLAREOLE Le Centre international OMS de ref6rence pour les lots de semence de BCG et le controle des preparations de BCG, Service du BCG, Statens Seruminstitut, Copen- hague, Danemark, poursuit actuellement une serie de recherches chez diverses especes animales en vue de caracteriser et de classer les souches de BCG, en particu- lier celles qui sont couramment utilis6es chez l'homme. Le present rapport expose les r6sultats de six 6tudes sur les doses minimales de BCG conferant une protection au campagnol (Clethrionomys glareolus). 82 PROTECIVE EFFECT IN BANK VOLES OF SOME STRAINS OF BCG 83 II suffit de tr&s petites doses de BCG pour immuniser le campagnol- de meme d'ailleurs que le cobaye. Cela tient probablement au fait que le BCG, 6tant suffisam- ment virulent pour ces deux especes, peut se multiplier librement jusqu'a ce que l'immunisation ait atteint son plus haut niveau. Les fortes doses de BCG assurent une protection plus pr6coce mais qui finalement n'est pas de meilleure qualit6. Cependant la dose immunisante mini- male n'est pas la meme pour toutes les souches de BCG, et on dispose ainsi d'un moyen de les distinguer. Onze souches ont ete 6tudiees au cours de six expe- riences realisees dans des conditions identiques. Apres avoir 6t6 cultiv6es en surface sur milieu de Sauton, toutes les souches ont ete r6colt6es au boe jour et ont servi a pr6parer un vaccin liquide selon la technique standard en usage au Service du BCG du Statens Seruminstitut. Les diff6rents vaccins ont fait l'objet de nombreux tests in vitro. Chaque experience comportant l'examen de cinq souches, toutes les souches ont ete 6tudi6es aL deux ou trois reprises. Pour chaque preparation de BCG, des doses allant, en progression de raison 4, de une a mile unit6s viables (concentration estim6e par numeration des colonies sur milieu solide) ont ete inject6es (0,1 ml par voie intraperiton6ale) a des campagnols, six animaux mAles et six animaux femelles etant utilis6s pour chaque dosage. Huit semaines apres la vaccination, les animaux immunises et un groupe t6moin de campagnols non vaccin6s (50 environ par experience) ont recu en injection intraveineuse d'6preuve 11 000 a 27 000 unit6s viables de Mycobacterium bovis virulent. (En raison de sa faible pathogenicite pour le campagnol, Myco. tuberculosis ne peut etre utilise a cet effet.) Les animaux ont Wt6 observ6s jusqu'h leur mort, et la dur6e de la survie apr6s l'injection d'epreuve a ete choisie comme indice de la protection relative conf6r6e par le vaccin. Aucune autopsie n'a ete pratiqu6e, mais on a inclus dans chaque experience un groupe t6moin d'environ 40 campagnols non vaccin6s et non soumis 'a l'injection d'6preuve pour d6celer toute epid6mie g6n6ra- trice de mortalite d'origine non tuberculeuse parmi les animauxc. Au cours d'une experience, une defaillance de l'installation de chauffage a provoqu6 la mort d'un grand nombre d'animaux au 94e jour suivant l'injection d'6preuve; dans ce cas particulier, l'analyse porte sur les r6sultats exprim6s en pourcentage d'animaux survivants au 93e jour. Les figures 1-6 donnent la repr6sentation graphique des r6sultats exprimes par la dur6e de la survie en fonc- tion de la dose vaccinale en millioniemes de milligramme (poids demi-sec). Les diff6rences entre souches sont particulierement apparentes pour les dosages tres faibles, de l'ordre de 0,000001 mg, ou 5-50 unites viables, de BCG. Avec des dosages plus 6lev6s ou plus faibles, les courbes ont tendance a se situer a des niveaux semblables, ce qui est conforme aux pr6visions faites lors de la conception des experiences. Si l'on admet que les courbes sont du type a sigmolde *, on peut raisonnablement, du point de vue statistique, associer la notion de dose protectrice minimale au point oiu chaque courbe presente le maximum d'ascendance. Les r6sultats font apparattre un rapport d'activit6 de 20 a 1 entre les souches les plus fortes et les souches les plus faibles. Les souches de Paris et de Rio de Janeiro, ainsi que la souche utilis&e autrefois a Madras, figurent parmi les premieres, tandis que les souches de Londres, de Tokyo et de Prague sont les moins actives. II ne convient pas d'attribuer une valeur absolue aux resultats enregistres avec la souche de Londres, car celle-ci, qui est habituellement maintenue en culture profonde, a ete cultivee en surface sur milieu liquide de Sauton en vue des presentes experiences. Cependant, les resultats d'une experience pr6liminaire men6e avec des vaccins d'origine lyophilises plaident egalement en faveur d'une faible activite de la souche de Londres (fig. 7). Ces etudes confirment que la dose immunisante mini- male chez le campagnol est tres faible pour certaines souches (par exemple, 5-10 particules cultivables pour la souche de Paris). On en conclut que de toute evidence le BCG se multiplie in vivo de maniere intensive, et il est vraisemblable que ce modele animal mesure en realite le plus petit nombre de particules cultivables d'une souche donn6e capable de provoquer une infection par le BCG chez cette espece. En d'autres termes, ce modele fournit une estimation d'un aspect de la virulence residuelle. Les auteurs ont etudie ces memes souches sous l'angle de leur pouvoir letal a forte dose pour le hamster dor6. Les resultats obtenus grace a ces deux modeles animaux sont dans l'ensemble tres concordants, malgre une exception notable, celle de l'ancienne souche de Madras qui fait preuve d'une tres grande activite chez le cam- pagnol mais n'est que peu virulente pour le hamster. Etant donne que la preparation internationale de reference derive de la souche de Tokyo, la signification de la faible virulence residuelle de cette souche pour l'utilisation de la preparation internationale de reference est examin&e et discutee. REFERENCES Bunch-Christensen, K., Ladefoged, A. & Guld, J. (1968) Bull. Wld Hlth Org., 39, 821 Bunch-Christensen, K., Ladefoged, A. & Guld, J. (1970) Bull. Wld Hlth Org., 43, 65 Comstock, G. W. & Palmer, C. E. (1966) Amer. Rev. resp. Dis., 93, 171 Comstock, G. W. & Webster, R. G. (1969) Amer. Rev. resp. Dis., 100, 839 84 A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD Conge, G. & Dubos, R. J. (1966) In: Recherches sur le B.C.G., Paris, Flammarion, pp. 245, 247, 248 Dubos, R. J., Pierce, C. H. & Schaefer, W. B. (1953) J. exp. Med., 97, 207 Dubos, R. J. & Pierce, C. H. (1956) Amer. Rev. Tuberc., 74, 699 Jespersen, A. (1954) Acta path. microbiol. scand., 35, 396 Jespersen, A. (1956) Acta Path. microbiol. scand., 38, 203 Jespersen, A. & Weis Bentzon, M. (1964) Acta tuber. scand., 44, 253, 276 Tolderlund, K., Bunch-Christensen, K. & Waaler, H. (1960) Bull. Wld Hlth Org., 22, 185 Willis, S. & Vandiviere, M. (1961) Amer. Rev. resp. Di, 84, 288 WHO Expert Committee on Biological Standardization (1966) WId Hlth Org. techn. Rep. Ser., No. 329 PROTECTIVE EFFECT IN BANK VOLES OF SOME STRAINS OF BCO 85 APPENDIX TABLE I EXPERIMENT 1: INDIVIDUAL SURVIVAL a OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA Copenhagen Paris Moscow Tokyo Y Placebo, Controls, 1331 1173 P2 3522 172 challenged challenged MI F Im I I I I F d F _______ I F Cotrls r r T~~~~~I T I 1-6 0.24 x 10 sag 38 43 50 68 66 70 79 77 83 78 92 96 -6 0.96 x 10 sag D D 55 0 55 94 73 96 79 105 104 S -63.8 x lI sag 49 58 77 89 95 98 103 99 107 114 192 130 1S x 16 g D 66 61 85 88 139 96 157 117 167 149 S 60 x10 6 mg 82 91 98 111 127 160 0 73 92 146 160 184 0.48 x 49 49 72 117 208 212 1.9 x D 90 94 94 109 145 7.5 x D 83 109 118 150 223 30 x D 58 87 99 142 162 120 x 30 58 98 119 149 163 -6 10 sag D 45 69 83 91 95 -6 10 sag 87 130 137 138 139 214 -6 10 sag U 96 100 136 152 153 10smg 75 84 85 105 122 147 -6 10 sg 98 103 156 173 186 S -6 0. 24 x 10 mg 35 53 39 75 6O 76 70 96 127 122 140 125 0.96 x 10 amg D 35 51 49 60 57 134 65 175 89 188 94 -6 3.8 x 10 sag 81 66 85 67 101 111 113 141 118 142 180 163 -6 1S x 10 6mg 71 D 91 0 102 83 175 98 206 136 S 200 0 x 10 smg 27 79 32 106 131 121 136 140 148 168 159 171 0.06 x ] 38 49 54 62 64 65 -60 mag 0 50 80 84 114 117 1-60. 24 x 10 mg 51 33 54 54 56 61 61 81 65 81 103 105 0.96 x 106 mg D 39 32 59 38 60 53 64 71 70 94 91 3.8 x lo s6g D 45 52 77 52 100 60 107 88 109 117 131 15 x 10 g 79 64 80 67 94 81 98 104 113 145 183 200 60 x 7-6 D 82 D 105 91 133 91 139 95 167 120 217 0.24 31 4] 5: 7: 10' 0.96 6J 6: 8] 8], 3.8 2' 8] 91 91 10' 13] 15 8] 81, 101 11I 141 60 61' 91 12: 13] -6x 10 mg U 36 9 61 2 72 3 72 3 79 5 84 -6 x 10 mg 5 61 1 97 3 98 1 115 a 164 ) 179 x 10- mg 9 69 1 102 1 128 3 153 5 156 1 216 -6 Y 10 6mg 7 D 1 85 7 99 5 106 L 115 5 130 Xl1O sag 5 92 7 96 1 96 3 98 2 113 1 184 D 32 32 34 34 35 35 44 45 46 48 49 52 55 57 58 60 61 64 64 64 64 72 75 96 42 43 43 44 45 49 53 58 60 61 63 64 65 67 72 75 77 77 95 99 99 102 107 119 S D D D 27 109 119 S S S S S S S S S 8 S S S 53 185 195 S S S S S S S 8 8 S S S S S S we" oonZ3 a Days. M - Male F - Female D - Animal died before date of challenge S - Animal surviving at the end of the experiment (225 days) U - Animal unidentiflable (escaped from cage) A. LADEFOGED, K. BUNCH-CHRISTENSEN & J. OULD APPENDIX TABLE 2 EXPERIMENT 2: INDIVIDUAL SURVIVAL a OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA . Controls, Copenhagen Paris Moscow Tokyo Y Placebo, ngn- 1331 1173 P2 3522 172 challenged cFllaengediiIiIIFlIIEI:ZEI ZZ U F I -60.24 x 10 Dlg 29 23 29 24 29 26 60 45 82 46 124 73 0.96 x 10 6mg 36 7 54 15 56 52 69 53 77 56 92 69 3.8 x 10- 6 mgl 7.5 45 36 58 44 85 76 90 95 114 100 147 109 -6 15 x 10- mg 62 72 73 73 84 78 96 91 98 152 133 177 60 x 10 6 mg 79 66 82 85 87 90 91 94 102 147 173 153 I -60.48 x 10 mg 29 23 51 51 60 61 88 85 97 87 99 106 1.9 x 106 mg 68 79 82 85 85 91 91 100 109 124 109 156 6( 81 9C 10( 114 30 55 5' 6: 7: 104 10l 120 6~ 7 1 8( 89 94 - - x10 ,mg 6 23 62 ) 63 ) 74 5 78 S 106 -6 x 10 6mg 72 83 a 91 a 102 1 141 B S -6 x 106 mg a 72 7 77 ) 86 3 90 6 125 S 134 0.24 x 106 mg 27 24 37 29 46 39 63 39 65 41 71 47 0.96 x 10 6mg D 52 49 57 55 64 89 67 109 79 130 87 3.8 x 106 mg 49 62 59 75 60 85 86 105 89 120 125 130 -6 15 x 106 mg 62 62 73 89 89 99 102 100 119 114 136 144 -6 60 x 10 mg 55 6 82 122 88 129 97 130 107 151 109 151 -60.06 x 10 mg 27 29 37 35 40 36 49 45 51 146 51 86 0.24 x 25 31 43 53 56 70 0.96 x 24 35 42 43 72 84 10 mg 40 44 44 58 61 70 -610 mag 19 39 45 47 79 83 -63.8 x 10 mg D 30 0 33 25 42 42 50 68 83 77 91 -6 15 x 10-6mg 59 54 73 62 80 73 80 77 91 96 105 107 -6 60 x 10 mag 75 53 85 71 98 82 115 83 118 86 S 104 0.24 x 10 mag 31 22 36 44 53 46 56 47 62 50 62 50 s -60.96 x 10 *g 33 42 43 49 49 56 51 59 63 70 72 72 3.8 x 106 mg 50 33 65 35 70 80 78 87 94 108 S 147 I-615 x 10 mg 17 57 49 70 72 72 103 78 107 90 108 122 60 x 10a6,, 72 62 80 68 92 72 93 85 104 106 115 126 24 26 29 29 30 30 31 31 35 39 39 42 42 46 47 47 48 49 49 52 54 59 59 66 66 24 26 28 29 29 33 34 35 37 39 41 42 43 46 46 47 48 48 49 51 51 53 59 62 62 D 73 76 138 140 s 8 S S 8 S S S S D 136 153 168 S S .8 S S S S S S S S S S a Days. M - Male F - Female D = Animal died before date of challenge - 0012% S - Animal surviving at the end of the experiment (177 days) U - Animal unidentiflable (escaped from cage) 86 Ie PROTECTIVE EFFECT IN BANK VOLES OF SOME STRAIN OF DM APPENDIX TABLE 3 EXPERIMENT 3: INDIVIDUAL SURVIVALa OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA Copenhagen London R Madras Prague Placebo, Controls, 1331 FIO 809 725 chalned challenged m F F M | F M F U ]| F X , , , , , , ,.[ , . ,~~~~~~~~~~~~~~ 1 -60.24 x 10 mg 1 '35 50 54 92 64 96 69 96 109 121 S -6 0.96 x 10 mg 49 36 74 41 76 58 78 87 133 96 182 133 3.8 xlO mg 83 3 84 53 101 61 117 65 118 101 170 124 -6 15 x10 6mg 87 32 99 96 147 109 154 154 172 155 S 191 -6 60 x 10 mg 39 97 127 109 140 119 142 122 178 136 198 S 0.24 x 10 mg 33 53 36 56 47 58 74 64 75 69 140 77 0.96 x 10 6mg 3 44 61 58 72 62 79 67 81 68 92 79 3.8 xo06 mg 51 36 76 36 79 54 86 68 90 76 121 79 -6 15 x10 6mg 55 D 66 D 75 87 97 118 108 144 114 S I-6 60 x 10 mg 68 63 88 120 105 120 107 132 126 144 140 151 1-6 0.12 x 10 6mg 46 58 53 60 54 61 57 65 57 68 86 76 0.48 x 10 mg 11 12 46 39 60 49 66 53 87 80 90 119 1.9 x 45 55 62 77 81 81 7.5 x 36 53 63 -6 .0 mug 40 44 65 68 70 71 -6 11 38 53 74 53 76 92 118 118 I-6 30 x 10 mg 36 D 47 50 53 57 55 57 56 116 164 125 -6 120 x 10 6mg 61 1 68 71 80 79 98 85 101 119 108 176 a Days. M - Male F - Female D - Animal died before date of challenge 1 -60.24 x 10 mg 1 47 75 70 81 126 86 127 109 158 153 165 0.96 x 10 6mg 85 78 94 109 126 110 129 110 158 176 165 S 3.8 x 0 6mg D 72 56 118 101 123 119 127 124 137 156 S 15 x 10-6 112 82 118 107 132 121 133 137 165 151 S 200 60 x10 6mg 86 54 130 86 134 109 159 120 S 170 S 194 1 -60.24 x 10 tg 1 43 40 53 54 55 57 63 64 63 65 71 -6 0.96 x 10 mg D 1 47 56 60 57 67 72 73 73 76 92 3.38 x 106 mg39 D .59 40 66 62 76 66 92 96 129 99 15 x 10 6 mg 6 41 68 42 72 87 83 104 109 116 150 119 60 x 42 94 97 116 120 S L -6 54 115 123 124 126 179. 38 43 43 47 48 49 53 54 57 57 57 58 62 62 62 63 64 64 66 68 69 74 76 81 110 D 27 36 45 49 -54 57 57 59 59 60 60 68 73 73 73 75 77 77 78 82 88 90 96 104 D 60 78 S S S S S S S S S' S S S 187 196 S S S S S S S S S S S S S wJu U!ZM S = Animal surviving at the end of the experiment (207 days) U - Animal unidentiflable (escaped from cage) 87 88 A. LADEPOGED, K. BUNCH-CHRISTENSEN & J. GULD APPENDIX TABLE 4 EXPERIMENT 4: INDIVIDUAL SURVIVAL a OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA Copenhagen London R Rio de Gothenburg Placebo, Controls, 1331 F10 Janeiro challenged challenged u pF 1 F u F F 1{ F F 0.24 X 6 Illg D 42 26 46 52 46 121 91 160 171 S 179 -6 0.96 x 107mg D D D D 95 D 95 45 102 53 S 121 -6 3.8 xl s6g 31 D 75 D 96 116 108 152 128 152 149 159 -6 lSxl10 sag D D D D 45 2 47 83 95 104 131 145 -6 60 x 10 mg 46 D 71 45 102 67 106 112 121 136 168 180 1 6 0.48 x 10 smg D 61 27 66 33 72 37 76 38 76 76 82 1.9 x 10 mg 30 D 35 D 36 36 47 58 71 63 72 109 7.5 x 106 sg D 58 49 61 59 84 69 120 116 125 118 S 3 -6 30 X. g D D D 33 104 146 120 x D 109 127 136 142 156 44 86 105 129 150 171 -6 10 sag D 104 108 113 122 137 a Days. M - Male F - Female D - Animal died before date of challenge -6 0.12 x 10 mg D D 35 41 35 42 47 53 62 67 S 71 0-6 0.48 x 10 mlg D D 41 D 45 D 51 41 53 47 126 57 1.9 x 1O sg 20 D 25 D 34 D 38 38 59 61 65 103 7.5 x 106 sg D D 42 D 47 48 59 54 83 59 101 116 -6 3010-6sg D D D 46 45 82 67 .142 129 153 138 182 120 x10 g6g 21 D 44 33 105 101 142 102 8 129 134 1-6 0.24 x 10 mg 22 D 34 D 99 48 100 81 114 93 S S -6 0.96 x 10 mg D D D D 71 D 100 34 138 47 172 127 -6 3.8 x 10 mg D D 48 D 69 63 76 131 144 163 159 S -6 15 x 10 mg D 84 D 102 138 113 148 119 151 172 183 S -6 60x10 mg D U D D 37 15S 125 164 134 168 138 S 0.06 x 106 mg 27 D 38 22 41 48 67 99 68 102 101 136 0.24 x 1o6 mg D D D D 45 52 109 66 135 122 149 172 -6 0.96 x 10 mg D D 18 46 39 126 56 137 106 151 163 180 3.8 x 106 ug 45 D 47 46 74 89 83 111 109 135 119 S 15 x 10 mg D 71 D 119 141 155 142 162 172 183 S S D D D D D D D 22 29 31 34 34 35 39 43 48 57 57 58 58 61 67 70 79 84 D D D D D 7 20 34 37 43 44 46 54 54 55 55 56 58 64 68 71 77 89 90 124 D D D D 169 183 S S S S S S S S S S J* I I____ D D D 48 120 S S S S S S S S S S S S S S WHO 00726 S - Animal surviving at the end of the experiment (184 days) U - Animal unidentifiable (escaped from cage) 89PROTECTIVE EFFECT IN BANK VOLES OF SOME STRAINS OF BCO APPENDIX TABLE 5 EXPERIMENT 5: INDIVIDUAL SURVIVAL a OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA Copenhagen Madras Prague Rio de Gothenburg Placebo, non- 1331 809 725 Janeiro challenged cha1engon ._1. 1 F_ I F Fr F MI F M F 1 -60.12 x 10 mg 47 43 49 55 56 65 58 71 S 80 8 -610 mg 50 58 66 S S S 1-610 ug 60 80 S S S S -6 10 mg S S S S S S -6 10 mg 88 S .SS S I-6 0.03 x 10 mg 11 44 47 47 53 53 8 53 S 68 77 0.12 x 10 g6mg 58 37 75 57 S 69 S 80 S 87 S S 0.48 x 10 mg 20 47 S 79 S 86 S 89 S S S S 1.9 x 106 mg 92 10 S S S S S S S S S S 7.5 x 106 mg 86 S S S S S S S S S S S I -6 0.48 x 10 ug D 52 36 60 38 60 83 75 8 88 S 93 -6 1.9 x 10 mg 10 51 60 62 66 73 69 S 77 S 89 S 7. 5 x 10 6 mg 29 60 35 64 62 70 64 79 60 81 82 S -6 30 x 10 ug 51 D 87 61 90 81 S S S S S S -6 120 x 10 mg 85 88 S 91 S 92 S S S S S S 0.24 44 52 63 S8 s 0.96 21 S S SSS 3.' D 83 S S S S 15 x 85 S S S S S 60 x 46 S S S S S x 106 mg 49 61 S 8 8 x 10-6 g 42 , S , S , S S -6 U I 51 , S S , S -6 10 ug 59 S S S S S -6 10 ug S S S S S S 0..06 x 26 31 55 S5 S -6 L0 ug 38 44 55 78 S S '-6 0.24 x 10 ug 76 51 83 86 8 89 s 93 S S S S 0.96 x 10 mg 84 83 s S S S S S S S S S 3.8 x 10-6 67 S S s S S S S S S S S -6 15 x 10 mg D 0 64 S 91 S S S S S S S D 17 30 32 35 39 40 45 46 47 51 53 54 56 61 61 62 63 66 66 68 70 71 73 81 U 29 47 50 60 61 64 66 67 67 68 71 72 73 74 75 76 78 81 81 83 92 S S S D 2 6 35 63 75 83 93 S S S S S S S S S S S S S S S S S S a Days. 51 93 S S S S S S S S S S S S S S S S S S S S S WHO 0072? M - Male F - Female D - Animal died before date of challenge S - Animal surviving at the end of the experiment (93 days) U Animal unIdentiflable (escaped from cage) 0.48 x 36 45 55 84 S S 1.9 x 66 S S 8S 7.5 x 82 88 S. S S S 30 x 83 84 S S S 8 . LADEFOGED, K. BUNCH-CHRISTENSEN & J. GULD APPENDIX TABLE 6 EXPERIMENT 6: [INDIVIDUAL SURVIVAL a OF BANK VOLES AFTER CHALLENGE WITH A STRAIN OF VIRULENT MYCO BACTERIA Copenhagen London R Rio de Gothenburg Placebo, Controls, 1331F10 Janeiro challlenged chal .nged _ _ _ _ ! X F _ _ _ _ _ _ _ _ _ J NJ FF __!!! 1-6 0.24 x 10 mg 10 50 14 57 23 63 50 70 54 81 60 100 -60.96 x 10 mg 27 43 57 53 77 68 130 83 191 156 193 161 -6 3.8 x lO mg 4 26 27 66 44 69 118 96 131 128 134 134 15 x 10 mg 85 97 102 124 130 164 -6 :10 mg 22 75 96 103 183 192 I 6 0. 48 x 10 mg 0 26 25 39 35 49 42 49 43 64 75 68 -6 1.9 x 1 m 34 35 44 35 54 51 72 84 106 110 114 142 7.5 x 16 ug 39 D 81 26 86 26 94 94 124 109 151 112 -6 30 x 10 mg 25 32 84 104 108 138 120 x 10 25 49 110 120 169 D 39 45 109 119 139 -6 10 ug 26 113 129 135 146 167 1 -60.48 x 10 mg D D 28 35 39 57 52 62 56 73 58 76 -6 1.9 x 7 mg 47 46 94 70 109 131 127 161 170 165 S 181 -6 7.5 x mg 18 1 67 93 157 101 163 106 225 117 S 194 30 x 10-6 mg 41 16 103 37 108 65 110 133 160 176 210 S 120 x D 36 74 124 131 131 -6 10 ug 82 90 139 154 -168 187 1-6 0.24 x 10 mg 41 6 69 25 90 37 137 97 160 101 S 142 0.96 x o6 mg 100 D 126 D 136 105 182 130 214 147 S 175 3.8 x 106 mg 36 1 120 113 122 148 133 154 149 170 152 177 -6 15 x 10 mg D 70 60 78 83 103 101 135 135 187 140 199 60 x D D 54 140 146 157 lo- g 3 10 152 155 171 175 I..6 0.06 x 10 mg 13 47 24 50 34 63 44 66 47 67 64 70 0.24 x 10ic mg 25 D 53 55 54 60 80 66 150 72 223 106 0.96 x 10 mg 48 '64 51 74 95 97 99 99 145 119 180 179 -6 3.8 x 10 mg 26 26 37 78 51 95 76 126 110 131 131 169 I-6 15 x 106 mg 11 76 27 91 36 95 86 128 122 173 206 204 D D 1 25 26 26 38 43 44 47 48 48 49 52 54 56 57 65 68 68 69 71 84 104 151 25 26 26 30 31 45 48 48 49 51 .51 54 55 57 59 62 65 66 68 68 70 76 81 81 99 D D D D 7 16 25 26 27 34 54 65 82 90 S S S S S S S S S S S S S S S S S S S S S S S D 20 97 8 8 S S S 8 a Days. M = Male F - Female D - Animal died before date of challenge S = Animal surviving at the end of the experiment (227 days) U = Animal unidentifiable (escaped from cage) 90 27 48 115 124 138 S 60 x 25 57 99 102 132 177 wm W fz83