Vol. 100, No. 3 Printed in U.S.A.
JOURNAL OF BACI8R10LOGY, Dec. 1969, p. 1237-1241 Copyright 0 1969 American Society for Microbiology
Yersinia philomiragia sp. n., a New Member of the Pasteurella Group of Bacteria, Naturally Pathogenic for the Muskrat (Ondatra zibethica) WAYNE I. JENSEN, CORA R. OWEN, AND WILLIAM L. JELLISON Bureau ofSport Fisheries and Wildlife, Bear River Research Station, Brigham City, Utah 84302, and Rocky Mouniain Laboratory, National Inistitute of Allergy and Infectious Diseases, Hamilton, Montana 59840
Received for publication 19 September 1969
A bacterium experimentally pathogenic for muskrats (Ondatra zibethica), white mice, mountain voles (Microtus montanus), and deer mice (Peromyscus maniculatus) was isolated from the tissues of a sick muskrat captured on the Bear River Migratory Bird Refuge (Brigham City, Utah) and from four surface water samples collected within 15 miles of that point. In culture, the cells are chiefly coccoid, but in the tissues of muskrats and voles they resemble the bizarre forms of Yersinia pestis, except for their smaller size. The characteristics of the organism are described and the name Yersinia philomiragia sp. n. is proposed. MATERIALS AND ME7ODS Study areas. Figure 1 is a map of the refuge and the surrounding territory in which isolates were found. These marshes, ranging from brackish to highly saline, are divided into shallow waterfowl management units by dikes on the Ogden Bay and Bear River Refuges. Aquatic and terrestrial invertebrates, fish, aquatic birds, muskrats, and other small mammals abound. Collection of water samples. Sterile 4-oz (118-ml) screw-cap bottles were filled with water by immersion at collection points and were held in ice water until used. Isolation, media, and growth conditions. The first isolation of the organism was made by streaking tissues from the muskrat on plates of Brain Heart Infusion Agar (BHIA) and Cystine Heart Agar with 1% Hemoglobin (CHHA), all Difco products. Stock cultures were maintained on BHIA slants. Isolations from marsh water were accomplished by animal inoculation and subsequent culturing of infected tissues. A 2-ml volume of water was injected intraperitoneally into each of two white mice. Upon death of a mouse, its spleen was cultured on cystineglucose-blood-agar (CGBA) plates, and a triturate of the tissue was injected into two more mice. Isolated colonies on plates were transferred to CGBA slants. CGBA was prepared from Cystine Heart Agar (Difco). After the medium was autoclaved and cooled to 60 C, 8% fresh defibrinated rabbit blood was added, and the mixture was held at 60 C for 90 min with occasional shaking before being dispensed. Blood agar (BA) was prepared by adding 5% horse blood to Tryptose Blood Agar Base (Difco). 1237
A moribund muskrat (Ondatra zibethica), found on the shoreline of a marsh on the Bear River Migratory Bird Refuge in northern Utah, was submitted to the Bear River Research Station for examination in July 1959. The weak and dyspneic young animal died within a few hours. In the course of routine diagnostic procedures, a small gram-negative bacillus was isolated from the lungs and liver, the only tissues examined bacteriologically. Although it resembled members of the genus Pasteurella in many respects, it was not identical with any previously described species. No other naturally infected animals were found, but interest in the organism was revived in 1960 as a result of an attempt by Rocky Mountain Laboratory personnel to demonstrate the presence and distribution of Francisella novicida (Pasteurella novicida), which was first isolated from a water sample taken on the Ogden Bay Refuge in 1950 (1). Although the primary goal of their study was not attained, additional isolations of the muskrat pathogen were made from four water samples collected in marshy areas within a few miles of the Bear River Research Station. This report describes morphological, biological, and biochemical characteristics of the new organism, Yersinia philomiragia sp. n. (ATCC 25015), and, briefly, the gross pathological changes that it induced in the one naturally infected muskrat.
1238
JENSEN, OWEN, AND JELLISON
Five liquid media were used in studies of growth characteristics (Table 1). Ground meat medium was prepared at the Rocky Mountain Laboratory according to the following procedure: 500 g of fresh ground beef heart was infused in 1,000 ml of distilled water for 18 to 24 hr at 4 C, cooked in flowing steam for 75 min, autoclaved at 15 lb of pressure for 30 min, and then filtered through gauze; the volume was restored to 1,000 ml, 10 g of peptone and 5 g of NaCI were added, and the pH was adjusted to 7.2 with 1 M NaOH before sterilization. LYA broth was composed of 3% Lactalysate (BBL), 2% Yeast Autolysate (Albimi Laboratories, Inc., Flushing, N.Y.), and 0.5% glucose; the pH was 7.3 to 7.4 before sterilization. Phytone-Ground Meat Medium was obtained from BBL, and Brain Heart Infusion, from Difco. Carbohydrate broth base was made from 1% Tryptose (Difco), 0.5% NaCl, 0.1% Na2HPO4, 0.05% agar, and phenol red or bromcresol purple indicator (pH 6.8). For biochemical studies, filter-sterilized carbohydrates were added to the base to a final concentration of 0.5 or 1.0%. Lead Acetate Agar (Difco) was used to test for H2S production. Nitrate broth was prepared from 1% tryptone (Difco), 0.5% NaCl, and 0.02% NaNO3 (pH 7.4). An incubation temperature of 37 C was used routinely. The growth temperature range was determined by incubating inoculated BHIA and CGBA plates for 7 days within a range of 15 to 42 C. BBL sensitivity discs were used to test the organism's susceptibility to antibiotics. Heavy suspensions of 24-hr growth from BHIA slants were streaked evenly on BHIA plates, and two discs were placed on the surface of each. Plates were observed daily during a 96-hr incubation period at 37 C. Except where deviations are noted, classification tests were done according to standard procedures (5). Experimental animals. White mice of the Rocky Mountain Laboratory stock were used at both laboratories. Guinea pigs were obtained from local breeders (Hamilton, Mont.). Muskrats, voles (Microtus montanus), and deer mice (Peromyscus maniculatus) were live-trapped near the two laboratories. Mallard ducks (Anas platyrhynchos) were obtained from a commercial breeder in Illinois. Pathogenicity. Susceptibility of muskrats, white mice, guinea pigs, voles, and deer mice to infection with the organism was usually determined by intraperitoneal injection of suitable dilutions of 24-hr BHIA or CGBA slant cultures in physiological saline solution. Ten muskrats were inoculated by the intranasal route: heavy suspensions of the organism were dripped into the nares of lightly anesthetized animals with a hypodermic syringe and needle. Ducks were inoculated intravenously (brachial vein).
RESULTS Nineteen of 192 water samples were lethal for mice. The 19 samples yielded 28 isolates, four of which were the organism described here. Figure 1 shows the points of origin of these four and the area in which the sick muskrat was found. Al-
J. BACTERIOL.
though water samples were collected within a radius of 200 miles of the Research Station, all isolates of the new species were found within 15 miles of that point. CeU morphology. When first isolated, individual cells could not be clearly distinguished in Gram stains of culture or tissue smears because of a mucoid material that enveloped them. Wright, Giemsa, carbol fuchsin, methylene blue, and safranin stains were equally unsatisfactory, but the bacterium could occasionally be stained successfully with malachite green. However, when growth from solid media was washed in 0.85% NaCl solution, stained smears revealed small organisms, mainly coccoid, but with a considerable number of bizarre forms (balloons, safetypins, spindles, clubs) and a few frankly bacillary cells. Recently, perhaps from long maintenance on artificial media, the bacteria have become more easily visible with Gram, Giemsa, and Wright stains, but their indistinct outlines suggest that a thin layer of mucoid material still adheres to the cell walls (Fig. 2). In smears from cultures, the cells are now mostly coccoid and, although their small and variable size makes precise measurement difficult, they appear to range from 0.2 to 1.0 Mgm in diameter. In tissues of infected mice or guinea pigs, the organism was not demonstrated with Gram, Giemsa, or malachite green stains. In tissues of voles and muskrats, however, it was easily made visible by Giemsa or Wright stains. Even in animals inoculated with cultures largely made up of the small coccoid form, the bacterial cells in tissue smears were much larger, with the bizarre and bipolar-staining bacillary forms predominating (Fig. 3). The bacterium was gram-negative and was not acid-fast. Capsules were not demonstrated by Hiss's technique. It was nonmotile in liquid media and semisolid agar incubated at both 37 C and room temperature. Lack of motility was also indicated by the sharply demarcated areas of liquefaction in gelatin even at 37 C, if mixing was avoided when tubes were moved to cooler temperatures to test for the presence of hydrolysate. Colonial morphology at 37 C. On BHIA plates at 1 day colonies were punctiform, translucent, and confluent in the area of heaviest inoculum; at 2 days, they were translucent, cream-colored, circular, entire, convex, mucoid, and glistening, with a size of 1.5 to 3.0 mm; at 6 days, wellisolated colonies as large as 8 mm were visible. On BHIA slants, thin to moderate, confluent, translucent, glistening growth was present at 1 day, and became more abundant with age. On BA plates, punctiform, round, nearly
VOL. 100, 1969
YERSINIA PHILOMIRAGIA, SP. N.
1239
eS STREAMS a CANALS ROADS ... a--RAILROAD
EJ
ISOLATE NO.
0O671
FiG. 1. Outline map of the Bear River Migratory Bird Refuge showing the points of origin of the five isolates of Y. philomiragia. No. 319-L is the isolate from a muskrat.
F1G. 2. Gram stain of Y. philomiragia from 48-hr culture on cystine-glucose-blood-agar. X 3,000.
colorless, transparent colonies were visible at 1 This medium was found at the Rocky Mountain day. At 6 days, colony size was 1.0 to 1.5 mm; Laboratory to be superior to other blood-agars there was no hemolysis, but blood turned brown. with respect to rate and abundance of growth. Colonies on BA slants appeared as a thin, On CGBA slants at 1 day, colonies were shiny, confluent, colorless film at 1 day; at 6 days, abundant, confluent, grayish, and shiny; at 6 they were thicker and slightly raised on edges. days, they were more abundant and heaped On CGBA plates, growth was more rapid slightly in the center. than on BHIA and BA. At 1 day, colonies were On Nutrient Agar (Difco) at 1 day, colonies gray, opaque, entire, shiny, and mucoid (aspect were punctiform and translucent, growing only reminiscent of Francisella tularensis), with a where inoculum was heavy; at 6 days, their size size of 1 mm; at 6 days, their size was 3 to 5 mm. was 0.25 to 1.0 mm.
1240
JENSEN, OWEN, AND JELLISON
J. BAc-rERioL.
cycline, chloramphenicol, erythromycin, and streptomycin. Unlike P. multocida (3), but like F. tularensis, it was resistant to penicillin. Pathogenicity. Y. philomiragia was pathogenic for muskrats, mountain voles, deer mice, and white mice by the intraperitoneal route. The susceptibility of muskrats to intranasal instillation of suspensions of the organism was inconsistent and unrelated to the dose. Of 10 animals so inoculated, 3 died, but the other 7 showed no signs of infection. Six adult mallards appeared FIG. 3. Wright stain of a blood smear from Microtus to be unaffected by intravenous injection of montanus experimentally infected with Y. philomiragia. doses ranging from 105 to 4.0 X 105 bacteria. Pathology. The lungs of the naturally infected x 1,500. muskrat were almost entirely hepatized. Small Growth of the bacterium in several liquid media yellowish foci were distributed throughout the liver. Other organs and tissues appeared to be is compared in Table 1. Biochemical reactions. Stratiform liquefaction normal. Histopathological examinations of this occurred in gelatin stabs and, when the organism animal were not made. Experimental infections and pathological was first isolated, appeared to be incomplete. Continued incubation at 37 C increased the depth, changes in laboratory animals will be discussed but not the degree, of liquefaction. This property in greater detail in a forthcoming paper. has changed, however, and gelatin is now apparently completely liquefied. Litmus milk was TABLE 1. Type and abundance of growth of Y. philomiragia in liquid media slightly reduced in the bottom of the tube and became alkaline. H2S was not produced; nitrates Growth were not reduced to nitrites. Indole production Medium was usually not demonstrated by the Kovacs 1 day 6 days method, but occasionally a trace could be detected. No growth occurred in Clark-Lubs me- Ground meat Faint, even tur- Same dium, and, when it did occur in urea medium, medium bidity the specific substrate was not utilized. The Lactalysate- No visible Heavy to very bacterium was catalase-positive. Dextrose, malYeast autol- growth to heavy turbidtose, and sucrose were utilized with the producfaint turbidity ity throughysate tion of acid but no gas; galactose and levulose at top only out; heavy ring pellicle were occasionally attacked; lactose, sorbitol, Growth doubt- Very slight to salicin, glycerol, dextrin, mannitol, rhamnose, PhytoneGround Meat ful moderate turraffinose, dulcitol, arabinose, inulin, inositol, Medium bidity with mannose, xylose, and starch were not utilized, (BBL) ring pellicle and the medium usually became alkaline. Heart In- Slight to moder- Moderate turRelationship to free oxygen. The bacterium was Brain ate turbidity bidity fusion aerobic and facultatively anaerobic. Table 2 (Difco) illustrates the effect of atmospheric composition Carbohydrate Moderate tur- Moderate turon growth on BA and CGBA plates. Broth Base bidity in upper bidity 1.9 cm throughout The thermal death point of all five strains was 56 C for 5 min. Growth temperature range. Growth at various temperatures was as follows: 15 C, no visible TABLE 2. Size of colonies of Y. philomiragia on two media in three atmospheres after 4 to 6 days growth within 1 week; 20 C, growth slower than at 37 C, but almost as heavy after 5 days; 40 C, Colony size (mm) moderate growth on BHIA, slight growth on Atmosphere CGBA; 41.5 C, no growth on BHIA, slight growth on CGBA; 43 C, no growth on either medium. Antibiotic sensitivity. When tested against Aerobic............. sensitivity discs, the bacterium was susceptible to Anaerobic........... 10% CO2............ tetracycline, dimethylchlortetracycline, oxytetra-
VOL. 100, 1969
YERSINIA PHILOMIRAGIA, SP. N.
DISCUSSION The importance of this bacterium as a wildlife pathogen cannot now be assessed. Only one infected muskrat was found, and a serious attempt to find others was not made. Large numbers of sick and dead animals along the shorelines or in the water could hardly have escaped notice, but many could have died in their lodges without being observed. The presence of the organism in 4 of 192 water samples may be evidence of an unobserved epizootic. Moreover, the white mouse (as compared with the muskrat or mountain vole) is relatively resistant to the experimental disease, and it is possible that more isolations would have been made if a more susceptible species had been used. The new bacterium has no known characteristics that would exclude it from the genus Pasteurella, as now defined in Bergey's Manual. Morphologically, cells in stained smears of cultures are indistinguishable from those species (tularensis and novicida) that are placed in the new genus Francisella (2). In smears of cultures shortly after isolation from the muskrat, as well as in tissue smears from infected voles and muskrats, the organism resembles Y. pestis (Pasteurella pestis), except for its smaller size. Although the general description of the species of Pasteurella in Bergey's Manual states that they do not liquefy gelatin, two species, septicaemiae and anatipestifer, that do so have been placed in the genus. The new species' lack of conformity with the majority in this respect, therefore, is not a deviation from accepted classification criteria.
1241
Other biochemical properties of the organism are quite unremarkable for members of the group. Deoxyribonucleic acid hybridization studies by Ritter and Gerloff (4) indicate that this bacterium, referred to as Ui, is more closely related to the two species of Yersinia (pestis and pseudotuberculosis) than to the species of Francisella and Pasteurella tested, although the evidence is not conclusive. Because of this finding and the organism's morphological resemblance to Y. pestis in some animal tissues, we have placed it in the genus Yersinia. The specific epithet, derived from the Greek combining form philo (loving) and the plural of the Latinized English word mirage, was chosen because of the spectacular mirages that are seen in the area where the isolations were made. ACKNOWLEDGMENT We are grateful for the assistance of Emery Buker in conducting both the field and the laboratory investigations. LITERATURE CITED 1. Larson, C. L., W. Wicht, and W. L. Jellison. 1955. A new organism resembling P. tularensis isolated from water. Public Health Rep. 70:253-258. 2. Philip, C. B., and C. R. Owen. 1961. Comments on the nomenclature of the causative agent of tularemia. Int. Bull. Bacteriol. Nomencl. Taxon. 11:67-72. 3. Queen, F. G., and E. R. Quortrup. 1946. Treatment of Pasteurella multocida (fowl cholera) infection in wild ducks with autogenous bacterin and penicillin. J. Amer. Vet. Med. Ass. 108:101-103. 4. Ritter, D. B., and R. K. Gerloff. 1966. Deoxyribonucleic acid hybridization among some species of the genus Pasteurella. J. Bacteriol. 92:1838-1839. 5. Society of American Bacteriologists. 1957. Manual of microbiological methods. McGraw-Hill Book Co., Inc., New
York.
