iiitjf'fu: ‘9.’ c‘ :07]. .v . 0 ' ‘ ""‘\g {'13.}; IZ-41 ‘,.~J;.s"‘ )‘Zr'_‘.-‘-,‘)"3.--’ . ' -- . ZI‘ "‘I‘I‘v I,1;ie‘:u I“. , L .348325: II-H;t.*£?I-t~.tac r 12.21": A LIVE STREPTOMYCIN-DEPENDENT PAST EURELIA . MULTOCIDA VACCINE FOR THE PREVENTION OF HEMORRHAGIC SEPTICEMIA Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY BETTY DONG WEI 1977 v o a. -- —-- ~" 0.... ‘ - .4...'-.‘.-.C.'_'-..l_'12_!’ ”mm ABSTRACT A LIVE STREPTOMYCIN-DEPENDENT PASTEURELLA MULTUCIDA VACCINE FOR.THE PREVENTION OF HEMDRRHAGIC SEPTICEMIA By Betty Dong wei A type B Pasteurella multocida was used for the development of a streptomycin—dependent (StrD) vaccine. .P. multocidh R9473, a hemor- rhagic septicemia strain, was mutagenized with N-methyl-N'-nitro-N- nitrosoguanidine to increase the likelihood of encountering a strepto- mycin-dependent mutant and plated on agar containing 400 ug/ml of strep- tomycin. Replica plating was used to differentiate dependent from re- sistant colonies. Mice and rabbits were vaccinated with a StrD mutant and challenged along with unvaccinated controls 21 days later with the wild type R9473. Protection of greater than 4 logs was shown for the vaccinated mice. All vaccinated rabbits were protected and all unvac- cinated controls succumbed to a challenge of 500 or 1000 LDSO' High mortality in clinical cases of hemorrhagic septicemia creates severe economic losses especially in southeast Asia and Africa. A.live preparation might prove more immunogenic than some of the killed vac- cines presently in use. Hemorrhagic septicemia is only one of a wide range of diseases caused by P. multocidh. StrD organisms may be of use as vaccines for other pasteurelloses such as rabbit "snuffles", fowl cholera, and pneumonic bovine pasteurellosis. A LIVE STREPTOMYCIN-DEPENDENT PASTEURELLA MULTUCIDA VACCINE FOR THE PREVENTION OF NEMORRHACIC SEPTICEMIA BY Betty Dong Wei A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1977 Dedicated to my father and mother Mr 8 Mrs Yen 0. Dong and to my husband Bill 11 who sta suggest peciall critica R. J. M mittee. for her E. Sand. Special laborat. Dr. M. I VaCCinal ACKNOWLEDGEMENTS I wish to express my sincere appreciation to Dr. G. R. Carter, who started my interest in streptomycin-dependence, for his helpful suggestions and guidance throughout my graduate studies. I would es- pecially like to thank Dr. R. R. Brubaker and Dr. L. Snyder for their critical suggestions in the course of my research work and to Drs. R. J. Moon and M. J. Patterson for kindly serving on the advisory com- mittee. Grateful acknowledgement is also made to Mrs. Dorothy Boettger for her efforts in securing necessary supplies and media and to Drs. E. Sanders and C. H. Coy for their donations of experimental animals. Special thanks to Mr. A. Wayne Roberts for his assistance in the I laboratory and for proofreading the manuscript. I would like to thank Dr. M. M. Chengappa and Mr. Amar Gupta for their assistance in the vaccination studies. iii TABLE OF CONTENTS Page REVIEW OF THE LITERATURE . . . . . . . . . . . . . . . . . . l HEMORRHAGIC SEPTICEMIA. . . . . . . . . . . . . . . . . 1 History. . . . . . . . . . . . . . . . . . . . . . l The Disease and Its Epidemiology . . . . . . . . . 2 The Microorganism. . . . . . . . . . . . . . . . . 4 Clinical Signs and Lesions . . . . . . . . . . . . S The Antigens of Pasteurella multocidh. . . . . . . 6 Typing of Pasteurella multocidh. . . . . . . . . . 9 Mechanism of Immunity. . . . . . . . . . . . . . . ll Vaccines Presently Used Against Hemorrhagic Septicemia . . . . . . . . . . . . . . . . . . . . ll STREPTOMYCIN. . . . . . . . . . . . . . . . . . . . . . l4 Streptomycin and Conformation Changes in the Ribosome O O I O O O O O C ' O O O O O O O O O O O O 16 The Lethal Effect of Streptomycin. . . . . . . . . 17 Point Mutations as a Basis for Resistance and Dependence O O O O O O O O O O O O O O O O O O O O 18 Mutations That Mask the Dependent Phenotype. . . . l9 LITERATURE CITED . . . . . . . . . . . . . . . . . . . iv ARTICLE: A LIVE STREPTOMYCIN-DEPENDENT PASTEURELLA MULTOCIDA Page VACCINE FOR THE PREVENTION OF HEMORRHACIC SEPTICEMIA 27 SUMMARY . . . - . . INTRODUCTION. . . . MATERIALS AND METHODS RESULTS , . . . . DISCUSSION. . . . . LITERATURE CITED. . Q 0 27 28 29 33 41 43 Table LIST OF TABLES Intraperitoneal Vaccination and Challenge . . . . Subcutaneous Vaccination and Challenge — Trial 1. Subcutaneous Vaccination and Challenge - Trial 2. Protection Afforded by StrD Vaccine . . . . . . . Subcutaneous Vaccination and Challenge of Rabbits vi Page 35 36 37 39 4O Pa and the descrih ganism plague tween lectiv BaciZZ When 1 trodu. Scrib. Paste Agrj itj m2. HEMDRRHACIC SEPTICEMIA A Review of the Literature History Pasteurellosis was first described in cattle by Bollinger in 1875 and the causative agent was isolated by Kitt in 1885. In 1880, Pasteur described the organismtwhich caused fowl cholera. Gaffky found the or- ganism responsible for rabbit septicemia in 1881 and Loeffler for swine plague in 1886. A German pathologist, Hueppe noting similarities be- tween the diseases in various hosts and their causative organisms, col- lectively named the disease hemorrhagic septicemia and the organism Bacillus septicemiae hemorrhagiae. The disease was extended to buffalo when it was described by Create and Armanni in 1887. In 1896, Kruse in- troduced the binomial, Bacillus bovisspticus. In 1900, Ligniers de- scribed the organism and diseases more completely and used the name, Pasteurella, which had been previously suggested by Trevisan. Another name, Bacterium.multoc£dhmb came into use in 1899. Rosenbusch and Merchant's binomial of 1937, Pasteurella multocidh, has wide but not universal usage. .A review is due by the International Congress for Microbiology to clear questions about classification and nomenclature. The name, hemorrhagic septicemia was at one time extended to include bovine shipping fever, a respiratory disease of complex etiology with pasteurellas often secondary invaders to Phrainfluenza-S virus (46). The current definition of hemorrhagic septicemia decided by the Food and Agriculture Organization of the United Nations in 1962 (7) states that it is an acute disease of cattle and buffalo caused by type B P. multocidh with high mortality in clinical cases. A.hemorrhagic l ax septicemia second se' Hemi north, c has not and its Epizoot Yellows Moiese , affectE 80“ 9 WI an inc in DOC OfLEn ture‘ regie Q1im' the ‘ Smal medj bre‘. 2 septicemia isolant from Central Africa was found by Carter (12) to be a second serotype, (type E). The Disease and Its Epidemiology Hemorrhagic septicemia (H—S) occurs in southern Europe, the USSR, north, central and east Africa, the Near East, and southern Asia. 'It has not been reported in South America, Australia, south Africa or Japan, and its occurence in North America has been confirmed only a few times. Epizootics in American bison were reported in 1912 (27), in 1922 at Yellowstone National Park (22), and in 1965 at the National Bison Range, Moiese, Montana (26). Two groups of young Holstein-Friesian cattle were affected in Pennsylvania in 1969. Serotype 6:B, the same as in the bi- son, was isolated (32). The greatest incidence of H-S is in the rainy season, when there is an increase in work demanded of draft animals. Cattle are likely to be in poor condition after the dry weather. Nutrition and management are often changed at this time. Stress, high humidity, changes in tempera? ture, and various minor infections may precipitate H—S. Cattle in dry regions of Equatorial Africa show naturally acquired immunity but no clinical disease, so extraneous factors must be needed to precipitate the disease. Cattle and buffalo of all ages are affected. The percent of animals infected in a particular country may be small but certain areas might sustain high losses, up to seventy per- cent. Usually, losses in a village amount to one to ten percent. Imr mediate vaccination with plain broth bacterin in the midst of an out- break saves some animals. Thailand and India are countries with high yearly most As kind of trol, 3 Be have be ease, j of entI (7) in debilii Clinic; from a1 cattle can ha diseas PEICEn acquit bodieS yefirs. harbol SPTEac numbe] ditiol Ih’hen 1 due t< of Sue Since the d; 3 yearly losses of 10,000 to 50,000 cattle and buffalo. The losses in most Asian countries remain the same year after year, indicating some kind of balance being reached despite present vaccinal methods of con- trol, a balance that can only be changed with improved prophylaxis. Because of the seasonal nature of the disease, parasitic vectors have been sought but never confirmed. From the pathology of the dis- ease, it is reasonable to assume that the nasopharynx is the main route of entry of the organism, either by inhalation or ingestion. Baldrey . (7) in 1911, sprayed fodder with a culture of Pasteurella, but only the debilitated became ill. Normal animals became immune, indicating a sub— clinical infection. Large numbers of bacteria (on the order of 108) from an aerosolized culture deposited in the nose can cause infection in. cattle. The carrier animal is the source of the microorganism. Soil 1 can harbor the organism for short periods and indirectly transmit the disease but there is no permanent soil reservoir (7). Greater than ten percent of cattle and buffalo in India and southeast Asia have naturally acquired immunity to hemorrhagic septicemia as shown by a rise in anti- bodies (7). The ten percent figure has been constant for the past fifty years. An outbreak begins with a decrease in resistance in an animal harboring Pasteurella types B or E. Once infected, the organism is spread in the feces, saliva, and nasal discharges which harbor large numbers of the organism. The size of the outbreak depends on the con- dition of the animals and the persistence of Pasteurella in the-herd. When half of the herd is immunized, outbreaks can be prevented, possibly due to lessening the chance for a first case and decreasing the number cfifsusceptibles. A village affected one year is usually spared the next since immunity presumably developed the previous year. The incidence in the district though remains the same. The Microorganism Pasteurella multocida is a small, gram negative, often pleomorphic rod ranging from cocco—bacillary to filamentous forms. Nikiphorova (7) demonstrated with the electron microscope a capsule on the type B organ- ism that is 0.5 um thick, smaller than some other Pasteurella types. Tryptose agar with carbohydrates show iridescent colonies 1 mm in dis- meter at 24 hours at 37 C. Strains dissociate to duller, smaller "blue" variants that are non capsulated. Carter (11) proposed S (smooth), SR (intermediate), and R (rough) as the main colonial types. Type I (B) does not form mucoid colonies. Staining of the fresh isolates reveal bipolar staining with methylene blue. Different strains of P. multocidh show great variations in their ability to ferment carbohydrates. In general, glucose, sucrose, levu- lose, saccharose, mannose, galactose, mannitol and fructose are ferment- ed by most strains. Lactose, maltose, trehalose, rhamnose, inositol, salicin, and inulin are usually not attacked. Some strains of P. multocida isolated from dogs and cats do ferment maltose and on oc- casion, lactose. Variable utilization is observed with xylose, arabi- nose, dulcitol, sorbitol, raffinose, dextrin, and glycerol (28,51). 0r- ganisms in the genus Phateurella characteristically are catalase posi- tive, reduce nitrates to nitrites, are oxidase positive and gelatinase negative. Ornithine decarboxylase is usually produced. Arginine di- hydrolase and lysine and glutamic acid decarboxylases are absent. Mal- onate and citrate are not utilized. Motility, urease production, and growth on MacConkey Agar are negative. Indole production is positive. Urease activity has been reported in canine and feline strains. Results of the examination of 1,268 isolates from various animal hosts show no 4 5 correlation between host preference and biochemical reactions (28). Clinical Signs and Lesions The mortality rate, once the clinical disease is established is 100 percent. Cattle show more variable pathologic signs than buffalo be— cause they tend to be affected less acutely and live longer. More pneu— ‘ monic lungs are found in cattle than in buffalo. The symptoms are fever, salivation, nasal discharge, dullness, respiratory distress, and pros- tration. Bacteremia can be demonstarted in six to twelve hours after experimental infection and saliva often is positive culturally at this time. The initial bacteremia lasts a short time due to the filtering action of the liver and spleen where the organism continues to multiply. A second bacteremia occurs and counts of 1 x 106 organisms per ml have been found (17). Blood fibrinogen increases and venous pressure de- creases. Pusteurella can be cultured from feces, urine, and milk. The terminal condition is reminiscent of and indeed may be endotoxic shock. Edema of the ventral neck and often the forelegs is observed. The tongue is swollen and protruding. In atypical cases, throat edema is not ob- served. After death, edema of the glottis, perilaryngeal, and peritra- ‘cheal tissues is found. Punctate hemorrhages may be seen in those tis- sues but the edema fluid, which is loaded with Pasteurella remains clear or slightly tinged with blood. Petechial hemorrhages are found in the auricles and under the serous membranes throughout the body. Lungs are often congested with thickening of the interlobular septa. Lymph nodes of the thoracic and peritoneal regions are congested and sometimes hemorrhagic. Early signs of peritonitis occur and calves may show 6 hemorrhagic gastritis or gastro—enteritis. The spleen is often un— changed except for occasional small hemorrhages. Intravenous injection of type B lipopolysaccharide gives severe blood stained diarrhea. The LPS is thought to be the cause of the inr testinal lesions and signs. The toxic changes are usual for a septice— mia (45). The hemorrhages, despite the name, are not spectacular. -The Antigens of Pasteurella multocida The antigenic components of Pasteurella multocidh, as with other gram negative bacteria, have been difficult to elucidate. There are difficulties in getting pure macro-molecular fractions to work with and risks of discarding important components in extraction and purification procedures. Very small quantities of antigens can stimulate an antibody response in vivo. The cell wall and capsule form a continuum.and anti- gens are distributed without notion of clear cut boundaries between cellular layers. The idea of somatic versus capsular antigens is some- what arbitrary and expresses quantitative differences in antigen derived ' from.various extraction procedures. Successive extractions with saline yield mixtures of antigens. Amounts of bound protein-polysaccharide and polysaccharide decrease as the numbers of extractions increase while amounts of lipopolysaccharide remain relatively constant. Saline ex- tracts (2.5 2) of Optimally antigenic (Phase I) Pbsteurella multocida yield 12 lines of precipitation on Ouchterlony plates (7,17). Other strains, especially if subcultured, yield less than 12 lines. It is possible that some of the precipitin lines are artifacts of fractiona- tion or represent enzyme proteins. Of these 12 lines, 2 are thought to 7 be polysaccharide, 1 lipopolysaccharide, and 9 are proteins or protein complexes (7). Sodium chloride extraction techniques yielded polysac- charides of high nitrogen content thought to be mucopolysaccharide and was represented by 1 precipitin band. It is thought that the other band represents a polymer of fructose and other sugars. No single homogeneous protein could be found which could account for the protection conferred by active immunization. The lipopolysaccharide fraction is constitutionally and biological- ly similar to that of other gram negative bacteria (5,6,35). It is pre- sumed that the LPS makes up a large portion of the cell wall as in other gram negatives. Lipopolysaccharide can be isolated by the phenol-water method of Westphal, but can be found released at all steps of extraction with 2.5 Z NaCl. The administration of either whole culture or endotoxin fractions caused similar, widespread vascular lesions (45). Widely distributed hemorrhages, edema and general hyperemia were the most obvious changes, ‘with pneumonia a constant finding. Toxins of Pasteurella have been ‘known since Pasteur's time. Baldrey in 1907 (7) discovered that fil- trates from old cultures were lethal for rabbits. Bain found that filtrates of infected sera produced fever in rabbits and mice suggesting the presence of lipopolysaccharide (7). Presently, there is controversy .as.to the presence of an exotoxin. Dhanda (15) isolated from type B Mukteswar 52 strain a toxic protein which comprised one percent of the cellular dry weight, and was inactivated at 56 C for 30 minutes. How- ever, Bain has not been able to obtain such a toxin from the Mukteswar or other strains (7). Two major antigenic components, capsular antigen and endotoxin have been identified in saline and phenol-water extracts (40,41). Indirect 8 evidence, obtained by adsorption with specific antigens of mouse pro- tective antibodies from bovine antidwhole cell serum against type B in— dicate that the type B capsular antigen played a part in protection (41). Capsular antigen, free of endotoxin was prepared by fractional precipi- tation from aqueous solution by addition of polar organic solvents. Purity from endotoxin was determined by rabbit pyrogenicity and chick embryo lethality tests. The capsular antigen is a high molecular weight acidic polysaccharide. Polyacrylamide gel electrophoresis showed some molecular weight heterogeneity so criteria of purity, as distinct from 'freedom from endotoxin, was difficult to establish. The antigen was not protein since it was heat stable, resistant to pronase, unable to stain with protein stains in polyacrylamide gel and had low light absorption at 280 nm. Mouse protection tests showed that adsorption of antidwhole cell bovine serum by the type B capsular antigen eliminated protection. However, this does not eliminate the possibility that there exist other protective antigans, since the antiserum may not have contained suffi- ciently high levels of antibodies against these to confer protection. Carter and Annau (9), and Bain (2) found fresh encapsulated isolates to be better antigens. Other workers, Bain (3), Dhanda (16), and Knox and Bain (31) described capsular antigens associated with protein components as having a protective role. Cell walls have been found to induce more protection than cytoplasm or culture filtrates (53). A heat stable, particulate lipopolysaccharide—protein complex iso- lated from encapsulated P. multocidh by extraction with formalinized saline was found to be antigenic (1,25,44). Injection of milligram amounts into mice, rabbits, and calves produced toxic reactions and fre- quently death (44). Survivors demonstrated high immunity to challenge without toxicity. The LPS-protein antigen resembled endotoxin 9 preparations in chemical Composition, heat stability. and toxicity but was more immunogenic than LPS isolated from the same strains by the West- phal procedure (44). It cannot be overlooked that the active immunity of the LPS-protein complex could be due to traces of nontoxic, specific, - capsular antigen. Typing of Pasteurella multocida . Various methods have been used in typing strains of P. multocida. The mouse protection test of Roberts (47) and the indirect hemagglutina- tion test of Carter (13) have been most commonly used. Roberts' types I, II, III, and IV have been correlated with Carter's types B, A, C, and D. A fifth group, type E (12), has been confirmed as separate from B in reciprocal serum protection tests in mice (14). Type C, once thought to be prevalent in the nasopharynx of dogs and cats, has since been dropped. Types B and E are associated with hemorrhagic septicemia. type B occur- ring in southeast Asia and E occurring in central Africa. The antigen typed by Carter's method is the polysaccharide component of the capsule which is adsorbed to the red blood cells in the indirect hemagglutina- tion test. The substance responsible for the specificity of the IRA test is thought to be carbohydrate or polysaccharide in nature because proteins are not adsorbed to unmodified erthrocytes (10). It was observed by Carter that many type A strains possesed large capsules of hyaluronic acid. Bain noted that the type II, III, and IV of Roberts also had large hyaluronic acid capsules. These immunologic types of Roberts paralleled occurrences of differing lipopolysaccharide substances or somatic "0" antigens. There is now reason to believe that . 10 Roberts' II, III, and IV are type A strains. Types A and D occur most widely geographically. Namioka (37) found that some type A strains differed in pathogeni— city for chickens in causing fowl cholera. Some strains killed three month old chickens in 48 hours in small doses while other strains did not kill even at doses of two million organisms. Therefore, as far as chicken-pathogenicity was concerned, there were two types of group A. Namioka went on to investigate the somatic or "0" antigen (38). The somatic antigen was derived from 1 N HCl treatment. It is thought to be LPS combined with protein in nature. Namioka found 12 different "0" groups from studies of 50 strains of P. multocidh. Along with Carter's capsular antigens,.this represents 15 serotypes: 6 somatic antigens for type A, 2 for type B, 6 for type D, and 1 for type E. The P. multocida group B responsible for the etiology of hemor- rhagic septicemia has only one "0" group, somatic group 6. Bain (4) also reported an organism isolated from an Australian cattle wound that belongs to type B (Australian strain 989). This group B organism did not cause hemorrhagic septicemia in experimental cattle. It had no sero— logical or biochemical differences in its capsule when compared with other type B strains. No protection was afforded in mice against type 6:B when the animals were vaccinated first with strain 989. Namioka found a new somatic antigen for 989, group 11. The central African H-S strain type B was determined to be "0" 6. Partial cross protection oc- curs in mice between type B and E. The Asian strain protects against challenge with the African strain but not vice versa. Prince and Smith found 20 different antigens by means of gel pre- cipitation and immunoelectrophoresis (42). They concluded that the most important antigens relating to serotype and protection were in 11 their terminology, the 5 fraction (capsule) and the X antigen (LPS) which corresponded to Carter's capsular and Namioka's somatic antigens. Mechanism of Immunity 'The precise mechanism of immunity is unknown. Passive transfer studies indicate that immunity is largely humoral (7,60). The most sig- nificant antibodies are antibacterial. Bactericidal antibodies have not been conclusively demonstrated in H-S. The antibody role could be op- sonization with subsequent phagocytosis. Phagocytosis is increased in the immune animal, but the fate of phagocytosed cells have not been in- vestigated. In challenge experiments, survival is proportional to the presence of circulating antibodies as determined by IHA and mouse pro- tection tests. Prince (42) observed that immune animals had circulating antibodies to «,B , or 3’ antigens. Dhanda (16) showed good correlation between circulating antibody and immunity. Vaccines Presently Used Against Hemorrhagic Speticemia (7) Plain Bacterin. Plain broth bacterin grown from.solid or broth me— dia initiate immunity quickly, in five days, so its use has merits in the face of an outbreak. However, immunity rarely lasts over six weeks. Im- munity is low grade as demonstrated in an epizootic in Malaysia in cattle vaccinated just three weeks previously (55). The density of bacterin should be 0.15 grams dry weight per liter. A 10 ml dose contains 1.5 mg Pasteurella. Formalin is used as the killing agent. 12 Delpy's Vaccine. Delpy's vaccine is presently being used only in Iran. A suitably encapsulated strain is chosen that sufficiently lyses in distilled water and merthiolate. Saponin (0.5 2) is added after 50 Z clearing. A 2 ml dose contains 2 mg of bacteria and produces an edema- tous swelling of 12 to 20 cm at the site of subcutaneous injection. The disadvantages of this vaccine is that it causes an unpopular swelling and must be agar grown, which makes harvesting difficult. Alum Precipitated Vaccine. This vaccine has been used in the Philippines, southern India, central Africa, and the USSR. Ten percent or hot twenty percent potash alum is added to formalinized, aerated bacterial broth suspension to give 1 Z alum in the vaccine. The pH is adjusted to 6.5 to obtain maximum flocculation. The route of vaccination is subcutaneous and dosage is the same as that for the plain broth bac- terin. A 2 mg dry weight dose affords immunity for up to five months. However, some tissue reactions to the vaccine occur and the preparation is not free from causing shock. Oil Adjuvant Vaccine. A 2 mg dose provides good immunity for one year. Liquid paraffin and lanolin are added as emulsifiers. Two mg of Pasteurella are added to a mineral oil emulsion. Lanolin was initially used as a stabilizing agent and to decrease the toxicity of the vaccine, but a decrease in adsorption was noted. The oil adjuvant depot at the site of inoculation serves as a prolonged source of antigen. In 1962, the Food and Agriculture Organization recommended the use of killed cul- tures with adjuvants. Live Attenuated Vaccines. Oreste and Armani in 1887 attenuated cul- tures by growing them at 32 C and also by passages through pigeons. In 1954, Hudson used decapsulated "blue" variants to immunize cattle based on a decrease in pathogenicity in mice. A few immunization trials in 13 Thailand (55) showed that immunity in cattle lasted several months but no controlled studies were done. With the initiation or improvement of lyophilization techniques in countries where hemorrhagic septicemia is endemic, live vaccines may eventually be the vaccine of Choice. STREPTOMYCIN A Review of the Literature Streptomycin was discovered in the laboratory of Waksman in 1944 (48) as a fermentation product of Streptomyces griseus. Dihydrostrep- tomycin, a chemical derivative, is nearly as active as the natural anti- biotic but is less toxic. The principle conclusion reached on strep- tomycin's mode of action is that it is,a specific inhibitor of protein biosynthesis at the ribosome level (8,18,19,21,23,34,36,49,52,57). Spotts and Stanier (52) proposed a unitary hypothesis for the action of streptomycin, postulating that the ribosome is the site of action of the antibiotic and that it is also the site responsible for phenotypic ex- pressions of susceptibility, resistance and dependence (23,29,57,59). In E. coli, susceptibility, resistance and dependence to streptomycin are expressions of multiple alleles of a single genetic locus (52). Streptomycin dependence (StrD) is the third alternative state governed by a triple allelic gene (43). Since ribosomes can be separated into their 308 and 508 subunits and reconstituted into active 705 particles, hybridization studies of subunits from streptomycin sensitive (Strs) and resistant (StrR) E. coli showed that the site of streptomycin action is on the 308 subunit (8,19, 21,33). Streptomycin binds specifically to the 308 subunit (18,43). The attachment site is generated during the associatibn of the entire - 308 particle (8). Equally bactericidal dihydrostreptomycin binds at the same site but with lower affinity. After the addition of streptomycin, the rate of protein biosynthesis declines to zero within 1/4 to 1/3 of the normal doubling time (23). In cell free systems, streptomycin 14 15 inhibits the incorporation of radioactive amino acids into proteins while amino-acyl transfer RNA accumulates (23). Sensitive or resistant E. coli are correspondingly sensitive or resistant to inhibitions of incorpora- tion of phenylalanine (23). Phenylalanine incorporation studies using synthetic polynucleotides (poly-U) have shown that the 308 ribosomal sub- ‘unit is involved (19,36). The binding of phenylalanine increases in the presence of streptomycin when streptomycinvdependent 30$ subunits are reconstituted with SOS subunits from either StrD or StrS cells (19). Cell free extracts remain susceptible to streptomycin if 308 subunits from Strs cells were used regardless of origin of the SOS subunits. Furthermore,~the 308 subunit can be dissociated into its 168 ribonucleic acids and 20 proteins (23). One of these proteins P10 (nomenclature of Ozaki,39) or $12 (nomenclature of Wittman,59) determines an organism's response to streptomycin. 812 is responsible for streptomycin sensi- tivity . Only mutations in that specific protein account for resistance, or dependence phenotypes (18,29,34). The StrA locus, mapping between argG and malA in E. coli, determines the protein product 812 (39,43). Chromosome mediated streptomycin resistance is governed by this gene. In vitro ribosome reconstitution studies with isolated ribosomal pro- teins showed 812 to be the determinant for the dependent phenotype (33). One StrD E. coli differed from its wild type parent by one amino acid in protein 512. Lysine was replaced by glutamine in position 42 (20,30). Studies were done to determine whether the 812 alteration was always the same. Amino acid replacements in two mutants to streptomycin dependence in 812 were found to be different from each other and from a mutant studied by Funatsu and Wittman (20). Therefore, it is likely that many amino acid replacements can lead to mutants dependent on streptomycin ' (30). Mutations can arise at different but closely linked loci of the 16 same gene, Biochemical analysis of ribosomal protein 812 from three StrR alleles showed that a single amino acid exchange correlated with .each mutant type (20). A similar approach can be used with the strep- tomycin dependence. Different loci can be affected so dependent mutants may ultimately have a range of variation in phenotype including differ- ing abilities to revert and differing concentration requirements for streptomycin (34). Streptomycin and Conformation Changes in the Ribosome It has been proposed that the effects of streptomycin are due to distortions of ribosomal binding sites for t~RNA (36). Ribosomes are inactivated by low Mg++ concentration and reactivated with heat and pro- per ionic conditions. Streptomycin hinders both inactivation and re- activation of ribosomes and there are indications that changes in ribo- somal conformation are involved in these interconversions (36). The 308 subunit from StrS E. coli is able to bind two molecules of streptomycin (8). Streptomycin resistant ribosomes only bind five percent of the amount of streptomycin as sensitive cells. When streptomycin is bound, the ribosome cannot bind phenylalanine-tRNA. This binding further changes the conformation of the ribosome so that aminoeacyl tRNA binding is not efficient (36). It appears that different concentrations of streptomycin cause different conformational changes (49). Catalytic levels of streptomycin cause inhibition of phenylalanine incorporation with no effect on misreading of the ribonucleic acid code. Misreading is stimulated at the higher streptomycin to ribosome ratios. Different streptomycin concentrations can cause inhibitions of Phe-tRNA binding 17 or stabilization of this binding. The stabilizing effect may be the cause of misreading since it may allow mismatched tRNA-mRNA complexes which otherwise might be shortelived (36). The Lethal Effect of Streptomycin It has not been resolved whether the lethal effect of streptomycin is due to 1) inhibitions of chain elongation by immobilizing peptidyl tRNA on its ribosome binding site (23) or 2) inaccuracies in transla- tion. The binding phenomenon itself does not kill (8). Streptomycin induced killing may not require tight binding to ribosomes. Protein synthesis need not completely stop before cell death occurs (34). A dual mechanism of action (34,57) may be in operation with killing sep- arate from the initial lesion on protein synthesis. Dihydrostreptomycin and streptomycin have the same bactericidal potency but have different affinities for ribosomes. Different strains of bacteria bind differing amounts of streptomycin. It was once proposed that high levels of strep- tomycin exert an irreversible effect on ribosome function in vivo by forming aberrant initiation complexes which permanently inactivate ri- bosomes (23,43). Formation of initiation complexes apparently takes place (57,58). .Studies using the poly-U system proved that ribosomes were slowed but not permanently inactivated (49). Only after the block on protein synthesis is partially released can stimulation of misreading begin. If all ribosomes were inactivated at initiation, there would be no chance for misreading to occur. Killing by streptomycin is antago- nized by chloramphenicol and other substances that form a stable block— ade of ribosomes in polysomes (58). 18 Streptomycin causes a cyclic blockade of ribosome initiation, blockage of chain elongation, gradual release, and reinitiation (57). This cyclical blockage accounts for the dominance of sensitivity in Strs/StrR heterozygotes. Streptomycin interferes with only a proportion of ribosomes which is enough to halt protein synthesis. Inactive StrR ribosomes regain activity when the cells are lysed and excess mRNA is provided. Streptomycin is also known to cause a gradual release of ri- bosomes from polysomes (36). Once affected, inactive ribosomes do not resume activity in extracts. However, the ribosomes are not inert, par- ticipating in continuous attachment and release from polysomes. Although no longer able to synthesize protein, the ability to attach to mRNA is retained. Reinitiated affected ribosomes carrying fMet—tRNA do not pro- ceed with chain elongation since pulse labelling studies show little va- line is incorporated as compared to methionine (57). The'StrS phenotype in heterozygotes is dominant because StrR ribosomes are deprived of the opportunity to initiate. Streptomycin susceptible ribosomes attach to mRNA for a longer time period (half life is five minutes vs 15 to 20 ri- bosomes per second in normal initiation) and successfully competes with StrR ribosomes for binding sites. Point Mutations as a Basis for Resistance and Dependence Mutations to streptomycin resistance and dependence usually arise by base substitutions (50). N-methyl-N'-nitro-N-nitrosoguanidine (NTG) produces point mutations by substituting one base for another. Acridine half mustard ICRr191 induces the addition or deletion of bases causing frame shift mutations. ICE—191 proved a poor mutagen for genes 19 affecting streptomycin resistance while MTG was 250 to 1000 times more effective (50). NTG induced approximately equal numbers of dependent and resistant mutants. That ICR9191 was a poor mutagen indicates that frame shift mutations are not allowed or are lethal. A change in a single amino acid is allowed, whereas changes in a whole sequence of amino acids in an important organelle protein must produce non-viable cells. Resistance to aminoglycoside antibiotics can be achieved in several ways, 1) Refactor genes for a variety of drug inactivating enzymes have been detected. Streptomycin and spectinomycin can be inactivated by adenylating enzymes (43,54). Phosphorylating enzymes have been dis- covered for streptomycin, paromamine, kanamycin, and neomycin (43,56). 2) An altered response to streptomycin may be due to permeability changes (34). 3) Mutations in the 812 protein genotypically alters response to streptomycin. Cell free ribosomal extracts have proven that the streptomycin-dependent phenotype is due to major genotypic and not permeability changes. Streptomycin-sensitive ribosomal extracts ceased synthesizing proteins in the presence of streptomycin while resistant cell free ribosomal extracts continued protein synthesis (52). Optimal protein synthesis was restored to cells or cell free extracts of depen- dent strains when streptomycin was added back to the medium (18). Mutations That Mask the Dependent Phenotype Genetic analysis of revertants from streptomytin-dependence (StrD) to streptomycin-independence (StrI) demonstrated that many have 20 mutations near but not coincident with the StrA locus. Using ribosomal reconstitution experiments, two dimensional gel electrophoresis, and phosphocellulose column chromatography, several laboratories have re— ported that a number of revertants have altered proteins S4 or $5 of the 308 subunit (24,29,30,33,34,59). Regardless of the StrD 812 allele, StrI mutants arise due to changes only in S4 or $5 (24). Revertants contain- ing mutations in other than protein 812 support the suggestion that these are not true revertants but mutations that suppress the dependent pheno- type. Streptomycin—dependence may be suppressed by alterations in either S4 or 85. Alterations in 54 give proteins which drastically differ in amino acid composition, molecular weight and amino acid sequence (30) which lends evidence for frame shift type mutations. Single amino acid replacements are found in 85 mutations as determined by two dimensional gel electrophoresis; base substitution is suspected as the mechanism of reversion here. A third alteration in S4 can compensate for an altera- tion in $5 in such a way that the original streptomycin-dependent pheno- type produced by 812 is expressed (59). In 190 mutants studied (59), mutations to StrD, StrI, and StrR were accounted for by alterations only in the three proteins 84, SS, and 812. A study by Hasenbank (24) showed that 40 of 100 revertants from StrD had electrophoretically altered pro- teins; 24 were altered in S4, 16 altered in SS. Deusser (24) showed in a study that 5 of 14 revertants had changes in S4 or $5, and an addition- al 3 of 14 revertants were discovered by immunologic and not electro- phoretic means since neutral amino acids were replaced by neutral amino acids. Conceivably, Hasenbank's study could have had as high as 60 to 65 out of 100 mutants with alterations only in S4 or $5 if undetected neutral substitutions were taken into account (24). In a study of StrI revertants of StrD Bacillus stearothermophilus (29), 2 of 13 and 4 of 13 21 had alterations on 84 and 55 respectively. Not all mutations of S4 and 85 occurred in the same position. The 85 proteins migrated to 5 differ- ent positions on two dimensional electrophoresis, and 84 proteins to 17 positions. The locations of S4, 85 and 812 in the assembled ribosome are quite different. 54 is an internal protein, 85, a fractional pro- tein added late in the assembly and 812 is not well defined. There was no direct correlation between the specific ribosomal pro- tein altered and the ultimate phenotype. In addition to the elimination of dependence, many differed from the parent strain and from each other in their growth rate, response to streptomycin in vivo, ability of ri- bosomes to bind streptomycin, and activity of ribosomes in cell free systems. 85 could produce the same phenotypic change as a change in S4 but some-changes in 84 could produce varying phenotypes (34). These results clearly support the concept of extensive, complex interaction among ribosomal proteins leading to vastly different structural confor- mations. LITERATURE CITED LITERATURE CITED 1. Babs, T.: Immunogenic Activity of a Ribosomal Fraction Obtain- ed From.Pasteurella multocida. Infect Immun, 15, (1977): 1—6. 2. Bain, R. V. 8.: Vaccination Against Hemorrhagic Septicemia of Bovines. Nature, 173, (1954): 584-585. 3. Bain, R. V. 8.: Studies on Hemorrhagic Septicemia of Cattle IV. A Preliminary Examination of Antigens of Pasteurella multocida Type I. Brit Vet J, 11, (1955): 492-498. 4. Bain, R. V. S.: A Note on Some Pasteurella Types Found in Aus- tralian Animals. Aus Vet J, 33, (1957): 119-120. 5. Bain, R. V. 8.: Hemorrhagic Septicemia of Cattle, Observations on Some Recent Work. Brit Vet J, 115, (1959): 365-369. 6. Bain, R. V. S. and Knox, K. W.: Antigen of Pasteurella malto- cida Type I, II. Lipopolysaccharide. Immunology, 4, (1961): 122-129. 7. Bain, R. V. S.: Hemorrhagic Septicemia. FAO Agricultural Study, Food and Agricultural Organization of the United Nations, (1963). 8. Biswas, D. R. and Gorini, L.: The Attachment Site of Strepto- mycin to the 308 Ribosomal Subunit. Proc Nat Acad Sci, USA, 69, (1972): 2141-2144. . 9. Carter, G. R. and Annau, E.: Isolation of Capsular Polysaccha- ride From Colonial variants of Pasteurella multocida. Amer J Vet Res, 14, (1953): 475-478. 10. Carter,-G. R.: Studies on Pasteurella multocida. I. A Hemagglutination Test for the Identification of Serologic Types. Amer J Vet Res, 16, (1955): 481-484. 11. Carter, G. R.: Studies on Pasteurella multocida. II. Iden- tification of Antigen and Colonial Characteristics. Amer J Vet Res, 18, (1957): 210-213. 12. Carter, G. R.: A New Serological Type of Pasteurella multo- cida From Central Africa. Vet Rec, 73, (1961): 1052. 22 23 13. Carter, G. R. and Rappay, D.: A Hemagglutination Test Employ- ing Specific Lipopolysaccharide for the Detection and Measurement of Pasteurella Antibodies to Pasteurella multocida. Brit Vet J, 119. (1963): 73-77. 14. Carter, G. R.: Immunological Differentiations of Type B and E Strains of Pasteurella multocida. Can Vet J, 4, (1963): 61-63. 15. Dhanda, M. R.: Purification and Properties of the Soluble Antigen of Pasteurella septica Type 1. Indian J Path Bact, 2, (1960): 16. Dhanda, M. R.: Studies on Immunochemistry. 1. Isolation and Characterization of the Soluble Antigens of Pasteurella septica. Indian J Vet Sci, 30, (1960): 114. 17. Dhanda, M. R. and Sen, G. P.: Hemorrhagic Septicemia and .Allied Conditions in Sheep, Goats, Pigs, and Poultry. United Indian Press. New Delhi, (1972). 18. Dixon, H. and Polgase, W. J.: Effect of Dihydrostreptomycin on Protein Synthesis in Whole Cells and in Cell-Free Extracts of a Strep- tomycin-Dependent Strain of E. coli B. J Bact, 100, (1969): 247-253. 19. Felsenfeld, 0., Stegherr-Barrios, A., Holmes, JS, and Parrott, ‘M; W.: Incorporation of Phenylalanine and Uridine by Str and StrD Vibrio cholerae Ogawa Ribosomal Particles. Res Comm in Clin Path Pharm, 1, (1970): 9-15. ' 20. Funatsu, G. and Wittman, H. G.: Ribosomal Proteins, XXXIII. Location of Amino Acid Replacements in Protein 812 Isolated From E. coli Mutants Resistant to Streptomycin. J Molec Biol, 68, (1972): 547. 21. Garvin, R. T., Rosset, R., and Gorini, L.: Ribosomal Assembly Influenced by Growth in the Presence of Streptomycin. Proc Nat Acad Sci, USA, 70, (1973): 2762-2766. 22. Gochenour, W. S.: Hemorrhagic Septicemia Studies. Amer Vet Med Assoc J, 65, (1924): 433-445. 23. Hahn, F. E.: Streptomycin. Antibiot Chemother, 17, (1971): 29-51. . 24. Hasenbank, R., Gutherie, C., Stoeffler, G., Wittman, H. G. Rosen, L., and Apirion, D.: Electrophoretic and Immunological Studies on Ribosomal Proteins of 100 Eacherichia coli Revertants From Strepto- mycin Dependence. Molec Gen Genet, 127, (1973): 1-18. 25. Heddleston, K. L., Rebers, P. A., and Ritchie, A. E.: Im- munizing and Toxic Properties of Particulate Antigens From Two Immuno- genic Types of'PasteurelZa.multocida of Avian Origin. J Immunol, 96, (1966): 124-133. 24‘ 26. Heddleston, K. L., Rhoades, K. R., and Rebers, P. A.: Experi- mental Pasteurellosis: Comparative Studies on Pasteurella multocida -From.Asia, Africa, and North America. Amer J Vet Res, 28, (1967): 1003- 1016. 27. Heddleston, K. L., and Wessman, G.: Vaccination of American Bison Against Pasteurella multocida Serotype 2 Infection (Hemorrhagic Septicemia). J Wildlife Dis, 9, (1973): 306-310. 28. Heddleston, K. L.: Physiologic Characteristics of 1,268 Cul- tures of Pasteurella multocida. Amer J Vet Res, 37, (1976): 745-747. I 29. Isono, R.: Altered 30$ Ribosomal Proteins in Streptomycin Re- sistant, Dependent and Independent Mutants of Bacillus stearothermophilus. Mblec Gen Genet, 133, (1974): 77-86. 30. Itol, T. and Wittman, H, G.: Amino Acid Replacements in Pro- tein SS and $12 of Two E. coli Revertants From StrD to StrI. Molec Gen Genet, 127,(1973):19-32. 31. Knox, K. W. and Bain, R. V. 8.: The Antigens of Pasteurella multocida Type I. I. Capsular Lipopolysaccharides. Immunology, 3, (1960): 352-363. 32. Kradel, D. C., Heddleston, K. L., Risser, J. V., and Manspeaker, J. E.: Septicemic Pasteurellosis (Hemorrhagic Septicemia) in Young Dairy Cattle. Vet Med/Small Anim Clin, 64, (1969): 145-147. 33. Kreider, G. and Brownstein, B. L.: Ribosomal Proteins Involv- ed in Suppression of Streptomycin Dependence in Eacherichia coli. J Bacteriol, 109, (1972): 780-785. 34. Kreider, G. and Brownstein, B. L.: Pleiotropic Effects Result- ing From.Mutations in Genes for Ribosomal Proteins: Analysis of Rever- tants From Streptomycin Dependence. J Mol Biol, 84, (1974): 159-171. 35. MacLennan, A. P. and Rondle, C. J. M.: Pasteurella septica: The Occurrence of Type-Specific Polysaccharides Containing Aldoheptose Sugars. Nature, 180, (1957): 1045-1046. 36. Miskin, R. and Zamir, A.: Effect of Streptomycin on Ribosome Interconversion, a Possible Basis for the Action of the Antibiotic. Nature, 238, (1972): 78-80. 37. Namioka, 8.: Antigenic Analysis of Pasteurella multocida. Nat Inst Anim Hlth Quarterly, 10, (1970):. 97-108. .38. Namioka, 8.: Serological Studies on Pasteurella multocida. Especially on "O" Antigenic Analysis of the Organism. Contrib Microbiol Immunol, 2, (1973): 177-178. 39. Ozaki, M., Mizushima, 8., and Nomura, M.: Identification and Functional Characterization of the Protein Controlled by the Streptomy- cin Resistant Locus in E. coli. Nature, 222, (1969): 333-339. 25 40. Penn, C. W. and Nagy, L. K.: Res Vet Sci, 16, (1974): 251. 41. Penn, C. W. and Nagy, L. K.: Isolation of a Protective, Non- toxic Capsular Antigen From Pasteurella multocida Type B and E. Res Vet Sci, 20, (1976): 90-96. 42. Prince, G. H. and Smith, J. E.: Antigenic Studies on Pasteur- ella multocida using Immunodiffusion Techniques. 1. Identification and Nomeclature of the Soluble Antigens of a Bovine Hemorrhagic Septicemia Strain. J Comp Path, 76, (1966): 303-321. 43. Quesnel, L. B. and Hussain, H. S. N.: Drug Dependence and Phenotypic Masking in Gram Positive and Gram Negative Bacteria. Micro- bios, 4, (1971): 33-48. 44. Rebers, P. A. and Heddleston, K. L.: Isolation From Pasteur- ella multocida of a Lipopolysaccharide Antigen With Immunizing and Toxic Properties. J Bacterol, 93, (1967): 7-14. 45. Rhoades, K. R., Heddleston, K. L., and Rebers, P. A.: Experi- mental Hemorrhagic Septicemia: Gross and Microscopic Lesions Resulting From Acute Infection and From Endotoxin Administration. Can J Comp Med, 31, (1967): 226-233. 46. Roberts, A. W. and Carter, G. R.: Essentials of Veterinary Virology. Michigan State University Press, East Lansing, Michigan, (1976): 59. 47. Roberts, R. S.: An Immunological Study of Pasteurella septica. J Comp Path Ther, 57, (1947): 261-278. 48. Schatz, A., Bergie, E., and Waksman, S.: Streptomycin, a Substance Exhibiting Antibiotic Activity Against Gram Positive and Gram Negative Bacteria. Proc Soc Exp Biol, 55, (1944): 66-69. 49. Sherman, M. 1.: Role of Ribosomal Conformation in Protein Synthesis. Further Studies With Streptomycin. Eur J Biochem, 25, (1972): 291-300. 50. Silengo, L., Schlessinger, D., Mangiarotti, C., and Apirion, D.: Induction of Mutations to Streptomycin and Spectinomycin Resistance in Escherichia coli by N-methyl-N'-nitro-N-nitrosoguanidine and Acridine Half-Mustard ICR-191. Mutation Res, 4, (1967): 701-703. 51. Smith, J. R.: Genus Pasteurella in Bergey's Manual of Deter- minative Bacteriology. 8th Ed. Edited by R. E. Buchanan, et al. Williams and Wilkins, Baltimore, Maryland, (1974). 370-373. 52. Spotts, C. R. and Stanier, R. Y.: Mechanism of Streptomycin on Bacteria: A Unitary Hypothesis. Nature, London, 192, (1961): 633- 637. 26 53. Srivastava, K. K., Foster, J. W., Darve, D. L., Brown, J., and Davis, R. B.: Immunization of Mice With Components of Pasteurella multocida. Appl Microbiol, 20, (1970): 951-956. 54. Takasawa, S., Utahara, R., Okanishi, M., Maeda, K., and Umezawa, H.: Studies on Adenylyl Streptomycin, a Product of Streptomy- cin Inactivation by E. coli Carrying the R-Factor. J Antibiot, 21, (1968): 477-484. 55. Thomas, J.: Control of Hemorrhagic Septicemia in West Malay- sia. Trop Anim Hlth Prod, 4, (1972): 95-101. 56. Umezawa, H., Okanishi, M., Kondo, 8., Banana, K., Utahara, R., Maeda, K., and Mitsuhashi, S.: Phosphorylative Inactivation of Amino- glycoside Antibiotics by E. coli Carrying R-Factor. Science, 157, (1967): 1559-1561. 57. Wallace, B. J. and Davis, B. V.: Cyclic Blockade of Initiation Sites by Streptomycin-Damaged Ribosomes in E. coli: An Explanation for Dominance of Sensitivity. J Mol Biol, 75, (1973): 377. 58. Wallace, B. J., Tai, P. C., Herzog, D.: Partial Inhibition of Polysomal Ribosomes of E. coli by Streptomycin. Proc Natl Acad Sci, USA, 70, (1973): 1234. D 59. Wittman, H. G.: Analysis of Ribosomal Proteins in StrR and Str Mutants Isolated From StrI E. coli Strains. Mulec Gen Genet, 141, (1975): 331- 341. , 60. Woolcock, J. B. and Collins, F. M.: Immune Mechanism.in Pasteurella multocida—Infected Mice. Infect Immun, 13, (1976): 949— 958. ARTICLE A LIVE STREPTOMYCIN-DEPENDENT PAST EURELLA MULTOCIDA VACCINE FOR THE PREVENTION OF HEMORRHAGIC SEPTICEMIA By B. D. Wei and G. R. Carter (Manuscript to be submitted to the Journal of the American Veterinary Medical Association) SUMMARY A type B Pasteurella multocida was used for the development of a streptomycin-dependent (StrD) vaccine. P. multocida R-473, a hemorrha- gic septicemia strain, was mutagenized with N-methyl-N'-nitro-N-nitroso- guanidine to increase the likelihood of encountering a streptomycin-de- pendent mutant and plated on agar containing 400 ug/ml of streptomycin. Replica plating was used to differentiate dependent from resistant colonies. Mice and rabbits were vaccinated with a StrD mutant and challenged along with unvaccinated controls 21 days later with the wild type R-473. Protection of greater than 4 logs was shown for the vac- cinated mice. All vaccinated rabbits were protected and all unvaccinat- ed controls succumbed to a challenge of 500 or 1000 LDSO' 27 INTRODUCTION The use of live attenuated bacteria in a vaccine against disease caused by Pasteurella multocida was first tried by Louis Pasteur. Pres- ently, killed oil adjuvant preparations are the vaccines recommended for the prevention of hemorrhagic septicemia (H-S) (2). A live, attenuated Pasteurella multocida, the Clemson University or CU strain, is available as a commercial vaccine for fowl cholera and is distributed in lyophi- lized form (3). Killed bacterial preparations often protect against homologous but not heterologous challenge (11). Administration of anti- biotics in the feed of turkeys vaccinated with the CU strain suppressed the immune response (5), proof that the organism.in a vaccine must sur- vive to some extent in the animal before optimal immunity is attained. Other workers have had success with an avirulent, high temperature mu- tant in the prevention of fowl cholera (15). The higher efficiency of live as opposed to killed bacteria as vaccines may be due to 1) the preservation of intact antigens of the virulent organism by obviating the use of harsh chemicals and physical procedures, 2) the formation of antigens produced in vivo that may have been lost after continuous subculturing, 3) the greater immunity produced by allowing some multi- plication in the animal (5,12,14). Vaccines that have shown much promise in inducing immunity, in innocuousness and in ease of administration are the live, streptomycin- dependent (StrD) preparations. Streptomycin-dependent mutants have been developed for Escherichia coli (10,14), Salmonella typhi (7,22), salmon- ella enteritidis (6,23,24), Shigella flexneri (16,17,18), and Vibrio cholera (9). 28 29 Because of asymptomatic carrier states and the acute nature of hemorrhagic septicemia, control of the disease depends on prephylaxis. With the improvement of culturing and lyophilization techniques in areas where H-S is endemic, live vaccines may be the vaccine of choice. A StrD P. multocida vaccine was developed, found to be stable against re- version, and was protective for mice and rabbits. MATERIALS AND METHODS Bacterial Strain. Pasteurella multocida strain R-473 (type 6:3) was the parent strain from which the streptomycin-dependent vaccine was derived. This isolant originally came from.a bovine hemorrhagic septi- cemia case in Egypt and had been maintained in lyophilized form in our laboratory. The culture was confirmed to be a type B by the indirect hemagglutination test (4). N-methyl-N'-nitro-N-nitrosoguanidinea (NTG) Mutagenesis. p. multo- cida R-473, a streptomycin-sensitive strain was mutagenized by the mod- ified methods of Adelberg (1), and Osechger and Berlyn (19) to obtain resistant and dependent colonies. Ten ml of a four hour logarithmic phase culture grown in tryptose broth with 0.3 2 yeast‘extractb was cen- trifuged for 10 minutes at 20,000 x g. The pellet was resuspended in .05 M Tris-maleic acid buffer, pH 6. The cells in buffer were a Aldrich Chemical 00., Milwaukee, Wisconsin b Difco Laboratories, Detroit, Michigan 30 mutagenized with a final concentration of NTG at 0.1 mg/ml for 15 min at 37 C. The suspension was centrifuged, washed once with broth, centri- fuged again and resuspended in 100 ml of tryptose broth with yeast ex- tract and incubated overnight. The next day, the culture was centri- fuged at 20,000 x g for 10 min and resuspended in only 1 ml of tryptose broth with yeast extract. Ten streptomycinc agar plates (tryptose agar (TA), 0.3 2 yeast extract, 100 ug/ml streptomycin (Str)) were inoculated with a sterile, bent glass rod. The plates were incubated for 48 hours. Selection of'StrD From StrR colonies. Each of the ten spread plates were replica plated by the method of Lederberg (13), first onto plain tryptose agar with yeast extract and then onto tryptose, yeast extract agar with 100 ug/ml of streptomycin. The replicated plates were in- cubated for 48 hours. Colonies that grew better on the TAI+ Str plates than on the plain TA were individually picked with a straight wire, one half of the colony put on another TA and the other half on TAI+ Str in order to check the dependency. Two StrD colonies #34 and #37 which grew exceptionally well on TA-+ Str but not at all on plain TA were chosen for the production of vaccine. These colonies were maintained by weekly transfers on TA.+ Str and also lyophilized. Reversion Studies. 1) Heavy loopfuls of the StrD Pasteurella were periodically streaked onto plain TA plates in order to determine revertants to streptomycin independence. 2) Overnight broth cultures of StrD R-473 were centrifuged and washed 3 times to rid the broth of c Streptomycin-sulfate, Pfizer Laboratories, New York 31 streptomycin and plated heavily on plain TA along with a known number of streptomycin-sensitive wild type R-473. The same dilution of wild type R-473 was plated alone on another set of TA plates. A significant increase in the number of colonies growing on the combined StrD and wild type plates would be interpreted as indicating the presence of revertants, yet assuring, by the growth of a known number of wild type colonies, that conditions allowed for revertants to be detected. Since the TA plates were devoid of streptomycin, no StrD colonies would be found. vaccine Preparation. The chosen StrD colonies were grown in 25 ml of tryptose broth with 0.3 2 yeast extract and 400 ug/ml streptomycin- for 24 hr at 37 C in a shaking water bath. The culture was centrifuged, washed once with saline, centrifuged again and resuspended in saline to the original volume. The vaccine was checked for purity by a gram stain, a culture on TA-+ Str agar, and by biochemical tests. For each vaccination study, the number of organisms in the washed broth was de- termined by viable cell counts of ten fold dilutions on spread plates. A non streptomycin-containing plate was always inoculated to assure that no streptomycin-independent organisms were present. Viability of'a Rehydrated vaccine at Room Temperature. A lyophi- lized vial of the StrD vaccine strain R-473 was rehydrated and grown in 25 ml of tryptose, yeast extract broth with 400 ug/ml of streptomycin. Its viability at room.temperature was checked by plating appropriate dilutions in duplicate for two weeks. 32 vaccination Studies. Swiss Webster ICR.albino miced, females weigh- ing 18 to 22 g were placed 6 in a cage and fed ad'libitum. Groups of 36 were vaccinated intraperitoneally or subcutaneously with 0.1 ml of the StrD vaccine. Groups of 6 were challenged 21 days later with 0.5 ml of ten fold dilutions of frequently mouse passaged wild type Rr473 by the same route of inoculation as the vaccination. Mice that died subsequent- ly to challenge were randomly selected and organs were cultured on plain and streptomycin incorporated tryptose agar plates. Organisms were iden- tified as Pasteurella multocida when these reactions were observed: ' slight acid production in the triple sugar iron agar slant and butt with no production of gas or H28, oxidase and indole production, reduction of nitrates to nitrites, negative lactose and maltose fermentation, nega- tive urease production, failure to grow on MacConkey Agar, and typical nonhemolytic, translucent colonies on bovine blood agar plates. Sixteen New Zealand albino rabbits, nine months old, were vac- cinated subcutaneously with 0.5 ml of vaccine. They were maintained in separate cages and fed ad Zibitum. Eight control rabbits remained un- vaccinated. After 21 days, eight vaccinated and four control rabbits were challenged with approximately 500 LD of the wild type Rr473. A 50 similar group was challenged with 1000 LDSO' The LDso was previously determined to be 34 organisms. Rabbits that succumbed to challenge were necropsied and cultured on plain TA and TA-+ Str. ' Challenge of both mice and rabbits was done with a recently mouse passaged P. multocida Rr473 grown for 8 to 10 hours in a shaking water bath at 37 C. Purity of the culture was assured by gram stain and bio- chemical tests. Viable counts were made by the spread plate technique. d Spartan Research Animals, Haslett, Michigan 33 Seven Holstein calves, aged 4 to 16 weeks were subcutaneously vac- cinated on the shoulder with 2 ml of an unwashed 24 hour StrD culture. This vaccine represented 7.6 x 108 live StrD P. multocida. The calves were kept in isolation and observed for three weeks. RESULTS Reversion Studies. No revertants to streptomycin-independence were found by heavily plating the StrD colonies used for the vaccine on plain tryptose, yeast extract agar. Occasionally, a thin haze of growth could be detected at the point of heaviest inoculum. Subculturing some of this material on another TA plate failed to show any more growth and it was thought that this growth was due to residual streptomycin trans- ferred from.the original media. It is also possible that the dependent cells had enough of the antibiotic inside the cell to allow growth for one or two generations. Heavy inocula on TA.+ Str plates showed growth of colonies 1 mm in diameter in 48 hours. A 1 x10.6 dilution of wild type, streptomycin-sensitive Rr473 was plated on 3 plain TA plates. This yielded 280, 225, and 275 colonies. The same volume (0.1 ml) of the 1 x 10.6 dilution of wild type was plat- ed simultaneously with 4.7 x 107 StrD cells on 3 other TA plates and growth of 242, 219, and 249 streptomycin-independent colonies was found. Thus, no streptomycin-independent colonies due to reversion could be de- tected. D Stability of'the Str vaccine at Room Temperature. After 24 hours incubation in 25 ml of tryptose, yeast extract broth with 400 pg 34 streptomycin, a rehydrated vial of lyophilized StrD R-473 had a viable count of 2.5 x 109 cells/ml. No growth appeared on agar without strep- tomycin. The culture was kept on a shelf at room temperature and viable .counts made 7 times in 16 days. An average of 2.5 x 108 cells were main- tained in the vaccine from day 2 to day 16. On day 16, 4.0 x 108 cells remained. Parallel platings on plain agar showed no growth. Vaccination Studies. The accumulated mortalities after challenge in the mouse vaccination studies are given in Tables 1, 2, and 3. No adverse effects due to the vaccine were observed. The intraperitoneal route of challenge infrequently caused convulsions and immediate death, 4 times in vaccinated and 3 times in unvaccinated control animals. A second administration of vaccine 14 days prior to challenge did not al- ter the degree of protection (Table l). Subcutaneous vaccinations were not as effective as intraperitoneal vaccinations when comparable dilu- tions of the challenge strain were used. Subcutaneous trial 1 gave protection of 1.9 logs. Two logs protection is the minimal requirement for commercial P. multocida vaccines according to Ose and Muenster (20). It was found that more organisms were needed to overwhelm the controls when the subcutaneous route of injection was used. Perhaps the subcutis does not foster growth as well as the rich peritoneum or there is loss of some organisms through drainage of the 0.5 ml challenge inoculum out of the mouse. When lower dilutions of challenge culture were used, the subcutaneous vaccinations protected just as well as the intraperitoneal. The LD50 for each trial was calculated by the method of Reed and Muench (21). As few as S organisms were able to kill 50 Z of unvaccinated controls while the vaccinated could survive up to 50,000 to 500,000 35 . .mHHmo.on x o.~ .N Haste .on x m.m .H Huang .Ha ~.o mo meow maauom> «« Ha\mHHuu so” x o.m..~ Hague .Ha\mHHmu on x n.~ .H Hesse .auoua «mamaamnu sausages: « a m\s N\o o\s m\o muoH «\s m\o s\o s\m m\s A m-o~ o\o o\n o\o o\o «\H “.03 o\o o\o o\o o\o «\H _o-oH o\o o\o o\n o\o «\H muoH o\o o\o o\s o\o m\o sues msomezoo mmemoom + oma mnomazoo ««ama amaz= nmeazHoo<> «maqum zHamem N A A N N H N m.e.nn o.euon.wv omeazHoo<> N 4 N N mo memos an NON: 2H EH09; DMHm wm nmnmommdw ZOHHUmHomm II Q MAQH 40 TABLE 5 -- SUBCUTANEOUS VACCINATION AND CHALLENGE OF RABBITS CHALLENGE DOSE OF ACCUMULATED MORTALITY 7 DAYS AFTER P. MULTOCIDA R-473 CHALLENGE VACCINATED CONTROLS 500 LD50 0/8 4/4 1000 LD 0/8 4/4 50 DISCUSSION Live, streptomycin-dependent vaccines have been successfully dev- eloped and tested for enteric organisms. Several serotypes of entero- pathogenic Eacherichia coli have been made streptomycin-dependent and used to vaccinate young monkeys by the oral route. Protection against challenge with the virulent organism lasted one month (10). Human 12 adults were treated orally with 1 x 10 StrD E. coli without side ef- fects (14). Streptomycin-dependent Salmonella typhi was safely given to human volunteers in doses up to l x 1011 cells. Challenge with 1 x. 105 virulent bacteria showed significant protection (7). In 1963, live, StrD Shigella flexneri 2a was given to soldiers in bacillary dysentery endemic areas in five oral doses. Virulent Shigella challenge caused dysentery in 5.5 2 of unvaccinated controls and in none of the vac- cinated volunteers. There were no untoward reactions to the vaccine and the carrier rate remained unchanged. The StrD Shigella vaccine confer- red a significant degree of protection never before seen (17). Lyo- philized StrD Shigella vaccines were as effective as freshly prepared cultures and induced a lower incidence of post vaccinal reactions (18). AStrD Pasteurella multocida vaccine was developed and found to be protective for mice and rabbits against challenge from.high numbers of wild type virulent organisms. Protection of 4 logs was Observed after parenteral vaccination and challenge. Commercial P. multocida vaccines are considered adequately immunogenic when an efficiency of 2 logs pro- tection is attained. Administration of the StrD vaccine resulted in no detectable side effects or deaths in mice or rabbits. The StrD pheno- type remained stable from reversion to independence when large numbers 41 42 of cells were plated on non streptomycin media. Other workers have re- ported reversion rates from as many as 0.2 to 8 in 1 x 108 cells (24) to as low as 0.0 in 4 x 1012 (16). In an oral vaccine, even if a re- vertant appeared, it would not be able to compete with the large num- bers of normal bacteria in the alimentary tract (14,16,24). The most practical method of vaccine administration for the prevention of hemor- rhagic septicemia might be via the drinking water or by intranasal D aerosol. Oral or intranasal administration of a Str vaccine for this disease would take advantage of the probable natural route of infection, through the nasopharynx (2). High mortality in clinical cases of hemorrhagic septicemia creates severe economic losses especially in southeast Asia and Africa. A live preparation might prove more immunogenic than some of the killed vac- cines presently in use. Hemorrhagic septicemia is only one of a wide range of diseases caused by Pasteurella multocida. Streptomycin- dependent organisms may be of use as vaccines for other pasteurelloses such as rabbit "snuffles", fowl Cholera and pneumonic bovine pasteurel- losis. LITERATURE CITED 1. Adelberg, E. A., Mandel, M., and Chen, G. C. C.: Optimal Con- ditions for Mutagenesis by N-methyl-N'-nitro-N-nitrosoguanidine in Escherichia coli K12. Biochem Biophys Res Comm, 18, (1965): 788-795. 2. Bain, R. V. 8.: Hemorrhagic Septicemia. Food and Agriculture Organization of the United Nations, (1963). 3. Bierer, B. W., and Scott, W. F.: Comparison of Attenuated Live Pasteurella multocidb1Vaccine Given in the Drinking Water Every Two Weeks to an Injected Oil Base Bacterin Administered to Turkeys. Poul Sci, XLVIII, (1969): 520-523. 4. Carter, G. R.: Studies on Pasteurella multocidh 1. A Hemag- glutination Test for the Identification of Serological Types. Am'J Vet Res, 16, (1955): 481-484. 5. Derieux, W. T.: Immune Response of Medicated Turkeys Vac- cinated With Live Pasteurella multocidb. .mm J Vet Res, 38, (1977): 487-489. 6. Diena, B. B., Ryan, A., Wallace, R., Johnson, E. M., Baron, L. 8., and Ashton, F. E.: Effectiveness of Parenteral and Oral Typhoid Vaccination in Mice Challenged With a salmonella typhi - Salmonella typhimuriwnflybrid. Infect Immun, 15, (1977): 997-998. 7. Dupont, H, L., Hornick, R. B., Snyder, M. J., Libonati, J. P., and Woodward, T. E.: Immunity in Typhoid Fever: Evaluation of Live Streptomycin-Dependent Vaccine. Antimicrob Agents Chemother, (1971): 236-239. 43 44 8. Dupont, H. L., Hornick, R. B., Snyder, M. J., Libonati, J. P., (Formal, S. B., and Ganarosa, E. J.,: Immunity in Shigellosis. 1. Re- sponse of Man to Attenuated Strains of Shigella. J Infect Dis, 125, (1972): 5-11. 9. Felsenfeld, 0., Stegherr-Barrios, A., Aldova, E., Holmes, J., and Parrott, M. W.: In Vitro and In Vivo Studies of Streptomycin- Dependent Cholera Vibrios. Appl Microbiol, 19, (1970): 463-469. 1. 10. Felsenfeld, 0., Wolf, R. H., Greer, W. E., and Brannon, R. 8.: Oral Immunization With Polyvalent Streptomycin-Dependent Eboherichia coli rw' Vaccine. Appl Microbiol, 23, (1972): 444-448. 11. Heddleston, R. L., Gallagher, J. E., and Rebers, P. A.: Fowl Cholera: Immune Response in Turkeys. Avian Dis, 14, (1970): 626- 635. 12. Heddleston, K. L., and Rebers, P. A.: Fowl Cholera: Cross Immunity Induced in Turkeys With Formalin-Killed In-Vivo-PrOpagated Pasteurella multocida. Avian Dis, 16, (1972): 578-586. 13. Lederberg, J. and Lederberg, E. M.: Replica Plating and Indirect Selection of Bacterial Mutants. J Bacteriol, 63, (1952): 399- 406. 14. Linde, K., Koch, H., Koditz, H., Kittlick, M., and Stelzner, A.: Investigation of Oral Immunization With Streptomycin-Dependent Eacherichia coli. Folia Microbiol, 17, (1972): 153-156. 15. Maheswaran, S. K., McDowell. J.R., and Pomeroy, B. 8.: Studies on Pasteurella multocida. I. Efficacy of an Avirulent Mutant as a Live Vaccine in Turkeys. Avian Dis, 17, (1973): 396-405. 45 16. Mel, D.M., Papo, R. G., Terzin, A. L. and Vuksic, L.: Studies (on Vaccination Against Bacillary Dysentery 2. Safety Tests and Reacto- genicity Studies on a Live Dysentery Vaccine Intended fOr Use in Field Trials. Bull Wld Hlth Org, 32, (1965): 637-645. 17. Mel, D. M., Terzin, A. L., and Vuksic, L.: Studies on Vac- cination Against Bacillary Dysentery. 3. Effective Oral Immunization Against Shigella flexneri Za in a Field Trial. Bull Wld Hlth Org, 32, (1965): 647-655. 18. Mel, D. M., Arsic, B. L., Nikolic, B. D., and Radovanic, M. L.: Studies on Vaccination Against Bacillary Dysentery 4. Oral Immunization With Live Monotypic and Combined Vaccines. Bull Wld Hlth Org, 39, (1968): 375-380. 19. Oeschger, M. P. and Berlyn, M. K. B.: A Simple Procedure for Localized Mutagenesis Using Nitrosoguanidine. Molec Gen Genet, 134, (1974): 77-83. 20. Ose, E. E. and Muenster, O. H.: A Method for Evaluation of Vaccines Containing Pasteurella multocida. J Am Vet Med Assoc, 29, (1968): 1863-1866. 21. Reed, L. J. and Muench, H.: A Simple Method for Estimating Fifty Percent End Points. Amer J Hyg, 27, (1938): 493-497. 22. Reitman, M.: Infectivity and Antigenicity of Streptomycin- Dependent.Salmonella typhosa. J Infect Dis, 117, (1967): 101-107. 23. Vladoianu, I. R., Dubini, F., and Bolloli, A.: Contribution to the Study of Live Streptomycin-Dependent Salmonella Vaccines: the Problem of Reversion to a Virulent Form. J Hyg, Camb., 75, (1975): 203-214. 46 24. Vladoianu, I. R. and Dubini, F.: Experimental Model of Oral Antityphoid Vaccination With Live Streptomycin-Dependent salmonella typhimurium in C57BL/6 Mice. J Hyg, Camb., 75, (1975): 215-218. “7'1lllllllll'llflflllfll 1'“