Active and passive immunization with Pseudomonas aeruginosa ribosomal vaccines and antisera in the burned rat model

JOURNAL OF SURGICAL RESEARCH 4,

138-

144 ( 1986)

Active and Passive Immunization with Pseudomonas aeruginosa Ribosomal Vaccines and Antisera in the Burned Rat Model’** MICHAEL M. LIEBERMAN,PH.D.,*,~ HARREL L. WALKER,M.S.,-~ ELEANORAYALA, B.S.,$ ANDISIDOROCHAPAS *Department of Clinical Investigation, Tripler Army Medical Center, Tripler AMC, Hawaii 96859-5000;f’lnstitute for Surgical Research,Fort Sam Houston, Texas 78234; and $Department of Clinical Investigation, Brooke Army Medical Center, Fort Sam Houston. Texas 78234 Submitted for publication October 18, I984 Pseudomonasaeruginosa ribosomal vaccines were tested for their ability to protect rats subjected to a 20% total body surfacebum against the lethal effectsof infection with homologous organisms. When administered prior to burning, the vaccinesprovided 100%protection. When administered postbuming, the vaccine from one strain also provided 100%protection when the time interval between vaccination and infection was 3 days. When this time interval was reduced to 1 or 2 days, approximately 50% protection was obtained with the same vaccine. The vaccine from a second strain tested provided about 50% protection with a 3day time interval. In addition, passive immunization using antiserum to a ribosomal vaccine was also demonstrated to be effective in protecting burned and infected rats, especially when multiple dosesof antiserum were used.In this case,8m protection was obtained (with no protection observedusing multiple dosesof normal serum).Finally, a comparison of ribosomal and hpopolysaccharide vaccines and antisera was aho performed. 0 1986 Academic Fvess, Inc.

tients with an establishedPseudomonasinfection. The vaccine could also be used to hyPseudomonasbacteremia presentsa partic- perimmunize groups of human volunteers. ularly difficult problem of medical manage- Plasma taken from these volunteers could be ment because of the high associated patient used to produce the hyperimmune globulin. mortality [ 13, 161and the frequent occurrence Previous studies from our laboratory have of multiple antibiotic resistant organisms [3, described the preparation, properties, and im231.Patients with extensive thermal injury are munogenicity of ribosomal vaccines from at high risk of infection with Pseudomonas Pseudomonas aeruginosa [7, 81. These vacaeruginosa [ 1, 161.Thus, these patients would cines were shown to provide effective protecbe prime candidates to receive Pseudomonas tion in laboratory animals against challenge vaccine or hyperimmune globulin. The vac- with whole (live) cultures of P. aeruginosa. cine would be used as prophylaxis while the Protection could be achieved by active imhyperimmune globulin could be used as either munization of mice [8, 91 or passive immuprophylaxis or specific therapy for those pa- nization of mice using antiserum or hyperimmune globulin raised in rabbits [lo] or mice [9] to the ribosomal vaccine. In addition, ’ The research was supported by the Department of polyvalent antisera to ribosomal vaccineswere Clinical Investigation, Brooke Army Medical Center, under prepared and the capability of this material to work unit Number C-64-82. The opinions or assertions contained herein are the private views of the authors and protect mice against randomly chosen cliniare not to be construed as reflecting the views of the De- cally isolated strains of P. aeruginosa was partment of the Army or the Department of Defense. demonstrated [ 111.Furthermore, the immu* This paper was presented in part at the 82nd Annual nogenicity of these ribosomal vaccines obMeeting of the American Society for Microbiology, 7- 12 served in C3H/HeJ mice [9] demonstrated that March 1982,Atlanta, Ga. the protection achieved cannot be ascribed to 3To whom reprint requestsshould be addressed. INTRODUCTION

0022-4804/86 $1.50 Copyright 8 1986 by Academic Press, Inc. All right.5 of reproduction in any fomt reserved.

138

LIEBERMAN ET AL.: Pseudomonas IMMUNITY

a small amount of lipopolysaccharide (LPS) present in the vaccine. The burned rat model for Pseudomonas infection has been developed in the U. S. Army Institute of Surgical Research [20]. This model closely parallels human infection in burn patients. Seeding of the wound with virulent strains of Pseudomonas organisms in the immediate postbum period is followed by development of visceral lesions and clinical signs which mimic those which occur in burn patients and are associated with predictable, reproducible, strain-specific mortality [ 15, 161. Thus, it was obviously of considerable significance to evaluate the Pseudomonas ribosomal vaccine and antiserum in the burned rat model of Pseudomonas infection. MATERIALS

AND METHODS

Bacteria. P. aeruginosa strain 12-4-4 (Habs serotype 8) was originally isolated from a patient who expired from Pseudomonas septicemia following burn trauma [21]. P. aeruginosa strain VA 134 (serotype 13) was isolated from a similar patient. Rats. Sprague-Dawley albino rats 175-225 g were obtained from either Timco (Houston, Tex.) or Charles River Breeding Labs (Wilmington, Mass.). Vaccines and antisera. Preparation and chemical analysis of “high salt” washed, chromatographically purified ribosomal vaccines and lipopolysaccharide from P. aeruginosa [ 8, lo] and vaccination and bleeding of rabbits for immune serum [lo] have been described previously. 2Xet0, 34deoxyoctonate (KDO) was assayed by the thiobarbituric acid method modified by the addition of dimethylsulfoxide [6]. Rats were administered vaccines or sera by intraperitoneal injection at times indicated below. Burn and challenge. The standard rat bum model described by Walker and Mason [20] was used. Briefly, the model employs a 10-s scald (using boiling water) administered to pentobarbital anesthetized rats with 20% of their total body surface exposed. The exposure is limited by enclosing the rat in a device which

139

is impermeable to water and well insulated except for an aperture for the exposure. This results in a full-thickness burn (dorsal side) on the exposed area with no lesions on the nonexposed (insulated) area. Seeding of the wound site was performed by two different methods: (A) 1 ml of a fully grown broth culture was rubbed over the entire surface of the wound using a sterile, cotton-tipped swab. (B) Logarithmic phase cultures were grown, monitored by turbidity measurements, and viability determinations performed as described previously [8]. Cultures (strain 1244) were then diluted either 1:5 or 1:10 and 0.5 ml or 1.O ml (about 2-5 X 10’ cfu) injected, subcutaneously, directly under the bum site. Most deaths occurred between 6 and 13 days postchallenge. Survivors were scored 21 days postchallenge. When the challenge is performed 3 days postburning, the lethality titer is about 100-200 LDsO/ml. When the challenge is performed on the same day as burning, the lethality titer is increased by more than an order of magnitude. Statistics. P values were calculated by the one-tailed Fisher exact probability test [ 191. RESULTS

Physical characterizationof ribosomal vaccines by sedimentation velocityanalysis. The ribosomal vaccines, prepared from sonically disrupted cells, were purified by ammonium sulfate fractionation, ultracentrifugation (including a “high salt” wash), and chromatography on Sepharose 4B and “peak B” [8, lo] used in all experiments. The vaccines (“peak B”) consist of approximately 65% RNA and 35% protein [8] with no detectable LPS determined either radiochemically [8] or by analysis for KDO (unpublished data). The ratio of absorbance at 260 nm to absorbance at 280 nm is 1.9. This material was characterized by sucrose density gradient ultracentrifugation and a typical sedimentation profile is shown in Fig. 1. The ribosomes sediment as 70 S material in the presence of 10 mM MgCl*, but dissociate into 50 and 30 S subunits after dialysis into 1 mM MgClz. Active immunity with ribosomal vaccines.

140

JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 2, FEBRUARY 1986 0.8

0.6

FRACTION No. FIG. 1. Sucrosedensity gradient ultracentrihrgation of dissociatedand undissociatedribosomes.Ribosomes were either maintained in 10 mMTris-HCI, pH 7.4, containing 10 rn84 MgC& or dissociated by dialysis in 10 mM Tris-HCl, pH 7.4, containing 1 mM MgCl,. Gradients with 5-2096 sucrose in either the buffer containing 1 mM MgClz (for dissociatedribosomes)or the bulTercontaining 10 mM MgClr (for undissociated ribosomes) were prepared. Samples were layered on top of the gradients and centrifuged at 135,OOOg (maximum) for 6 hr. Fractions were collected from the top of the gradients by a peristaltic pump connected to an electrically operated meniscus-sensitiveprobe. Dissociated ribosomes: 0, undissociated ribosomes: 0. Inset: S value plotted as a function of the percentage sucroseat which the 30,50, and 70 S particles from Escherichia coli ribosomes sediment (used for reference).

In the first experiment rats were vaccinated twice with varying dosesof ribosomal vaccine prepared from strain 1244 with a 7day interval between vaccinations. Ten days postbooster vaccination, rats were burned and challenged with homologous organisms as described above (method A). Control animals were not vaccinated, but were burned and infected. The results presented in Table 1 demonstrated that at doses from 20 to 500 pg, 100%protection was achieved. In the second experiment, rats were vaccinated only once and then 7 days later burned and challenged (method A). Ribosomes horn two different strains (representing different se-

rotypes) were used, and animals were challengedwith the homologous strain. The results are given in Table 2 and again show excellent protection with the ribosomal vaccine and an abbreviated vaccination schedule. In all subsequent experiments, immunization was performed postburning. The data presented in Table 3 are the results obtained from experiments where rats were burned and then vaccinated immediately afterward. Three days postvaccination, rats were challenged with the homologous strain as outlined above. Excellent protection with strain 1244,but only marginal protection with strain VA1 34 was achieved.

LIEBERMAN

ET AL.:

TABLE 3

TABLE 1 Am

WITHP. aeruginosa

I~~oN

141

Pseudomonas IMMUNITY

hBOsOMAL VACCINE AS A FIJNCllON OF VACCINE DOSE IN THE BURNED RAT MODEL

ACTIVE IMMUNIZATION OF BURNEDRATSwrr~ IbOsOh4AL VACCINE POSTBuRNING P. aeruginosa survivors/totaIb

Dose of vaccine” w

500 100 20 Controls

Survivors/ totalb

Dose’ % survivaI

100 100 100 0

919

818 IO/10 O/9

P valuec

0.00002 0.00004 0.00001 -

’ Dose of vaccine (strain 1244)in pg of protein administered intraperitoneally twice, 7 days apart; rats were burned and infected 10 days after the second vaccination. b Survivors were scored 21 days after burning and infecting (method A). ’ P value calculated by the Fisher exact probability test.

Active immunity comparing ribosomal vaccines with lipopolysaccharides. To compare ribosomes with LPS for active immunity and to determine the minimum time required to generateimmune protection, rats were burned, then vaccinated and challenged at various times. Excellent protection was achieved by vaccination with either ribosomesor LPS (data Table 4) when vaccination and burning were conducted the same day followed by method B infection 3 days later. When the time interval between vaccination and challenge was reduced to 1 or 2 days, only 50% protection was TABLE 2 ACTIVE IMMUNIZATION m P. aemginosa R~BOSOMAL VAAS A SINGLE DOSE IN THE BURNED RAT MODEL suNivors/totaIb Strain

1244 VA 134

Dow’ k?3

100 100

Vaccinated

Controls

P value’

9110 lO/lO

l/10 l/IO

0.0006 0.00006

n Dose of vaccine in a of protein administered once, intraperitoneally; rats were burned and infected 7 days after vaccination. ’ Survivors were scored 2 1 days after burning and infecting (method A). ’ P value calculated by the Fisher exact probability test.

Strain

(a)

Vaccinates

Controls

P valuesc

l/10 l/4 o/7

0.00006 -

1244’

100

IO/10

VA 134d

100

S/14

VA 134’

100

419

’ Dose of vaccine in pg of protein administered once, intraperitoneally. b Survivors were scored 2 1 days after infection. c P value calculated by the Fisher exact probability test. d Rats were vaccinated immediately after burning and infected 3 days later by method A. eRats were vaccinated 8 hrs after burning and infected 3 days later by method B.

achieved with ribosomes whereas 80% protection was obtained with LPS. No protection with either ribosomes or LPS was found when vaccination was performed 1 day after infection. Passive immunity with antiserum to ribosomal vaccines. Passiveimmunization of rats postbuming was studied by administration of preformed rabbit antibody. Rabbits were vaccinated with the ribosomal vaccine administered in two doseswith a 14-day interval. Immune serum was colkcted 14 days postbooster vaccination. Preimmune serum was obtained prior to the first vaccination. Preimmune or immune serum was administered to rats immediately after burning. Three days later rats were challenged (method A) and survivors scored for up to 21 days. In some groups of rats, antiserum (or preimmune serum) was administered more than once at 7day intervals. The results of this experiment, presented in Table 5, demonstrate that significant passive protection was achieved with either one or three dosesof antiserum to ribosomes. Passive immunity comparing antiserum to ribosomes with antiserum to LPS. Passiveimmunization experiments were performed to compare the immunogenicity of ribosomes with that of LPS. Rabbit antisera were pre-

142

JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 2, FEBRUARY 1986 TABLE 4 ACTIVEIMMUNITY~BURNEDRATSW~THRIBOSOMALANDLIPOPOLYSA

CCHARlDE(L~)vACCINES"

Vaccine b

Day of vaccination

Day of challenge

survivors/ total’

Ribosomes Lps None

0 0 -

3 3 3

718

Ribosomes None

0’ -

2 2

5110 l/9

Ribosomes LPS None

1 1 -

2 2 2

317 8110 Of4

Ribosomes LPS None

3 3 -

2 2 2

l/7 014

515 l/7

l/6

P valued KO.01
-

a Rats were burned on Day 0 and vaccinated immediately afterward or as indicated in the table. bRats were vaccinated with ribosomes (100 pg protein) or LPS (100 @ calculated from 2-keto, 3deoxyoctonate (KDO) analysis assuming KDO to be 4% of LES [4]) prepared from strain 1244. ’ Survivors were scored 2 1 days after infection (method B) with strain 1244. dP value calculated by the Fisher exact probability test comparing each set of vaccinated and control (nonvaccinated) rats. ’ Rats were vaccinated 8 hrs after burning.

pared as described above against either ribosomesor LPS. In standard passivemouse protection tests [ lo], these antisera were found to have approximately equal mouse protective capability (data not shown). Either antisera or preimmune serum were administered to rats on Days 1,4, and 8, postburning. Infection by method B was performed 2 days postburning. The data obtained from this experiment (Table 6) show 80% protection using antiserum to ribosomes and 50% protection using antiserum to LPS. DISCUSSION

The results presented here clearly demonstrate that ribosomal vaccines prepared from P. aeruginosu can protect rats from the lethal effects of infection with this organism after having been subjectedto a 20% total body surface burn. When administered prior to burning, the vaccine has provided unassailable protection (Tables 1 and 2). When administered after burning, the protection afforded is

dependent on the time interval between vaccination and infection. A minimum of 2 to 3 days is required between vaccination with ribosomesand infection in order to achieve effective protection. LPS, however, appears to be effective with only 1 day between vaccination and challenge (Table 4). This might be due to the fact that antibodies to LPS are primarily of the IgM class [2] and arise earlier in the host’s immune response than IgG antibodies [ 171. Antibodies to Pseudomonas ribosomal vaccineshave previously been shown to be both of the IgG and IgM classes,and both were shown to be effective in passive mouse protection studies [lo]. The results of the passiveprotection experiments in burned rats demonstrate that antiserum to the ribosomal vaccine can also provide a signiiicant measureof protection against lethal infection when administered postbuming (Tables 5 and 6). These experiments indicate that multiple passive immunizations increase the survival of burned and infected rats.

LIEBERMAN

ET AL.: Pseudomonas IMMUNITY

143

Walker et al. [22] tested exotoxoid A and found no protection whatsoever afforded the PA.W~EIMMUNIZATIONOFBURNEDRATSWITH animals. “Nontoxic” LPS vaccines were preANTISERUMTO P. aeruginosu RIBOSOMALVACCINE pared from P. aeruginosaand found to provide sllNivors/ Number of significant protection in burned rats (J. Sadoff, P valued totalc dOseSb Serum” R. Seid, P. Altieri, H. Collins, H. Sidberry, S. Futrovsky, and S. Berman, Abst. Annu. Meet. -co.05 Immune 1 4110 Amer. Sot. Microbial., 1982, p. 19). Also, the 1 Preimmune o/10 -co.05 11/15 Immune 3 same group found various degreesof partial Preimmune 3 5115 protection in burned and infected rats using live, killed, or cell-surface vaccines (H. H. ’ Serum obtained from rabbits either prior to vaccination (preimmune serum) or after two vaccinations given 14 Collins, H. F. Sidberry, and J. C. Sadoff,Abst. days apart followed by bleeding 14 days after the second Annu. Meet. Amer. Sot. Microbial., 1980, vaccination (immune serum); vaccination of rabbits by p. 84). subcutaneous administration of 200 pg of protein using Another model of thermal injury that has ribosomes from strain 1244. been used to evaluate Pseudomonas vaccines b0.5 ml of undiluted serum administered intraperitoneally immediately alherburning; rats were inkcted 3 days is the burned mouse model of Stieritz and after burning and passive immunization; those groups of Holder [ 181.Holder et al. [5] found that flarats given three dosesof serum were given the secondand gellar preparations were effective in protecting third doses of serum 7 and 14 days after the tirst dose, burned, infected mice. Pavlov&is et al. [ 121 respectively. used exotoxoid A in the samemodel with varcSurvivors were scored2 1daysaher infection with strain ious degreesof protection achieved. Finally, a 1244 (method A). d P value calculated by the Fisher exact probability test pilus vaccine has also been tested in both the for immune versus preimmune serum for each set (one burned mouse and burned rat models with or three dosesof serum). significant protection against Pseudomonas obtained (J. D. Silipigni, A. M. Levine, J. SaRecently, Powanda et al. [ 141 discovered doff, and C. C. Brinton, Abst. Annu. Meet. certain biochemical indicators of infection in Amer. Sot. Microbial., 1981, p. 17). perchloric acid filtrates of whole blood taken from burned and infected rats. These indicaTABLE 6 tors, determined by increasedfluorescenceand optical density measurements, appear to respond primarily to infection rather than to the extent of thermal injury [ 141.Theseindicators survivors/ Serumb totalc P valued were also determined in burned and infected rats that had been immunized with the PseuAntiserum to ribosomes 8110 co.001 domonas ribosomal vaccine and were found Antiserum to LPS s/10 KO.02 to be significantly reduced relative to nonvac- Preimmune serum o/10 cinated, burned, and infected rats (M. C. Po’ Rats were burned on Day 0, and given serum (0.5 ml, wanda, M. M. Lieberman, S. A. Sham, J. Duundiluted) on Days 1,4, and 8; rata were infected (method bois, Y. Villarreal, and B. A. Pruitt, Jr., B) on Day 2. manuscript in preparation). The response of bSerum obtained from rabbits either prior to vaccination these indicators to vaccination correlated well (preimmune serum) or after two vaccinations (given 14 with survival rates in these rats. Thus, these days apart) with either ribosomes (500 a protein) or LPS indicators might be useful in assessingthe ef- (500 rkgbased on KDO analysis assuming KDO to be 4% of LPS [4]) prepared from strain 1244, rabbits were.bled fectiveness of vaccination. 14 days after the second vaccination for immune serum. The burned rat model used in this report ‘Surviv0rsweresc0red2 ldaysafterinfection. has been used to test the efficacy of Pseudod P values calculated by the Fisher exact probability test monas vaccinesother than ribosomal vaccines. for immune versus preimmune serum. TABLE 5

144

JOURNAL OF SURGICAL RESEARCH VOL. 40, NO. 2, FEBRUARY 1986

In summary, the evidence indicates that the ribosomal vaccine provides at least as good or better protection in the burned rat than any of the other Pseudomonas vaccines tested to date in either animal model of thermal injury and infection as reported in the literature. ACKNOWLEDGMENTS The authors arc most grateful to Ms. Shirley Armann, Ms. Dodie Bratten and Ms. Helen Smith for expert manuscript preparation, and to Mr. Bill Gregory and Ms. Wendy Blomgren for dedicated and skillful veterinary technical assistance.All animal studiesadhered to the National Research Council’s guide for the care and use of laboratory animals.

REFERENCES 1. Alexander, J. W., and Moncrief, J. A. Alterations of the immune responsefollowing severethermal injury. Arch. Surg. 93: 75, 1966. 2. Barrett, J. T. Basic Immunology and Its Medical Application. St. Louis: Mosby, 1980. P. 53. 3. Bryan, L. E. Resistanceto antimicrobial agents: The general nature of the problem and the basis of resistance. In R. G. Doggett (Ed.), Pseudomonasaeruginosa, Clinical Manifestations of Infection and Current Therapy. New York: Academic Press, 1979. P. 2 19. 4. Chester, J. R., Meadow, P. M., and Pitt, T. L. The relationship between the 0-antigenic lipopolysaccharide and serological specificity in strains of Pseudomonas aeruginosa of different 0-serotypes. J. Gen. Microbial. 18: 305, 1973. 5. Holder, I. A., Wheeler,R., and Montie, T. C. FIagellar preparations from Pseudomonasaeruginosa: Animal protection studies. Infect. Immun. 35: 276, 1982. 6. Karkhanis, Y. D., Zeltner, J. Y., Jackson, J. J., and Carlo, D. J. A new and improved microassay to determine 2-keto, 3deoxyoctonate.in lipopolysaccharide of gram-negative bacteria. Anal. Biochem. 85: 595, 1978. 7. Liebcrman, M. M. Direct evidence for the presence of lipopolysaccharide components in a Pseudomonas ribosomal vaccine. Infect. Zmmun. 17: 47 I, 1977. 8. Lieberman, M. M. Pseudomonasribosomal vaccines: Preparation, properties, and immunogenicity. Infect. Immun. 21: 76, 1978. 9. Lieberman, M. M., and Ayala, E. Active and passive immunity against Pseudomonasaeruginosa with a ribosomal vaccine and antiserum in C3H/HeJ mice. J. Zmmunol. 131: 1, 1983.

10. Liebetman, M. M., McKissock, D. C., and Wright, G. L. Passive immunization against Pseudomonas with a ribosomal vaccine-inducedimmune serum and immunoglobulin fractions. Infect. Immun. 23: 509, 1979. 11. Lieberman, M. M., Wright, G. L., Wolcott, K. M., and McKissock-De&o, D. C. Polyvalent antisera to Pseudomonasribosomal vaccines:Protection of mice against clinically isolated strains. Infect. Immun. 29: 489,198O.

12. Pavlov&is, 0. R., Edman, D. C., Leppla, S. H., Wretlind, B., Lewis, L. R., and Martin, K. E. Protection against experimental Pseudomonasaeruginosa infection in mice by active immunization with exotoxin A toxoids. Infect. Immun. 32: 68 1, 1981. 13. Pennington, J. E. Immunotherapy of Pseudomonas aeruginosa infection. In R. G. Doggett (Ed.), Pseudomonas aeruginosa, Clinical Mamfestations of Infection and Current Therapy. New York: Academic Press, 1979. P. 191. 14. Powanda, M. C., Dubois, J., Villarreal, Y., Walker, H. L., and Pruitt, B. A., Jr. Detection of potential biochemical indicators of infection in the burned rat. J. Lab. Clin. Med. 97: 672, 1981. 15. Pruitt, B. A., Jr. Infections of bums and other wounds causedby Pseudomonasaeruginosa. In L. D. Sabath (Ed.), Pseudomonasaeruginosa, the Organism, DiseasesIt Causes,and Their Treatment. Bern: Huber, 1980. P. 55. 16. Pruitt, B. A., Jr., and Lindberg, R. B. Pseudomonas aeruginosa infections in burn patients. In R. G. Doggett (Ed.), Pseudomonasaeruginosa, Clinical Manifestations of Infection and Current Therapy. New York: Academic Press, 1979. P. 339. 17. Roitt, J. M. Essential Immunology. London: Blackwell, 1977. P. 82. 18. Stieritx, D. D., and Holder, I. A. Experimental studies of the pathogenesisof infection due to Pseudomonas aeruginosa: Description of a burned mouse model. J. Infect. Dis. 131: 688, 1975. 19. Swinscow,T. D. V. Statistics at Square One. London: Brit. Med. Assoc., 1976. P. 54. 20. Walker, H. L., and Mason, A. D., Jr. A standard animal bum. J. Trauma 8: 1049, 1968. 21. Walker, H. L., Mason, A. D., Jr., and Rat&on, G. L. Surfaceinfection with Pseudomonasaeruginosa. Ann. Surg. 160: 297, 1964. 22. Walker, H. L., McLeod, C. G., Jr., Leppla, S. H., and Mason, A. D., Jr. Evaluation of Pseudomonasaeruginosa toxin A in experimental rat bum wound sepsis. Infect. Immun. 25: 828, 1979. 23. Zak, 0. Antibiotics and Pseudomonasaeruginosa. In L. D. Sabath (Ed.), Pseudomonasaeruginosa, the Organism, Diseases It Causes, and Their Treatment. Bern: Huber, 1980. P. 133.