- Email: [email protected]
Subacute inhalation toxicity study of an ice-nucleation-active Pseudomonas syringae administered as a respirable aerosol to rats
Toxicology Letters, 54 (1990) 1577167
157
Elsevier TOXLET 02459
Subacute inhalation toxicity study of an ice-nucleation-active P~~~d~~~~~s syringne administered as a respirable aerosol to rats
R.A. Goodnowl,
G. Katz2, D.C. Haines3 and J.B. TerriIP
‘Genencor International, Inc., Rochester. NY, ‘Eastman Kodak Company, Roehester, NY and ‘Ha&ton Laboratories America, Inc., RockviNe. MD (U.S.A.)
(Received 6 March 1990) (Revision received 7 May 1990) (Accepted 25 July 1990) Key words: Pseudomonas syringae: Ice-nucleation active; Respirable aerosol; Inhalation; Snow making; Biotechnology
SUMMARY The inhalation toxicity of a commercial sample of an ice-nucleation-active Pse~domonus syringae (strain 3 1a) was evaluated by repetitively exposing rats to about 700 mg/m’ of an aerosol consisting of a suspension of 0.0008, 0.4 or 0.8 g/l of bacteria in water for 2 h per day, 5 days per week for 13-14 exposures. No mortality, moribundity or biologically significant differences in clinical signs, body weight, food consumption or clinical pathology were observed. Animals tested at 500 times (0.4 g/l) and 1000 times (0.8 g/l) the recommended ice-nucleation concentration (0.0008 g/l) exhibited concentration-dependent increased lung weights. Several animals exhibited enlarged tracheobronchial lymph nodes. The pulmonary responses observed are considered compatible with a mild irritant reaction. There was no evidence of bacterial infection. Animals tested at a concentration typical for the discharge mouth of a snow gun (0.0008 g/l) demonstrated no significant biological effect.
INTRODUCTION
Specific isolates of the gram-negative fluorescing bacteria, Pseudomonas syringae, due to their ubiquity and high ice-nucleation activity, are considered one of the more important sources of natural ice nuclei [I]_ The recent commercial use of ice-nucleation-active P. syringae in water as an ice nucleator in the manufactu~ng of man-made Address.for correspondence:
Robert A. Goodnow, Genencor International, Inc., 1700 Lexington Avenue, Rochester, NY 14650-3605, U.S.A.
158
snow allows freezing to begin in -3 to - 5°C water compared to -8 to - 10°C in water without the presence of P. syringae. Although snow makers and skiers may be exposed to plumes emitted from snow guns, snow guns are designed to emit water droplets averaging 300 pm in diameter, a size that is generally considered non-respirable [2, 31. Moreover, the supercooled droplets quickly form ice or snow crystals that average greater than 3000 pm in diameter [4]. However, Eisel et al. [5] concluded that about 6% of the water may be lost from man-made snow through evaporation and sublimation, thereby reducing droplet diameter and exposing the snow maker and skier to respirable-size bacterial cells and cellular components. Public health officials have expressed concern that snow makers and skiers may exhibit adverse health effects if exposed to snow plumes containing gram-negative bacterial ice nucleators. This concern arises from reports of occupational lung disease in textile workers that was thought to be related to exposure to endotoxin from gram-negative bacteria 1691. The acute toxicity of an ice-nucleation-active P. syringae (strain 31a) has been studied in several animal species [unpublished data]. The oral LD50 value of the commercial material (strain 31a) for rats was greater than 5.0 g/kg. The inhalation L&o value for commercial material was greater than 5.5 mg/l for 2 h (aerodynamic mass median diameter, 10 pm; geometric standard deviation, 5.0) and greater than 5.1 mg/l for 4 h (aerodynamic mass median diameter, 3.2 pm; geometric standard deviation, 3.0) of exposure to a dry dust atmosphere, and greater than 5.3 mg/m’ of a 0.8 g/l aqueous suspension of strain 31a for 4 h of exposure (aerodynamic mass median diameter not reported). The acute dermal LDSs of strain 31a in rabbits was greater than 2.0 g/l, and the strain was considered non-irritating to the skin of rabbits. Strain 3 la did not cause dermal contact sensitization in guinea-pigs. It produced minor and transient changes of the conjunctiva in only 1 of 6 rabbits tested for eye irritancy. The objective of the following study was to determine the toxicity of strain 31a following repeated daily inhalation exposure corresponding to a 5-day work-week for people who make snow. In addition, the exposure concentrations and duration were selected to exceed those anticipated for skiers exposed to snow plumes emitted from snow guns. MATERIAL
AND METHODS
Preparation of bacterial suspensions P. syringae (strain 31a) was propagated
under fermentation in a defined sucrosebased mineral salts medium. The culture was dewatered by ultrafiltration to approximately a 20% solids cell slurry, rapid-frozen in liquid nitrogen, and desiccated. This preparation was commercially available for use as an ice-nucleator in the manufacturing of man-made snow and was supplied by Bio-Products Division of Eastman Kodak Company, Rochester, NY 14650. Prior to each animal exposure, desiccated bacterial culture containing 1.G 2.0 x 10” organisms (approximately lo4 viable and lo* non-viable cells of strain 31a)
159
per gram was prepared concentrations
(8 g/l) and then diluted
of 0.0008,0.4
with sterile spring water to provide
and 0.8 g/l. These concentrations,
1, 500 and 1000 times
the concentrations discharged from a snow gun, were selected to provide exposure levels equal to anticipated exposures, plus exposure levels that would provide data on safe levels of exposure greater than that which was anticipated. Standard plate count techniques of stock bacterial
were employed to determine suspension [lo].
the colony-forming
units (cfu) per ml
Lipopolysaccharide (LPS) (endotoxin) assay Eight separate dried culture preparations of P. syringae strain 31a were analyzed in duplicate to determine the average LPS content per gram of dry culture using the Limulus Amebocyte Lysate (LAL) assay [ 111. Test animals Male and female Sprague-Dawley rats (Crl:CD@BR) were obtained from Charles River Laboratories, Inc., Raleigh, NC. Animals were approximately 7 weeks of age at the start of the study and were randomly assigned to exposure groups composed of 10 males and 10 females. Purina Certified Rodent Laboratory Chow #5002 and tap water were available ad libitum throughout the 3-week acclimation and non-exposure periods. Neither food nor water was available during each 2 h exposure period. Aerosol generation and analysis Four lOOO-liter stainless steel and glass inhalation exposure chambers were operated with total airflow rates through each chamber of 216, 221, 214 and 218222 liters per minute (lpm) for the 0.0, 0.0008, 0.4 and 0.8 g/l bacterial suspension groups, respectively. For aerosol generation, each bacterial suspension was drawn from a reservoir to a Spraying System atomizer (Wheaton, Illinois) by a fluid metering pump. The suspension was mixed with compressed air in the atomizer to yield the bacterial aerosol, which was further diluted with air, as necessary, to provide the target concentration. The control group animals were exposed to an aerosol of spring water. All groups were exposed for 2 h daily, 5 days per week for 13-14 exposures. Exposure concentrations were measured by collecting samples of each chamber atmosphere and then assaying the samples for total protein concentration and viable fluorescing pseudomonads (colony-forming units). Protein concentration was analytically determined using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Richmond, CA) and pseudomonad colony-forming units were enumerated using a standard plate count technique. Particle size distribution was determined twice daily using an aerodynamic particle sizer (TSI Aerodynamic Particle Sizer, Model #3301/3302). Nominal aerosol concentrations (calculated from the ratio of test material used and total airflow) were determined daily. Homogeneity of aerosol distribution in each chamber was determined twice weekly.
160
Clinical observations All rats were observed twice daily for mortality and moribundity. Animals were observed once daily (immediately after exposure) for clinical manifestations of adverse health effects resulting from exposure to P. syringae. Animal observations included, but were not limited to, examination for clinical abnormalities associated with respiratory, circulatory and central nervous systems, hair, skin, eyes, feces and urine. Body weight measurements were taken prior to randomization, at test initiation {prior to and after the first exposure), and twice weekly thereafter. Feed consumption measurements were recorded weekly. Water consumption measurements were recorded daily. Clinical pathology
Animals were fasted overnight and anesthetized with ketamine, and blood samples were collected from the orbital sinus. Animals were sacrificed by exsanguination following intraperitonal injection of sodium pentobarbital (4 mg/lOO g body wt.). Red and white blood cell counts, hemoglobin concentration, hematocrit, red cell indices, and platelet count were performed using a Coulter Counter (Model S+IV System, Coulter Electronics, Inc., Hialeah, FL). Prothrombin, activated partial thromboplastin time, and clotting time were performed by Coag-A-MateTM x 2 (General Diagnostics, Morris Plains, NJ). Differential white blood cell counts were performed by microscopy. Clinical chemistry dete~inations (sodium, potassium, chloride, total protein, albumin, calcium, phosphorus, lactic dehydrogenase, total bilirubin, urea nitrogen, cholesterol, glucose, asparate aminotransferase, alanine aminotransferase, and alkaline phosphatase) were conducted using a BMD/Nitachi 737 Chemistry Analyzer (Boehringer Mannheim Diagnostics, Indianapolis, IN). Assays of overnight urine collections (urine volume, pH, color, ketone, bilirubin, occult blood, and glucose) were conducted using N-Multistix@C reagent strip (Miles Laboratories, Inc., Elkhart, IN). Gross histopathoiogy Animals were sacrificed over a 2-day period, following the 13th and 14th exposures. On each day the sacrifices were equally divided by sex and group. Lungs, liver, brain, heart, adrenal glands, kidneys, testes with epididymides, pituitary glands, thyroid glands, and spleen were weighed for all animals. Organ to terminal body weight ratios were calculated. A complete gross postmortem was conducted on all animals. The following tissues from all animals were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned and stained with hematoxylin and eosin: nasal passages, trachea, tracheobronchial lymph nodes, lungs, heart, aorta, esophagus, stomach, duodenum, jejunum, ileum, cecum, colon, rectum, liver, salivary glands, kidneys, urinary bladder, pituitary gland, adrenal glands, thyroid glands, thymus, spleen, mesenteric lymph nodes, brain, spinal cord (cervical, thoracic and lumbar regions), sciatic nerve, testes, epididymides, ovaries, uterus and gross lesions. All tissues
161
from
the high-exposure
and control
groups
were examined
microscopically.
The
lungs, trachea, tracheobronchial lymph nodes, nasal turbinates and gross lesions were examined from all animals in the lower- and middle-exposure groups. In addition, two sections of formalin-fixed liver tissue (frozen and stained with Oil Red-O), and a smear of femoral bone marrow (stained with May-Griinwald-Giemsa from the control and high-exposure groups were examined microscopically.
stain)
Statistical analysis Mean body weight prior to the initiation of exposures and on days 1, 4, 7, 11 and 15, total body weight gain (initial body weight through day 15), weekly feed consumption, total feed consumption, weekly water consumption, clinical chemistry and hematology data (except cell morphology gradings), and organ weight data from the treated animals were compared with the control animals of the same sex. Data were evaluated using tests for homogeneity of variances and one-way ANOVA at the 5.0% one-tailed probability level and a two-tail Dunnett’s t-test at the 5.0% probability level. RESULTS
Bacterial viability Daily reconstituted test material averaged 5.0 x lo4 viable organisms per gram of dried culture. Averages of 7, 129, 352 and 585 viable fluorescing pseudomonad cells per 135-liter atmospheric sample were recovered from the 0.0, 0.0008, 0.4 and 0.8 g/l P. syringae inhalation chamber atmospheres, respectively. Chamber atmospheres Particle size distribution determinations indicated that the test aerosols were within the respirable size range for rats, which is generally considered to be < 10 pm. The values for the 3 treated groups ranged from 1.23-l .66 pm for the aerodynamic mass median diameter and 2.64.8 for the geometric standard deviation. Total protein exposures (corrected for background from the control group) were approximately 39 and 78 ,ug per 135-liter air sample from the 0.4 and 0.8 g/l groups, respectively. The analytic concentration based on the protein assay was about 700 mg/m3 for both the 0.4 and 0.8 g/l suspensions. The protein levels of the 0.0 and the 0.0008 g/l test groups were below the sensitivity of the Bio-Rad protein assay (1 pg/ml). However, based on the similarity of the nominal concentrations (24 910 to 26400 mg/m3) among the 4 groups, it was concluded that rats exposed to the 0.0008 g/l and 0.0 g/l aerosols also received about 700 mg/m3 of bacterial suspension or spring water, respectively. Lipopolysaccharide (endotoxin) assay The mean and standard deviation of the 8 tested dried culture syringae (strain 3 1a) were 19 + 6 mg LPS/g of dried culture.
preparations
of P.
162
Clinical observations and clinical pathology All animals survived to termination and showed no biologically significant differences between the treated and control animals in physical and behavioral observations, body weight, feed consumption, water consumption, hematology, clinical chemistry or urinalysis. Organ weight analysis There were no treatment-related absolute or relative weight differences observed in any organ except the lungs. Mean absolute and relative lung weights of male and female rats in the mid- and high-concentration groups were statistically significantly increased (except for the absolute lung weights for the mid-concentration females) in a concentration-related manner compared to the control groups (Table I). Although lung weights for rats of both sexes in the low-concentration group were slightly higher than those of the control groups, the weights were not statistically significantly different from controls, nor were the weight differences considered treatmentrelated or biologically significant. Gross pathology Two male animals (one male animal each from the mid- and high-concentration groups) were found with enlarged mesenteric lymph nodes, and several male and female animals in the same groups had enlarged or unequally sized tracheobronchial lymph nodes. The findings in the tracheobronchial lymph nodes were considered to be treatment-related. Other findings at necropsy were considered sporadic and not treatment-related. Histopathology The lamina propria of the nasal turbinates contained a minimal to slight infiltrate of lymphocytes and a few neutrophils in all the mid- and high-con~ntration animals {Table II). Severity was mildly increased in the high-concentration animals when compared to the mid-concentration animals. No subacute inflammation was noted in either the control or low-concentration animals. In all the mid- and high-concentration animals examined, the epithelium lining the trachea exhibited hypertrophy (Table III). The severity (minimal to moderate) was similar in the two groups. No hypertrophy was noted in the control or the low-concentration animals. Pulmonary changes included a mild increase in the severity of chronic inflammation and a mild increase in the incidence and severity of macrophages in the alveoli (Table IV). The mid- and high-concentration rats also had minimal to moderately severe peribronchiolar lymphoid hyperplasia, and minimal to moderate hypertrophy of the bronchiolar epithelium. The pulmonary lesions were more pronounced in the high-concentration animals. In the tracheobronchial iymph node, minimal to moderate l~phoid hype~lasia
163
TABLE
I
MEAN
LUNG
WEIGHT
AND
LUNG/BODY
WEIGHT
RATIOS
OF RATS EXPOSED
TO P.SYR-
INGAE Concentration bacterial
(g/l) of
Lung weight(g)
Lung/body
weight ratio (%)
suspension Male
Female
Male
Female
0.0
1.302
1.045
0.385
0.497
0.0008
1.359
1.057
0.407
0.511
0.4
1.504
1.145
0.478’
0.565’
0.8
1.744
1.358’
0.527’
0.653’
* Statistically TABLE
significant
difference
from 0.0 g/l group at a = 0.05.
II
SUMMARY
OF NASAL
TURBINATE
HISTOPATHOLOGY
INCIDENCE Female
Male Concentration (g/l) of bacterial suspension Number
of animals
examined
0.0
0.0008
IO
10
Subacute inflammation Minimal
0.4 10
0.8
0.0
0.0008
0.4
0.8
IO
10
10
IO
10
0’
0
8
0
0
0
8
2
Slight
0
0
2
10
0
0
2
8
Total
0
0
10
10
0
0
10
10
*Number TABLE
of animals
exhibiting
response
III
SUMMARY
OF TRACHEAL
HISTOPATHOLOGY
INCIDENCE
Male
Female
Concentration (g/l) of bacterial suspension
0.0
Number
9’
10
Minimal
0 .*
Slight
0
Moderate Total
of animals
Hypertrophy,
*Trachea **Number
examined
0.0008
0.4
0.8
9’
10
0
3
0
6
0
0
0
0
0.0
0.0008
0.4
0.8
10
10
10
10
2
0
0
3
4
7
0
0
7
6
0
1
0
0
0
0
9
10
0
0
10
10
epithelium
was not in plane of section for one rat. of animals
exhibiting
response.
164
TABLE
IV
SUMMARY
OF LUNG HISTOPATHOLOGY Male
Concentration -
bacterial
suspension
Number
of animals
Chronic
Female
(g/l) of 0.0 examined ..-
10
0.0008
0.4
0.8
IO
10
IO
0.0
0.0008
IO _
IO
0.4
0.8
IO
IO
Minimal
8’
7
0
0
7
6
1
0
Slight
0
0
6
0
0
0
6
4
Moderate
0
0
4
IO
0
0
3
6
Total
8
7
10
10
I
6
IO
10
Minimal
6
4
5
0
4
2
0
4
Slight
0
0
5
IO
0
0
IO
6
Total
6
4
IO
IO
4
2
IO
IO
Minimal
0
0
1
0
0
0
4
0
Slight
0
0
7
2
0
0
6
9
Moderate
0
0
2
6
0
0
0
1
0
0
0
2
0
0
0
0
0
0
IO
IO
0
0
JO
10
Minimal
0
0
2
2
0
0
4
2
Slight Moderate
0 0
0 0
8 0
8 0
0 0
0 0
6 0
7
Total
0
0
10
to
0
0
IO
10
Alveolar
macrophages
Peribronchial
Moderately
lymphoid
hyperplasia
severe
Total Bronchiolar
*Number
-.-.
inflammation
epithelium
of animals
hypertrophy
exhibiting
I
response.
occurred in all nodes examined from male rats exposed to the middle and high concentrations and 8 of 10 and 8 of 8 nodes examined from the female rats exposed to the middle and high concentrations, respectively (Table V). Severity was slightly greater in the high-concentration females than in the mid-concentration females. Minimal lymphoid hyperplasia was noted in 1 low-concentration male and in 2 low-concentration females. These findings in rats exposed at the low concentration were not considered to be of biological significance to this study, however, since lymphoid hyperplasia is an occasional incidental finding in normal rats.
t65
TABLE
V
SUMMARY
OF TRACHEOBRONCHIAL
of animals
NODE
HISTOPATHOLOGY
examined’
Lymphoid hyperplasia Minimal
INCIDENCE
Female -.
Male __-.--.
Concentration (g/l) of bacterial susnension Number
LYMPH -
0.0
0.0008
0.4
0.8
0.0
0.0008
3
5
8
8
7
8
10
8
0.4
0.8
0 .I
1
I
0
0
4
0 0
5
0
1 I
2
Slight
0
3
6
Moderate
0
0
3
6
0
0
0
2
Total
0
I
8
8
0
2
8
8
*Some specimens ‘*Number
submitted
of animals
were not examined
exhibiting
because
the nodes were not in the plane of section.
response.
DISCUSSION
The advent of commercialization of new biologically derived products has created concerns about their safety. Even materials that have not been bioengineered, such as strain 31a, have come under increased scrutiny. The primary health-related concerns expressed about naturally occurring bacterial organisms such as strain 3 1a have centered around questions concerning infectivity, pathogenicity, and endotoxin exposure. P. syringae occurs ubiquitously in nature and is found at concentrations of 401000 ice-nucleation bacteria/m3 in the atmosphere where it contributes to background levels of ice-nucleators 112, 131. Challenge of 85 ~onomically important plants with strain 31a demonstrated no level of pathogenicity adequate to assign strain 3 la to a specific host plant [ 14, 151. Fate studies have demonstrated that strain 31a has low survivability in alpine soils, water and man-made snow [16]. P. syringae has not been reported to be a pathogen for humans or animals, Lipopolysaccharides, or endotoxins, from gram-negative bacteria are biologically active substances and are presumed to be the agents that cause injury to the lungs in people exposed to gram-negative bacteria. For example, endotoxin has been associated with byssinosis [ 171, febrile reactions [ 1S] and allergic alveolitis [ 191. Endotoxin concentrations in dust associated with these responses have been reported in the range of 0.1-10 pg/m3 [20], 0.006-0.78 pg/m3 [21] and 0.01-50 pg/m3 [22]. Yet, there is evidence that administration of endotoxin alone as nebulized solutions at these concentrations is not sufficient to cause respiratory symptoms [23, 241, suggesting that reactions to bacteria-laden dusts or mists might not be entirely associated with endotoxin but rather with other components in the mixtures or that other components in the mixtures may act as co-factors enhancing the potency of endotoxin. This suggestion has found some support in a ¢ study of farm environments [22]
166
in which extremely
high and variable
concentrations
of endotoxin
in air (from < 0.01
to > 50 pg/m3, median 42 pg/m3, geometric mean 29 pg/m3) produced no symptoms of febrile reactions or allergic alveolitis in exposed individuals, while farm environments with lower endotoxin concentrations in air (from < 0.01 to > 50 pg/m3, median 6.4 pg/m3, geometric mean 2.2 pg/m3) were associated with these symptoms. Although these studies do not eliminate endotoxins as a causal factor in respiratory responses to farm dust, they do suggest that respiratory toxic reactions may be caused by components other than endotoxins. In this study, the endotoxin content of strain 31a was 19 mg/g of dry cells. The animals administered 700 mg/m3 of the 0.8 g/l (0.08%) bacterial suspension were exposed to about 0.6 mg/m3 of strain 31a (0.0008 x 700 mg/m3). Assuming that each strain 31a bacterium is 100% endotoxin, this is equivalent to 11 yg of endotoxin/m3 (0.6 mg/m3 x 19 mg endotoxin/g dry cells x g/1000 mg). However, there was no evidence of infectivity or pathogenicity at this concentration (0.8 g/l of strain 3la), which was 1000 times higher than that of animals exposed to 0.0008 g/l, the concentration typical for the discharge mouth of a snow gun. Moreover, actual endotoxin concentration may have been underestimated since animals were exposed to nominal concentrations of 25 000 mg/m3 of the bacterial suspensions. In previously conducted toxicity studies, strain 31a has exhibited a low order of toxicity. It has an oral LDse of greater than 5.0 g/kg and liquid aerosol and dry dust inhalation LCses of greater than 5.0 mg/l in rats. It is a slight skin and eye irritant in rabbits and has not induced dermal sensitization in a standard guinea-pig assay. The present study of rats exposed to about 700 mg/m3 of 0.0008, 0.4 or 0.8 g/l bacterial suspensions for 2 h per day, 5 days per week for 13-14 exposures also did not demonstrate any evidence of toxicity. The primary response from exposure to strain 31a at high levels was a mild irritant reaction considered typical for deposition of particulate material in respiratory passages. No evidence was found in the current study for the infectivity, pathogenicity or toxicity of P. syringae in rats repetitively exposed to aerosol concentrations of about 700 mg/m3 of 0.0008,0.4 or 0.8 g/l of aqueous bacterial suspensions. These concentrations were 500 times (0.4 g/l) and 1000 times (0.8 g/l) the concentration typical for the discharge mouth of a snow gun (0.0008 g/l). ACKNOWLEDGEMENTS
The authors thank W. Newman, J. Morris, D.L. Thomas, D.R. Jagannath, P.A. Miller, R. McCrea and V. Bialecki for technical assistance, and D. Warren and C. Pergolizzi for preparation of the manuscript. REFERENCES 1 Maki, L.R., Galyan, E.L., Mei-Mon, CC. and Calwell, D.R. (t974) Ice nucleation ~~~~~ syringae. Appl. Microbial. 28,456-4.59.
induced
by Pseudo-
167
2 Chen, J. and Kevorkian, V. (1968) Mass production of 300-micron water droplets by air-water twophase nozzles. Ind. Eng. Chem. Process Des. Dev. 7,586590. 3 Chen, J. and Kevorkian. V. (1971) Heat and mass transfer making artificial snow. Ind. Eng. Chem. Process Des. Dev. IO, 75-78. 4 Knight, C. and Knight, N. (1973) Snow crystals. Sci. Am. 228, lO&IO7. 5 Eisel, M.L., Kimberly, D.M. and Leaf, CF. (1988) Estimated consumption loss from man-made snow. Water Res. Bull. Am. Water Res. Assoc. 24, 815-819. 6 Brigham, K.L. and Meyrick, B. (1986) Endotoxin and lung injury. Am. Rev. Respir. Dis. 133,913-927. 7 Rylander, R. and Lundholm, M. (1978) Bacterial contamination of cotton and cotton dust and effects on the lung. Br. J. fnd. Med. 35,204207. 8 Hefander, I., Salkinoja-SaIonen, M. and Rylander, R. (1980) Chemical structure and inhalation toxicity of lipopolysaccharide from bacteria on cotton. Infect. Immun. 29,859-862. 9 Helander, I., Salkinoja-Salonen, M. and Rylander, R. (1982) Pulmonary toxicity of endotoxins: comparison of lipopolysacharides from various bacterial species. Infect. Immun. 35, 5288532. 10 Koch, A.L. (1981) Growth measurement, In: P. Herhardt, R.G.E. Murray, R.N. Costilow, E.W. Nester, W.A. Wood, N.R. King and G.B. Phillips (Eds.), Manual of Methods for General Bacteriology. American Society for Microbiology, Washington, DC, pp. 185IS6 11 Levin, J. and Bang, F. (1964) The role of endotoxin in the extracellular coagulation of limulus blood. Bull. Johns Hopkins Hosp. 115,265-271. 12 Flanagan, P.W. and Jayaweera, K. (1980) Evidence for microbial ice nucleation over the arctic ocean. EOS 61,968. 13 Schnell, R.C., Miller, SW. and Allee, P.A. (1981) Ice nucleation studies on aerosols. In: Abstracts, 15th Conference on Agriculture, Forestry and Meteorology and 5th Conference on Biometeorology, pp. 22-24. 14 Harrison, M.D. (1988) Evaluation of the pathogenicity and ecological characteristics of P. syringae strain 31a: a potential pathogenicity of P. syringae 31a on 19 host plants. Colorado State University, Fort Collins. Unpublished data, Bio-Products Division, Eastman Kodak Company, Rochester, NY 1465@0820. 15 Lindow, S.E. (1984) Evaluation of P. syringue 31a for plant pathogenicity. University of California Berkeley. Unpublished data, Bio-Products Division, Eastman Kodak Company, Rochester, NY 14650-0820. 16 Goodnow, R.A., Harrison, M.D., Morris, J.D., Sweeting, K.B. and LaDuca, R.J. (1990) Fate of ice nucleation active Pseudomonas syringue strains in alpine soils and waters and in synthetic snow samples Appl. Environ. Microbial. 56, 2223-2227. 17 Rylander, R. (1981) Bacterial toxins and etiology of byssinosis. Chest (Suppl) 79,34-38. 18 Clark, S., Rylander, R. and Larsson, L. (1983) Airborne bacteria, endotoxin and fungi in dust of poultry and swine confinement buildings, Am. Ind. Hyg. Assoc. J. 44,537-541. 19 DoPico, G.A., Reddan, W., Flaherty, D., Reed, C. and Tsiatis, A. (1986) Health effects of occupational grain dust exposure. Am. J. Ind. Med. 10,298-299. 20 Haglind, P. and Rylander, R. (1984) Exposure to cotton dust in an experimental cardroom. Br. J. Ind. Med. 41,340-345. 21 Castellan, R.M., Olenchock, S.A., Kinsley, K.B. and Hankinson, J.L. (1987) Inhaled endotoxin and decreased spirometric values, an exposure-response relation for cotton dust. N. Engl. J. Med. 3 17, 6055610. 22 Rask-Andersen, A., Malmberg, P. and Lundholm, M. (1989) Endotoxin levels in farming: absence of symptoms despite high exposure levels. Br. J. Ind. Med. 46,412416. 23 Cavagna, G., Foa, V. and Vigliani, E.C. (1969) Effects in man and rabbits of inhalation of cotton dust or extracts and purified endotoxins. Br. J. Ind. Med. 26, 314-321. 24 Haglind, P., Bake, B. and Rylander, R. (1984) Effects of endotoxin inhalation challenges in humans. In: P.J. Wakelyn and R.R. Jacobs (Eds.), Proceedings, 8th Cotton Dust Research Conference, Atlanta, Georgia, pp. 105-106.
157
Elsevier TOXLET 02459
Subacute inhalation toxicity study of an ice-nucleation-active P~~~d~~~~~s syringne administered as a respirable aerosol to rats
R.A. Goodnowl,
G. Katz2, D.C. Haines3 and J.B. TerriIP
‘Genencor International, Inc., Rochester. NY, ‘Eastman Kodak Company, Roehester, NY and ‘Ha&ton Laboratories America, Inc., RockviNe. MD (U.S.A.)
(Received 6 March 1990) (Revision received 7 May 1990) (Accepted 25 July 1990) Key words: Pseudomonas syringae: Ice-nucleation active; Respirable aerosol; Inhalation; Snow making; Biotechnology
SUMMARY The inhalation toxicity of a commercial sample of an ice-nucleation-active Pse~domonus syringae (strain 3 1a) was evaluated by repetitively exposing rats to about 700 mg/m’ of an aerosol consisting of a suspension of 0.0008, 0.4 or 0.8 g/l of bacteria in water for 2 h per day, 5 days per week for 13-14 exposures. No mortality, moribundity or biologically significant differences in clinical signs, body weight, food consumption or clinical pathology were observed. Animals tested at 500 times (0.4 g/l) and 1000 times (0.8 g/l) the recommended ice-nucleation concentration (0.0008 g/l) exhibited concentration-dependent increased lung weights. Several animals exhibited enlarged tracheobronchial lymph nodes. The pulmonary responses observed are considered compatible with a mild irritant reaction. There was no evidence of bacterial infection. Animals tested at a concentration typical for the discharge mouth of a snow gun (0.0008 g/l) demonstrated no significant biological effect.
INTRODUCTION
Specific isolates of the gram-negative fluorescing bacteria, Pseudomonas syringae, due to their ubiquity and high ice-nucleation activity, are considered one of the more important sources of natural ice nuclei [I]_ The recent commercial use of ice-nucleation-active P. syringae in water as an ice nucleator in the manufactu~ng of man-made Address.for correspondence:
Robert A. Goodnow, Genencor International, Inc., 1700 Lexington Avenue, Rochester, NY 14650-3605, U.S.A.
158
snow allows freezing to begin in -3 to - 5°C water compared to -8 to - 10°C in water without the presence of P. syringae. Although snow makers and skiers may be exposed to plumes emitted from snow guns, snow guns are designed to emit water droplets averaging 300 pm in diameter, a size that is generally considered non-respirable [2, 31. Moreover, the supercooled droplets quickly form ice or snow crystals that average greater than 3000 pm in diameter [4]. However, Eisel et al. [5] concluded that about 6% of the water may be lost from man-made snow through evaporation and sublimation, thereby reducing droplet diameter and exposing the snow maker and skier to respirable-size bacterial cells and cellular components. Public health officials have expressed concern that snow makers and skiers may exhibit adverse health effects if exposed to snow plumes containing gram-negative bacterial ice nucleators. This concern arises from reports of occupational lung disease in textile workers that was thought to be related to exposure to endotoxin from gram-negative bacteria 1691. The acute toxicity of an ice-nucleation-active P. syringae (strain 31a) has been studied in several animal species [unpublished data]. The oral LD50 value of the commercial material (strain 31a) for rats was greater than 5.0 g/kg. The inhalation L&o value for commercial material was greater than 5.5 mg/l for 2 h (aerodynamic mass median diameter, 10 pm; geometric standard deviation, 5.0) and greater than 5.1 mg/l for 4 h (aerodynamic mass median diameter, 3.2 pm; geometric standard deviation, 3.0) of exposure to a dry dust atmosphere, and greater than 5.3 mg/m’ of a 0.8 g/l aqueous suspension of strain 31a for 4 h of exposure (aerodynamic mass median diameter not reported). The acute dermal LDSs of strain 31a in rabbits was greater than 2.0 g/l, and the strain was considered non-irritating to the skin of rabbits. Strain 3 la did not cause dermal contact sensitization in guinea-pigs. It produced minor and transient changes of the conjunctiva in only 1 of 6 rabbits tested for eye irritancy. The objective of the following study was to determine the toxicity of strain 31a following repeated daily inhalation exposure corresponding to a 5-day work-week for people who make snow. In addition, the exposure concentrations and duration were selected to exceed those anticipated for skiers exposed to snow plumes emitted from snow guns. MATERIAL
AND METHODS
Preparation of bacterial suspensions P. syringae (strain 31a) was propagated
under fermentation in a defined sucrosebased mineral salts medium. The culture was dewatered by ultrafiltration to approximately a 20% solids cell slurry, rapid-frozen in liquid nitrogen, and desiccated. This preparation was commercially available for use as an ice-nucleator in the manufacturing of man-made snow and was supplied by Bio-Products Division of Eastman Kodak Company, Rochester, NY 14650. Prior to each animal exposure, desiccated bacterial culture containing 1.G 2.0 x 10” organisms (approximately lo4 viable and lo* non-viable cells of strain 31a)
159
per gram was prepared concentrations
(8 g/l) and then diluted
of 0.0008,0.4
with sterile spring water to provide
and 0.8 g/l. These concentrations,
1, 500 and 1000 times
the concentrations discharged from a snow gun, were selected to provide exposure levels equal to anticipated exposures, plus exposure levels that would provide data on safe levels of exposure greater than that which was anticipated. Standard plate count techniques of stock bacterial
were employed to determine suspension [lo].
the colony-forming
units (cfu) per ml
Lipopolysaccharide (LPS) (endotoxin) assay Eight separate dried culture preparations of P. syringae strain 31a were analyzed in duplicate to determine the average LPS content per gram of dry culture using the Limulus Amebocyte Lysate (LAL) assay [ 111. Test animals Male and female Sprague-Dawley rats (Crl:CD@BR) were obtained from Charles River Laboratories, Inc., Raleigh, NC. Animals were approximately 7 weeks of age at the start of the study and were randomly assigned to exposure groups composed of 10 males and 10 females. Purina Certified Rodent Laboratory Chow #5002 and tap water were available ad libitum throughout the 3-week acclimation and non-exposure periods. Neither food nor water was available during each 2 h exposure period. Aerosol generation and analysis Four lOOO-liter stainless steel and glass inhalation exposure chambers were operated with total airflow rates through each chamber of 216, 221, 214 and 218222 liters per minute (lpm) for the 0.0, 0.0008, 0.4 and 0.8 g/l bacterial suspension groups, respectively. For aerosol generation, each bacterial suspension was drawn from a reservoir to a Spraying System atomizer (Wheaton, Illinois) by a fluid metering pump. The suspension was mixed with compressed air in the atomizer to yield the bacterial aerosol, which was further diluted with air, as necessary, to provide the target concentration. The control group animals were exposed to an aerosol of spring water. All groups were exposed for 2 h daily, 5 days per week for 13-14 exposures. Exposure concentrations were measured by collecting samples of each chamber atmosphere and then assaying the samples for total protein concentration and viable fluorescing pseudomonads (colony-forming units). Protein concentration was analytically determined using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Richmond, CA) and pseudomonad colony-forming units were enumerated using a standard plate count technique. Particle size distribution was determined twice daily using an aerodynamic particle sizer (TSI Aerodynamic Particle Sizer, Model #3301/3302). Nominal aerosol concentrations (calculated from the ratio of test material used and total airflow) were determined daily. Homogeneity of aerosol distribution in each chamber was determined twice weekly.
160
Clinical observations All rats were observed twice daily for mortality and moribundity. Animals were observed once daily (immediately after exposure) for clinical manifestations of adverse health effects resulting from exposure to P. syringae. Animal observations included, but were not limited to, examination for clinical abnormalities associated with respiratory, circulatory and central nervous systems, hair, skin, eyes, feces and urine. Body weight measurements were taken prior to randomization, at test initiation {prior to and after the first exposure), and twice weekly thereafter. Feed consumption measurements were recorded weekly. Water consumption measurements were recorded daily. Clinical pathology
Animals were fasted overnight and anesthetized with ketamine, and blood samples were collected from the orbital sinus. Animals were sacrificed by exsanguination following intraperitonal injection of sodium pentobarbital (4 mg/lOO g body wt.). Red and white blood cell counts, hemoglobin concentration, hematocrit, red cell indices, and platelet count were performed using a Coulter Counter (Model S+IV System, Coulter Electronics, Inc., Hialeah, FL). Prothrombin, activated partial thromboplastin time, and clotting time were performed by Coag-A-MateTM x 2 (General Diagnostics, Morris Plains, NJ). Differential white blood cell counts were performed by microscopy. Clinical chemistry dete~inations (sodium, potassium, chloride, total protein, albumin, calcium, phosphorus, lactic dehydrogenase, total bilirubin, urea nitrogen, cholesterol, glucose, asparate aminotransferase, alanine aminotransferase, and alkaline phosphatase) were conducted using a BMD/Nitachi 737 Chemistry Analyzer (Boehringer Mannheim Diagnostics, Indianapolis, IN). Assays of overnight urine collections (urine volume, pH, color, ketone, bilirubin, occult blood, and glucose) were conducted using N-Multistix@C reagent strip (Miles Laboratories, Inc., Elkhart, IN). Gross histopathoiogy Animals were sacrificed over a 2-day period, following the 13th and 14th exposures. On each day the sacrifices were equally divided by sex and group. Lungs, liver, brain, heart, adrenal glands, kidneys, testes with epididymides, pituitary glands, thyroid glands, and spleen were weighed for all animals. Organ to terminal body weight ratios were calculated. A complete gross postmortem was conducted on all animals. The following tissues from all animals were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned and stained with hematoxylin and eosin: nasal passages, trachea, tracheobronchial lymph nodes, lungs, heart, aorta, esophagus, stomach, duodenum, jejunum, ileum, cecum, colon, rectum, liver, salivary glands, kidneys, urinary bladder, pituitary gland, adrenal glands, thyroid glands, thymus, spleen, mesenteric lymph nodes, brain, spinal cord (cervical, thoracic and lumbar regions), sciatic nerve, testes, epididymides, ovaries, uterus and gross lesions. All tissues
161
from
the high-exposure
and control
groups
were examined
microscopically.
The
lungs, trachea, tracheobronchial lymph nodes, nasal turbinates and gross lesions were examined from all animals in the lower- and middle-exposure groups. In addition, two sections of formalin-fixed liver tissue (frozen and stained with Oil Red-O), and a smear of femoral bone marrow (stained with May-Griinwald-Giemsa from the control and high-exposure groups were examined microscopically.
stain)
Statistical analysis Mean body weight prior to the initiation of exposures and on days 1, 4, 7, 11 and 15, total body weight gain (initial body weight through day 15), weekly feed consumption, total feed consumption, weekly water consumption, clinical chemistry and hematology data (except cell morphology gradings), and organ weight data from the treated animals were compared with the control animals of the same sex. Data were evaluated using tests for homogeneity of variances and one-way ANOVA at the 5.0% one-tailed probability level and a two-tail Dunnett’s t-test at the 5.0% probability level. RESULTS
Bacterial viability Daily reconstituted test material averaged 5.0 x lo4 viable organisms per gram of dried culture. Averages of 7, 129, 352 and 585 viable fluorescing pseudomonad cells per 135-liter atmospheric sample were recovered from the 0.0, 0.0008, 0.4 and 0.8 g/l P. syringae inhalation chamber atmospheres, respectively. Chamber atmospheres Particle size distribution determinations indicated that the test aerosols were within the respirable size range for rats, which is generally considered to be < 10 pm. The values for the 3 treated groups ranged from 1.23-l .66 pm for the aerodynamic mass median diameter and 2.64.8 for the geometric standard deviation. Total protein exposures (corrected for background from the control group) were approximately 39 and 78 ,ug per 135-liter air sample from the 0.4 and 0.8 g/l groups, respectively. The analytic concentration based on the protein assay was about 700 mg/m3 for both the 0.4 and 0.8 g/l suspensions. The protein levels of the 0.0 and the 0.0008 g/l test groups were below the sensitivity of the Bio-Rad protein assay (1 pg/ml). However, based on the similarity of the nominal concentrations (24 910 to 26400 mg/m3) among the 4 groups, it was concluded that rats exposed to the 0.0008 g/l and 0.0 g/l aerosols also received about 700 mg/m3 of bacterial suspension or spring water, respectively. Lipopolysaccharide (endotoxin) assay The mean and standard deviation of the 8 tested dried culture syringae (strain 3 1a) were 19 + 6 mg LPS/g of dried culture.
preparations
of P.
162
Clinical observations and clinical pathology All animals survived to termination and showed no biologically significant differences between the treated and control animals in physical and behavioral observations, body weight, feed consumption, water consumption, hematology, clinical chemistry or urinalysis. Organ weight analysis There were no treatment-related absolute or relative weight differences observed in any organ except the lungs. Mean absolute and relative lung weights of male and female rats in the mid- and high-concentration groups were statistically significantly increased (except for the absolute lung weights for the mid-concentration females) in a concentration-related manner compared to the control groups (Table I). Although lung weights for rats of both sexes in the low-concentration group were slightly higher than those of the control groups, the weights were not statistically significantly different from controls, nor were the weight differences considered treatmentrelated or biologically significant. Gross pathology Two male animals (one male animal each from the mid- and high-concentration groups) were found with enlarged mesenteric lymph nodes, and several male and female animals in the same groups had enlarged or unequally sized tracheobronchial lymph nodes. The findings in the tracheobronchial lymph nodes were considered to be treatment-related. Other findings at necropsy were considered sporadic and not treatment-related. Histopathology The lamina propria of the nasal turbinates contained a minimal to slight infiltrate of lymphocytes and a few neutrophils in all the mid- and high-con~ntration animals {Table II). Severity was mildly increased in the high-concentration animals when compared to the mid-concentration animals. No subacute inflammation was noted in either the control or low-concentration animals. In all the mid- and high-concentration animals examined, the epithelium lining the trachea exhibited hypertrophy (Table III). The severity (minimal to moderate) was similar in the two groups. No hypertrophy was noted in the control or the low-concentration animals. Pulmonary changes included a mild increase in the severity of chronic inflammation and a mild increase in the incidence and severity of macrophages in the alveoli (Table IV). The mid- and high-concentration rats also had minimal to moderately severe peribronchiolar lymphoid hyperplasia, and minimal to moderate hypertrophy of the bronchiolar epithelium. The pulmonary lesions were more pronounced in the high-concentration animals. In the tracheobronchial iymph node, minimal to moderate l~phoid hype~lasia
163
TABLE
I
MEAN
LUNG
WEIGHT
AND
LUNG/BODY
WEIGHT
RATIOS
OF RATS EXPOSED
TO P.SYR-
INGAE Concentration bacterial
(g/l) of
Lung weight(g)
Lung/body
weight ratio (%)
suspension Male
Female
Male
Female
0.0
1.302
1.045
0.385
0.497
0.0008
1.359
1.057
0.407
0.511
0.4
1.504
1.145
0.478’
0.565’
0.8
1.744
1.358’
0.527’
0.653’
* Statistically TABLE
significant
difference
from 0.0 g/l group at a = 0.05.
II
SUMMARY
OF NASAL
TURBINATE
HISTOPATHOLOGY
INCIDENCE Female
Male Concentration (g/l) of bacterial suspension Number
of animals
examined
0.0
0.0008
IO
10
Subacute inflammation Minimal
0.4 10
0.8
0.0
0.0008
0.4
0.8
IO
10
10
IO
10
0’
0
8
0
0
0
8
2
Slight
0
0
2
10
0
0
2
8
Total
0
0
10
10
0
0
10
10
*Number TABLE
of animals
exhibiting
response
III
SUMMARY
OF TRACHEAL
HISTOPATHOLOGY
INCIDENCE
Male
Female
Concentration (g/l) of bacterial suspension
0.0
Number
9’
10
Minimal
0 .*
Slight
0
Moderate Total
of animals
Hypertrophy,
*Trachea **Number
examined
0.0008
0.4
0.8
9’
10
0
3
0
6
0
0
0
0
0.0
0.0008
0.4
0.8
10
10
10
10
2
0
0
3
4
7
0
0
7
6
0
1
0
0
0
0
9
10
0
0
10
10
epithelium
was not in plane of section for one rat. of animals
exhibiting
response.
164
TABLE
IV
SUMMARY
OF LUNG HISTOPATHOLOGY Male
Concentration -
bacterial
suspension
Number
of animals
Chronic
Female
(g/l) of 0.0 examined ..-
10
0.0008
0.4
0.8
IO
10
IO
0.0
0.0008
IO _
IO
0.4
0.8
IO
IO
Minimal
8’
7
0
0
7
6
1
0
Slight
0
0
6
0
0
0
6
4
Moderate
0
0
4
IO
0
0
3
6
Total
8
7
10
10
I
6
IO
10
Minimal
6
4
5
0
4
2
0
4
Slight
0
0
5
IO
0
0
IO
6
Total
6
4
IO
IO
4
2
IO
IO
Minimal
0
0
1
0
0
0
4
0
Slight
0
0
7
2
0
0
6
9
Moderate
0
0
2
6
0
0
0
1
0
0
0
2
0
0
0
0
0
0
IO
IO
0
0
JO
10
Minimal
0
0
2
2
0
0
4
2
Slight Moderate
0 0
0 0
8 0
8 0
0 0
0 0
6 0
7
Total
0
0
10
to
0
0
IO
10
Alveolar
macrophages
Peribronchial
Moderately
lymphoid
hyperplasia
severe
Total Bronchiolar
*Number
-.-.
inflammation
epithelium
of animals
hypertrophy
exhibiting
I
response.
occurred in all nodes examined from male rats exposed to the middle and high concentrations and 8 of 10 and 8 of 8 nodes examined from the female rats exposed to the middle and high concentrations, respectively (Table V). Severity was slightly greater in the high-concentration females than in the mid-concentration females. Minimal lymphoid hyperplasia was noted in 1 low-concentration male and in 2 low-concentration females. These findings in rats exposed at the low concentration were not considered to be of biological significance to this study, however, since lymphoid hyperplasia is an occasional incidental finding in normal rats.
t65
TABLE
V
SUMMARY
OF TRACHEOBRONCHIAL
of animals
NODE
HISTOPATHOLOGY
examined’
Lymphoid hyperplasia Minimal
INCIDENCE
Female -.
Male __-.--.
Concentration (g/l) of bacterial susnension Number
LYMPH -
0.0
0.0008
0.4
0.8
0.0
0.0008
3
5
8
8
7
8
10
8
0.4
0.8
0 .I
1
I
0
0
4
0 0
5
0
1 I
2
Slight
0
3
6
Moderate
0
0
3
6
0
0
0
2
Total
0
I
8
8
0
2
8
8
*Some specimens ‘*Number
submitted
of animals
were not examined
exhibiting
because
the nodes were not in the plane of section.
response.
DISCUSSION
The advent of commercialization of new biologically derived products has created concerns about their safety. Even materials that have not been bioengineered, such as strain 31a, have come under increased scrutiny. The primary health-related concerns expressed about naturally occurring bacterial organisms such as strain 3 1a have centered around questions concerning infectivity, pathogenicity, and endotoxin exposure. P. syringae occurs ubiquitously in nature and is found at concentrations of 401000 ice-nucleation bacteria/m3 in the atmosphere where it contributes to background levels of ice-nucleators 112, 131. Challenge of 85 ~onomically important plants with strain 31a demonstrated no level of pathogenicity adequate to assign strain 3 la to a specific host plant [ 14, 151. Fate studies have demonstrated that strain 31a has low survivability in alpine soils, water and man-made snow [16]. P. syringae has not been reported to be a pathogen for humans or animals, Lipopolysaccharides, or endotoxins, from gram-negative bacteria are biologically active substances and are presumed to be the agents that cause injury to the lungs in people exposed to gram-negative bacteria. For example, endotoxin has been associated with byssinosis [ 171, febrile reactions [ 1S] and allergic alveolitis [ 191. Endotoxin concentrations in dust associated with these responses have been reported in the range of 0.1-10 pg/m3 [20], 0.006-0.78 pg/m3 [21] and 0.01-50 pg/m3 [22]. Yet, there is evidence that administration of endotoxin alone as nebulized solutions at these concentrations is not sufficient to cause respiratory symptoms [23, 241, suggesting that reactions to bacteria-laden dusts or mists might not be entirely associated with endotoxin but rather with other components in the mixtures or that other components in the mixtures may act as co-factors enhancing the potency of endotoxin. This suggestion has found some support in a ¢ study of farm environments [22]
166
in which extremely
high and variable
concentrations
of endotoxin
in air (from < 0.01
to > 50 pg/m3, median 42 pg/m3, geometric mean 29 pg/m3) produced no symptoms of febrile reactions or allergic alveolitis in exposed individuals, while farm environments with lower endotoxin concentrations in air (from < 0.01 to > 50 pg/m3, median 6.4 pg/m3, geometric mean 2.2 pg/m3) were associated with these symptoms. Although these studies do not eliminate endotoxins as a causal factor in respiratory responses to farm dust, they do suggest that respiratory toxic reactions may be caused by components other than endotoxins. In this study, the endotoxin content of strain 31a was 19 mg/g of dry cells. The animals administered 700 mg/m3 of the 0.8 g/l (0.08%) bacterial suspension were exposed to about 0.6 mg/m3 of strain 31a (0.0008 x 700 mg/m3). Assuming that each strain 31a bacterium is 100% endotoxin, this is equivalent to 11 yg of endotoxin/m3 (0.6 mg/m3 x 19 mg endotoxin/g dry cells x g/1000 mg). However, there was no evidence of infectivity or pathogenicity at this concentration (0.8 g/l of strain 3la), which was 1000 times higher than that of animals exposed to 0.0008 g/l, the concentration typical for the discharge mouth of a snow gun. Moreover, actual endotoxin concentration may have been underestimated since animals were exposed to nominal concentrations of 25 000 mg/m3 of the bacterial suspensions. In previously conducted toxicity studies, strain 31a has exhibited a low order of toxicity. It has an oral LDse of greater than 5.0 g/kg and liquid aerosol and dry dust inhalation LCses of greater than 5.0 mg/l in rats. It is a slight skin and eye irritant in rabbits and has not induced dermal sensitization in a standard guinea-pig assay. The present study of rats exposed to about 700 mg/m3 of 0.0008, 0.4 or 0.8 g/l bacterial suspensions for 2 h per day, 5 days per week for 13-14 exposures also did not demonstrate any evidence of toxicity. The primary response from exposure to strain 31a at high levels was a mild irritant reaction considered typical for deposition of particulate material in respiratory passages. No evidence was found in the current study for the infectivity, pathogenicity or toxicity of P. syringae in rats repetitively exposed to aerosol concentrations of about 700 mg/m3 of 0.0008,0.4 or 0.8 g/l of aqueous bacterial suspensions. These concentrations were 500 times (0.4 g/l) and 1000 times (0.8 g/l) the concentration typical for the discharge mouth of a snow gun (0.0008 g/l). ACKNOWLEDGEMENTS
The authors thank W. Newman, J. Morris, D.L. Thomas, D.R. Jagannath, P.A. Miller, R. McCrea and V. Bialecki for technical assistance, and D. Warren and C. Pergolizzi for preparation of the manuscript. REFERENCES 1 Maki, L.R., Galyan, E.L., Mei-Mon, CC. and Calwell, D.R. (t974) Ice nucleation ~~~~~ syringae. Appl. Microbial. 28,456-4.59.
induced
by Pseudo-
167
2 Chen, J. and Kevorkian, V. (1968) Mass production of 300-micron water droplets by air-water twophase nozzles. Ind. Eng. Chem. Process Des. Dev. 7,586590. 3 Chen, J. and Kevorkian. V. (1971) Heat and mass transfer making artificial snow. Ind. Eng. Chem. Process Des. Dev. IO, 75-78. 4 Knight, C. and Knight, N. (1973) Snow crystals. Sci. Am. 228, lO&IO7. 5 Eisel, M.L., Kimberly, D.M. and Leaf, CF. (1988) Estimated consumption loss from man-made snow. Water Res. Bull. Am. Water Res. Assoc. 24, 815-819. 6 Brigham, K.L. and Meyrick, B. (1986) Endotoxin and lung injury. Am. Rev. Respir. Dis. 133,913-927. 7 Rylander, R. and Lundholm, M. (1978) Bacterial contamination of cotton and cotton dust and effects on the lung. Br. J. fnd. Med. 35,204207. 8 Hefander, I., Salkinoja-SaIonen, M. and Rylander, R. (1980) Chemical structure and inhalation toxicity of lipopolysaccharide from bacteria on cotton. Infect. Immun. 29,859-862. 9 Helander, I., Salkinoja-Salonen, M. and Rylander, R. (1982) Pulmonary toxicity of endotoxins: comparison of lipopolysacharides from various bacterial species. Infect. Immun. 35, 5288532. 10 Koch, A.L. (1981) Growth measurement, In: P. Herhardt, R.G.E. Murray, R.N. Costilow, E.W. Nester, W.A. Wood, N.R. King and G.B. Phillips (Eds.), Manual of Methods for General Bacteriology. American Society for Microbiology, Washington, DC, pp. 185IS6 11 Levin, J. and Bang, F. (1964) The role of endotoxin in the extracellular coagulation of limulus blood. Bull. Johns Hopkins Hosp. 115,265-271. 12 Flanagan, P.W. and Jayaweera, K. (1980) Evidence for microbial ice nucleation over the arctic ocean. EOS 61,968. 13 Schnell, R.C., Miller, SW. and Allee, P.A. (1981) Ice nucleation studies on aerosols. In: Abstracts, 15th Conference on Agriculture, Forestry and Meteorology and 5th Conference on Biometeorology, pp. 22-24. 14 Harrison, M.D. (1988) Evaluation of the pathogenicity and ecological characteristics of P. syringae strain 31a: a potential pathogenicity of P. syringae 31a on 19 host plants. Colorado State University, Fort Collins. Unpublished data, Bio-Products Division, Eastman Kodak Company, Rochester, NY 1465@0820. 15 Lindow, S.E. (1984) Evaluation of P. syringue 31a for plant pathogenicity. University of California Berkeley. Unpublished data, Bio-Products Division, Eastman Kodak Company, Rochester, NY 14650-0820. 16 Goodnow, R.A., Harrison, M.D., Morris, J.D., Sweeting, K.B. and LaDuca, R.J. (1990) Fate of ice nucleation active Pseudomonas syringue strains in alpine soils and waters and in synthetic snow samples Appl. Environ. Microbial. 56, 2223-2227. 17 Rylander, R. (1981) Bacterial toxins and etiology of byssinosis. Chest (Suppl) 79,34-38. 18 Clark, S., Rylander, R. and Larsson, L. (1983) Airborne bacteria, endotoxin and fungi in dust of poultry and swine confinement buildings, Am. Ind. Hyg. Assoc. J. 44,537-541. 19 DoPico, G.A., Reddan, W., Flaherty, D., Reed, C. and Tsiatis, A. (1986) Health effects of occupational grain dust exposure. Am. J. Ind. Med. 10,298-299. 20 Haglind, P. and Rylander, R. (1984) Exposure to cotton dust in an experimental cardroom. Br. J. Ind. Med. 41,340-345. 21 Castellan, R.M., Olenchock, S.A., Kinsley, K.B. and Hankinson, J.L. (1987) Inhaled endotoxin and decreased spirometric values, an exposure-response relation for cotton dust. N. Engl. J. Med. 3 17, 6055610. 22 Rask-Andersen, A., Malmberg, P. and Lundholm, M. (1989) Endotoxin levels in farming: absence of symptoms despite high exposure levels. Br. J. Ind. Med. 46,412416. 23 Cavagna, G., Foa, V. and Vigliani, E.C. (1969) Effects in man and rabbits of inhalation of cotton dust or extracts and purified endotoxins. Br. J. Ind. Med. 26, 314-321. 24 Haglind, P., Bake, B. and Rylander, R. (1984) Effects of endotoxin inhalation challenges in humans. In: P.J. Wakelyn and R.R. Jacobs (Eds.), Proceedings, 8th Cotton Dust Research Conference, Atlanta, Georgia, pp. 105-106.