Cultivation of Pasteurella haemolytica in a Casein

APPLIED MICROBIOLOGY, May, 1965 Copyright @ 1965 American Society for Microbiology Vol. 13, NO. 3 Printed in U.S.A. Cultivation of Pasteurella haemolytica in a Casein Hydrolysate Medium G. E. WESSMAN National Animal Disease Laboratory, Animal Disease and Parasite Research Division, U.S. Department of Agriculture, Ames, Iowa Received for publication 30 December 1964 ABSTRACT WESSMAN, G. E. (National Animal Disease Laboratory, Ames, Iowa). Cultivation of Pasteurella haemolytica in a casein hydrolysate medium. Appl. Microbiol. 13:426-431. 1965.-The growth of Pasteurella haemolytica strain H44L was studied under aerobic conditions in a medium of acid-hydrolyzed casein, supplementary cysteine, inorganic salts, vitamins, and a carbon source. The concentration of casein hydrolysate necessary for optimal growth was 1.5 or 2.0%, depending upon the carbon source employed. Essential vitamins were calcium pantothenate, nicotinamide, and thiamine. Concentrations as low as 0.01,ug/ml of thiamine monophosphate or thiamine pyrophosphate supported maximal growth, but thiamine hydrochloride or thiamine nitrate were active only at the unusually high levels of 10 to 20 jg/ml. The best carbon sources were D-galactose or sucrose. Maximal growth resulted from an inoculum containing fewer than 10 cells per milliliter of medium. Cellular yields averaged 6 X 109 to 7 X 109 cells per milliliter for the test organism and five other strains of P. haemolytica isolated from cases of bovine respiratory diseases. No studies dealing specifically with the nutritional requirements of Pasteurella haemolytica have been reported, although several studies have been done on the closely related species, P. multocida (Banerji and Mukherjee, 1953; Handa, 1958; Carter and Bain, 1960). Berkman (1942), studying the nutrition of various species of Pasteurella, included a single isolate of P. haemolytica; it grew in a medium of hydrolyzed gelatin, supplementary amino acids, glucose, and inorganic salts. The present report concerns the growth of P. haemolytica in a casein hydrolysate medium, and defines some optimal conditions for its culture. MATERIALS AND METHODS Smooth and nonsmooth variants of P. haemolytica H44L were employed in the major portion of the investigation. The smooth variant was isolated from the lung of an animal that died of bovine shipping fever; the nonsmooth variant appeared upon dissociation of the smooth form in static culture. All except the final experiment were conducted with both variants. Where significant differences in response resulted, data for both types are given; otherwise, only the data for smooth variants are reported. Other strains, isolated from cattle with respiratory disease, were tested to determine their ability to grow in the final medium, and are listed in Table 7. The maintenance of strains as stock cultures was described previously (Wessman, 1964). Growth experiments were conducted with 10- or 25-ml amounts of medium contained, respectively, in 50- or 250-ml Erlenmeyer flasks. Aeration was accomplished by agitation on a rotary shaker (156 rev/min). The cultures were incubated at 37 C. Numbers of viable cells were estimated by means of plate counts as described previously (Wessman, 1964). Turbidimetric measurements of growth were made by reading optical densities on a Bausch & Lomb Spectronic-20 colorimeter, with the use of light of 575-miA wavelength. Readings were made on cultures diluted 1:5 in distilled water with diluted medium as the blank. The medium was a modification of the deacidified casein partial hydrolysate (DCPH) developed by Higuchi and Carlin (1957) for the growth of P. pestis at 37 C. Magnesium, cysteine, vitamins, and carbon sources were prepared as separate filtersterilized solutions, and were added after autoclaving the rest of the medium. For inocula, cells were transferred from blood-agar plates to Brain Heart Infusion (Difco) broth, and grown for 10 hr, with aeration, at 37 C. The cells were separated from 10 ml of medium by centrifugation, washed, and suspended in 10 ml of 0.85% NaCl. This suspension was employed as the inoculum, 1% by volume being added to the medium. Dry-weight determinations were made after washing cells in distilled water and drying over- 426

VOL. 13, 1965 CULTIVATION OF PASTEURELLA HAEMOLYTICA 427 night at 110 C. The organism was grown in 25-ml amnounts of medium for these determinations. RESULTS Modifications in the medium of Higuchi and Carlin (1957) were made to compound the final medium for the growth of P. haemolytica. The two media, except for carbon source, are compared in Table 1. Changes in the original medium were: reduction in the concentration of casein hydrolysate, magnesium, and cysteine; elimination of gluconate, glycine, and biotin; replacement of thiamine hydrochloride with thiamine phosphate; and addition of nicotinamide. The concentration of casein hydrolysate needed for maximal growth of P. haemolytica depended partly on the carbohydrate added as carbon source. Casein hydrolysate at a concentration of 2% "solids" supported maximal growth when the medium contained D-glucose or sucrose (Table 2); this concentration inhibited growth when D- galactose was the carbohydrate present. The highest yield of cells with D-galactose present was obtained with 1.0 to 1.5% casein hydrolysate. Concentrations greater than 2.5% were markedly inhibitory under all conditions. The standard medium contained 2.0% casein hydrolysate, TABLE 1. Comparison of medium for Pasteurella pestis and medium for P. haemolytica Component Concn P. pestis P. haemolytica medium medium Casein hydrolysate. 2.25-2.5% 2.0%t K2HP04... 0.025 0.025 Citric acid.0.01 0.01 Sodium gluconate MgSO 4 7H20... 0.02 M 0.01 M FeSO4-7H 0 O.0001 0.0001 MnSO 4 -H20... O 00001 M 0. 00001 M Glycine... 0.027 L-Cysteine hydrochloride... 0.004 0.002 Calcium pantothenate... 1.0,ug/ml 1.0 jug/ml Thiamine hydrochloride log/ml 1.0 Thiamine monophosphate.0. 1 g/ml Biotin... Nicotinamide... 0.01 0.5 jug/ml 1.0 jug/ml Medium of Higuchi and Carlin (1957) for growth at 37 C. t Concentration was 1.5% when D-galactose was added as carbon source. Concentrations of casein hydrolysate are on the basis of "solids" content determined by drying samples at 100 C. unless the carbon source was D-galactose; then, 1.5% casein hydrolysate was used. An alkaline ph benefited initiation of growth of the organism. There was little difference in the rate of growth when the initial ph was 7.2, 7.5, or 7.8 (Table 3). However, a ph of 6.9 or less was inhibitory, and no visible growth occurred at ph 6.0 or 6.5. At initial ph values of 8.1 and 8.8, good growth was obtained, but precipitation of salts made these conditions undesirable; even TABLE 2. Effect of the relationship of casein hydrolysate concentration and kind of carbon source upon growth of the smooth variant of strain H44L Casein Optical density hydro- Carbon source lysate 6 hr 12 hr 24 hr 1.0 D-Glucose 0.22 0.34 0.37 D-Galactose 0.06 0.35 0.40 Sucrose 0.07 0.28 0.26 1.5 D-Glucose 0.24 0.37 0.41 D-Galactose 0.09 0.45 Sucrose 0.07 0.29 0.29 2.0 D-Glucose 0.27 D-Galactose 0.06 0.18 0.30 Sucrose 0.07 0.32 0.40 2.5 D-Glucose 0.28 0.35 0.28 D-Galactose 0.02 0.04 0.04 Sucrose 0.05 0.11 0.11 Concentrations of carbon sources were 0.5% D-glucose, 1.0% D-galactose, and 1.0% sucrose. TABLE 3. Carbon source Effect of initial ph upon growth of the smooth variant of strain H44L ph Optical density Initial Ter- 6 hr 12 hr 16 hr 24 hr D-Glucose, 6.9 6.0 0.18 0.11 0.10 0.5% 7.2 5.8 0.35 7.5 6.0 0.40 0.43 7.8 5.8 0.40 0.38 D-Galactose, 6.9 6.8 0.04 0.11 0.28 1.0% 7.2 6.9 0.07 7.5 6.9 0.09 0.47 7.8 7.0 0.07 0.40 Sucrose, 6.9 6.6 0.04 0.11 0.23 1.0% 7.2 6.6 0.06 0.33 0.47 7.5 6.5 0.08 0.35 0.47 7.8 6.6 0.06 0.32 0.40

428 WESSMAN APPL. MICROBIOL. w CL Wl Xi 7-0 LL 6- I 5-5 10 15 20 25 30 HOURS OF INCUBATION FIG. 1. Effect of the carbon source upon growth of the smooth variant of strain H44L. at ph 7.8, some precipitation occurred. A ph of 7.3 to 7.4 was selected as the most suitable, and the medium was routinely adjusted to this range. Experiments with various carbohydrates, alcohols, and salts of organic acids demonstrated that the best carbon sources were D-galactose or sucrose. Growth with sodium lactate was less, but the ph of the culture remained relatively constant. In Fig. 1, the growth responses of P. haemolytica with these compounds as carbon sources are compared with that resulting with D-glucose. As noted with many other aerobic microbial cultures, the organisms multiplied rapidly with glucose in the medium, but the rapid accumulation of acid resulted in early death of the culture. Growth was considerably slower with galactose, sucrose, or lactate, but the cells remained viable longer because the ph of the culture did not decline so rapidly. D-Xylose, D-mannitol, D-ribose, D-sorbitol, and fructose supported good growth, but with rapid development of limiting acidity. Sodium gluconate produced slow growth with occasional unexplained limited yields. Since the cells rapidly formed acids (particularly with glucose in the medium), which lowered the ph to inhibitory or lethal levels, attempts were made to control the ph by increasing the buffering capacity of the medium. However, phosphate, at concentrations high enough to 35 -J (-) -J LX LU 0.05M 0.025M 0.075M _OIM SMOOTH CELLS 0 5 10 15 20 25 0 5 10 15 20 25 HOURS OF INCUBATION FIG. 2. Effect of phosphate upon growth of smooth and nonsmooth variants of strain H44L. Concentrations are given for sodium phosphate buffer (ph 7.3). TABLE 4. Effect of vitamins upon growth of the smooth variant of strain H44L Vitamins Vitaminsadded added ~~~(16 hr) Optical density None... 0.02 Nicotinamide... 0.02 Calcium pantothenate... 0.01 Thiamine monophosphate... 0.02 Nicotinamide, calcium pantothenate. 0.06 Nicotinamide, thiamine monophosphate... 0.06 Calcium pantothenate, thiamine monophosphate... 0.04 Nicotinamide, calcium pantothenate, thiamine monophosphate... 0.49 Medium contained 1.0% sucrose. Concentrations of vitamins: nicotinamide, 1.0 ug/ml; calcium pantothenate, 1.0 Ag/ml; thiamine monophosphate, 0.1,g/ml. maintain a constant ph, inhibited the cultures. Concentrations of phosphate above 0.05 M inhibited growth of both smooth and nonsmooth variants, although the latter were more tolerant of phosphate than were the smooth cells (Fig. 2). Buffering with tris(hydroxymethyl)aminomethane (Tris) was not feasible because it was more toxic than phosphate.

VOL. 13, 1965 CULTIVATION OF PASTEURELLA HAEMOLYTICA Thiamine, pantothenic acid, and nicotinamide were required for maximal growth in the semidefined medium (Table 4). Single additions of the vitamins had no influence on growth; paired additions produced only slight stimulation of growth. Nicotinic acid was ineffective in replacing its amide. The following compounds did not affect growth of this organism: biotin, folic acid, p-aminobenzoic acid, pyridoxine, riboflavine, hemin, or oleic acid (as the free acid or the ester in Tween 80). TABLE 5. Effect of thiamine compounds upon growth of the smooth variant of strain H44L Compound added' Concn Optical density (16 hr) pg/mi None 0 0.02 Thiamine hydrochloride 1.0 0.05 5.0 0.10 10.0 0.41 20.0 0.51 Thiamine mononitrate 1.0 0.04 10.0 0.37 20.0 0.48 Thiamine monophosphate 0.0001 0.02 0.001 0.14 0.01 0.52 0.1 0.51 1.0 0.51 Thiamine pyrophosphate 0.0001 0.02 0.001 0.14 0.01 0.54 0.1 0.50 1.0 0.51 Medium contained 1.0% sucrose. Viable cells/ml (o hr) 2.97 X 108 2.97 X 107 2.97 X 106 2.97 X 105 2.97 X 104 2.97 X 10' 2.97 X 102 2.97 X 101 2.97 X 100 The most interesting response to a vitamin occurred with thiamine. Thiamine hydrochloride supported maximal growth only at concentrations of 10 to 20,ug/ml. This unusually high requirement could be replaced by either thiamine monophosphate or thiamine pyrophosphate at levels ordinarily used in nutritional studies (Table 5). A concentration of 0.01 to 0.1,g/ml of either derivative was optimal, and the two forms were equivalent in promoting growth. Thiamine nitrate produced responses similar to thiamine hydrochloride. In this medium, the need for a phosphorylated form of thiamine was clearly indicated. Mge+ stimulated the growth of P. haemolytica, affecting both the rate and total amount of growth. A 0.01 M concentration was optimal; the medium was not turbid as was that resulting when the higher concentration used by Higuchi and Carlin (1957) was employed. Either 0.002 M cysteine hydrochloride or 0.005 M sodium thiosulfate served as a supplementary source of sulfur. In the latter case, growth was slightly delayed. These compounds could not be replaced by DL-methionine. The medium supported growth from an initial TABLE 7. Viable-cell yields from the smooth variants of several strains of Pasteurella haemolytica Strain Viable cells/ml X 109 H23N... 6.81 H44L... 6.95 H57T... 6.03 H66T... 7.10 P-1148... 5.98 801... 6.72 Medium contained 1.0% sucrose; counts were made at 16 hr. TABLE 6. Effect of size of inoculum upon growth of the smooth variant of strain H44L 3 hr 0.22 6 hr 0.32 0.06 Optical de.-sity D-Glucose, 0.5% D-Galactose, 1.0% Sucrose, 1.0% 12 hr 0.52 0.48 0.34 0.25 24 hr 0.48 0.41 0.45 0.45 0.43 0.03 0.03 48 hr 3 hr 0.13 6 hr 0.07 0.02 12 hr 0.54 0.22 0.03 24 hr 48 hr 0.47 0.49 0.23 0.09 Readings were made only after visible turbidity appeared in culture. 3 hr 0.10 6 hr 0.22 0.07 12 hr 24 hr 0.47 0.27 0.41 0.01 0.41 0.40 0.02 429 48 hr 0.43 0.41 0.41 0.40

430 WESSMAN APPL. MICROBIOL. concentration of fewer than 10 cells per milliliter. Table 6 shows the growth response of smooth variants when inoculated at various levels up to more than 108 cells per milliliter. Even at a level theoretically lower than three cells per milliliter, good growth resulted in 48 hr. Dry-weight determinations were made to measure total cell yield in the standard medium, employing sucrose as carbon source. The results on cultures incubated for 24 hr showed average yields of 0.70 mg/ml for smooth cells; comparable values for nonsmooth cells were 0.725 mg/ml. Five smooth strains of P. haemolytica, in addition to strain H44L, were tested for growth in the complete medium (Table 7). There were definite variations in the rates of growth, but total cellular yield was essentially equivalent for all strains tested. DISCUSSION The principal alterations in the medium of Higuchi and Carlin (1957), effecting optimal growth of P. haemolytica, were the addition of nicotinamide, provision of thiamine in phosphorylated form, and reduction of the concentration of casein hydrolysate. Few viable cells were produced in the absence of nicotinamide, or in the presence of small amounts of unphosphorylated thiamine. Berkman (1942) showed that nicotinamide was essential for the propagation of P. multocida in a hydrolyzed gelatin medium, but nicotinic acid could not substitute for the amide. However, the single strain of P. haemolytica tested grew through repeated transfers in the medium without nicotinamide. The strains of P. haemolytica investigated in the present experiments definitely require nicotinamide, and, in their inability to utilize the acid, they behave like the hemorrhagic septicemia Pasteurella species studied by Berkman. Few microorganisms preferentially utilize thiamine pyrophosphate instead of thiamine, although some organisms can use either compound equally well. Many fail to metabolize the phosphorylated form because they cannot absorb it. There are rare reports of organisms having an absolute requirement for thiamine pyrophosphate; some gonococci need this form and cannot use thiamine (Lankford and Skaggs, 1946). The responses of P. haemolytica also are unusual, as the organisms efficiently utilize phosphorylated thiamine, but free thiamine has limited growthpromoting activity. The data are not adequately explained by simply suggesting that the cells cannot phosphorylate the vitamin efficiently. Thiamine monophosphate is not an intermediate in the conversion of thiamine to thiamine pyrophosphate in other systems. In baker's yeast (Camiener and Brown, 1960) and in rat liver (Mano, 1960), the monophosphate is first dephosphorylated to thiamine, and thiamine pyrophosphate is formed by direct pyrophosphorylation of thiamine with the aid of adenosine triphosphate. These reactions have been suggested, although not fully demonstrated, for bacteria. If they occur under our experimental conditions, free thiamine formed should limit growth when thiamine monophosphate is provided as the source of the vitamin. The utilization of thiamine is markedly affected by the cultural environment. MaciasR (1958) reported that oleic acid and Tween 80 enhance and sodium acetate depresses the utilization of thiamine by Lactobacillus fermenti. Studying the nutrition of this organism, BAnhidi (1960) examined the effects of reducing compounds in the medium, and showed that the growth-promoting activities of thiamine derivatives may depend upon the presence of various amounts of the disulfide forms of these compounds in different preparations. Thus, interactions between different nutrients often mask the true effect of thiamine and its derivatives. Since we employed a semidefined medium, undoubtedly containing some dispensable nutrients, the actual mechanism of thiamine utilization can be only cautiously suggested. The cellular yields are somewhat less than those of many other gram-negative bacteria in casein hydrolysate media. Maximal growth obtained with P. haemolytica reached 1010 cells per milliliter; yields of P. pestis were 2 X 1010 to 3 X 1010 cells per milliliter (Higuchi and Carlin, 1957). Growth was slightly greater, however, than that of P. multocida with similar media and cultural conditions. Bain (1963) reported dry-weight yields of 0.5 mg/ml; the average yield was approximately 0.7 mg/ml for P. haemolytica in the present work. When Bain added an autodigest of pancreas to his medium, however, the yields of P. multocida were considerably increased; probably a similar stimulation would occur with P. haemolytica. The definition of unknown factors in complex materials, such as pancreatic digest, is needed to enhance yields of these organisms. Supplying the carbon source in several small increments, as Higuchi and Carlin (1957) did, was not investigated in the present work, since the addition of a single quantity of carbohydrate was much more convenient for culturing the rapidly proliferating P. haemolytica. However, D-xylose, the best carbon source for P. pestis, would not be suitable for P. haemolytica except when added in extremely small increments, since it is very rapidly metabolized, producing large amounts of acid. Mannitol, ribose, sorbitol, or fructose would produce similarly unsatisfactory

VOL. 13, 1965 CULTIVATION OF PASTEURELLA HAEMOLYTICA 431 conditions. The use of 0.5 or 1.0% D-galactose or sucrose, with harvest of cells at 12 to 15 hr, before conditions became extremely acid, is a practical solution if the casein hydrolysate medium is employed. LITERATURE CITED BAIN, R. V. S. 1963. Haemorrhagic septicemia, p. 21-22. FAO Agricultural Study, Food and Agricultural Organization of the United Nations. BANERJI, T. P., AND R. MUKHERJEE. 1953. Nutritional requirements of Pasteurella septica. Current Sci. 22:177-178. BXNHIDI, Z. G. 1960. Activation of the disulfide forms of thiamine and its phosphates as growth factors for Lactobacillus fermenti. J. Bacteriol. 79:181-190. BERKMAN, S. 1942. Accessory growth factor requirements of the members of the genus Pasteurella. J. Infect. Diseases 71:201-211. CAMIENER, G. W., AND G. M. BROWN. 1960. The biosynthesis of thiamine. II. Fractionation of the enzyme system and identification of thiazole monophosphate and thiamine monophosphate as intermediates. J. Biol. Chem. 235:2411-2417. CARTER, G. R., AND R. V. S. BAIN. 1960. Pasteurellosis (Pasteurella multocida). A review stressing recent developments. Vet. Rev. Annotations 6:105-128. HANDA, R. K. 1958. Studies on Pasteurella septica (P. 52) (1) optimum requirements. J. Vet. Animal Husbandry Res. 3:40-53. HIGUCHI, K., AND C. E. CARLIN. 1957. Studies on the nutrition and physiology of Pasteurella pestis. I. A casein hydrolyzate medium for the growth of Pasteurella pestis. J. Bacteriol. 73: 122-129. LANKFORD, C. E., AND P. K. SKAGGS. 1946. Cocarboxylase as a growth factor for certain strains of Neisseria gonorrhoeae. Arch. Biochem. 9: 265-283. MACIASR, F. M. 1958. Effect of oleic acid on the response of Lactobacillusfermenti to thiamin and its moieties. J. Bacteriol. 75:561-566. MANO, Y. 1960. Studies on enzymatic synthesis of cocarboxylase in animal tissues. I. Fundamental properties of the reaction. J. Biochem. (Tokyo) 47:24-36. WESSMAN, G. E. 1964. Interrelationships of smooth and nonsmooth variants in the dissociation of Pasteurella haemolytica. J. Bacteriol. 87:356-360. Downloaded from http://aem.asm.org/ on September 19, 2018 by guest