INFECTION AND IMMUNITY, Jan. 1991, p. 1-6

Vol. 59, No. 1

0019-9567/91/010001-06$02.00/0 Copyright © 1991, American Society for Microbiology

A Protective Human Monoclonal Antibody Directed to the Outer Core Region of Pseudomonas aeruginosa Lipopolysaccharide MASAZUMI TERASHIMA,1* IKUKO UEZUMI,1 TEIJI TOMIO,1 MASUHIRO KATO,' KENJI IRIE,1 TAKAO OKUDA,' SHIN-ICHI YOKOTA,2 AND HIROSHI NOGUCHI2 Research Laboratories, Sumitomo Pharmaceuticals Co. Ltd., Osaka 554,1 and Laboratory of Biotechnology, Takarazuka Research Center, Sumitomo Chemical Co. Ltd., Takarazuka, Hyogo 665,2 Japan Received 21 May 1990/Accepted 5 October 1990 The protective activity against experimental Pseudomonas aeruginosa infection of a human monoclonal antibody, MH-4H7, which is thought to recognize L-rhamnose and its neighboring residues in the outer core region of P. aeruginosa lipopolysaccharide and which binds to strains of Homma serotypes A, F, G, H, K, and M, was studied in normal, burned, and leukopenic mice. MH-4H7 at doses of 0.1 to 1.0 ,ug per mouse (5 to 50 ,ug/kg) was effective against serotype A, F, G, H, and K clinical isolates of P. aeruginosa tested in normal mice but not against strains of serotype M, B, E, or I. The 50% protective doses were calculated to be 0.01 to 0.1 Fg per mouse against challenge with serotype G strains and 3 to 8 ,ug per mouse against challenge with serotype A strains. MH-4H7 promoted macrophage-mediated opsonophagocytosis of serotype A, F, G, H, and K strains but not of serotype M strains. The opsonophagocytic activity, expressed as the reduction rate of viable bacteria in the presence of MH-4H7, macrophages, and complement, was higher against serotype G strains (more than 90%) than against serotype A strains (60 to 80%) and serotype F, H, and K strains (50 to 86%). It was correlated with the protective activity but not with the binding intensity of MH-4H7 to the organisms. In addition, burned and leukopenic mice as well as normal mice infected with serotype G strains recovered from a very low dosage of MH-4H7. Thus, a monoclonal antibody directed to the outer core region of P. aeruginosa lipopolysaccharide was effective against infection with a wide range of 0-serotype strains of P. aeruginosa.

clinical isolates belonging to various serotypes of P. aerug-

Pseudomonas aeruginosa is a major pathogen for opportunistic infections in compromised hosts such as patients undergoing immunosuppressive therapy and those with burns, diffuse panbronchiolitis, or cystic fibrosis. Chemotherapy in such patients is often ineffective because the resistance of P. aeruginosa strains to many kinds of antibiotics has been increasing (3, 12, 16). In contrast, immunotherapy has been suggested (4, 15, 17, 19) to reduce infectivity of P. aeruginosa. Lipopolysaccharide (LPS), mucoid exopolysaccharide, flagella, and outer membrane protein F of P. aeruginosa have been extensively studied for their effectiveness as protective antigens (1, 6, 9, 14). Recently, monoclonal antibodies (MAbs) specific for LPS have been obtained and shown to be highly protective (2, 18, 21, 22, 25, 26, 29). LPS consists of three regions, 0 polysaccharide, core oligosaccharide, and lipid A, which are covalently bound to each other. Antibodies to 0 polysaccharide have been demonstrated to be more effective than antibodies to the other components of LPS in experimental animal infection models (13, 18, 21, 22, 25). However, the former antibodies have narrow efficacy spectra, because each of them is only effective against strains belonging to the specific 0 serotype that is recognized by the antibody. We have recently isolated human MAbs MH-4H7 and KN-2B11, which are thought to recognize the outer core region of P. aeruginosa LPS, especially the nonreducing terminal of the L-rhamnose residue in the outer core region, and we have demonstrated that the MAbs bind to strains of several serotypes (27). In this study, we examined the effects of MH-4H7 against infections in mice caused by some

inosa.

MATERIALS AND METHODS MH-4H7. MH-4H7 (immunoglobulin M [IgM]) (27) was produced by culturing the heterohybridoma cell line MH4H7 in a serum-free medium, Celgrosser H (Sumitomo Parmaceuticals, Osaka, Japan), that contained only insulin (5 mg/liter) and transferrin (5 mg/liter) as protein components. The preparation chromatographically purified to more than 95% purity was used for this study. Fluorescein isothiocyanate labeling of MH-4H7. The MAb (5 mglml) was incubated with 20 ,ug of fluorescein isothiocyanate isomer I (Dojin Chemical Laboratory, Kumamoto, Japan) per ml in 0.05 M sodium-hydrogen-carbonate buffer (pH 9.5) containing 0.85% sodium chloride at 4°C for 4 h. Then the labeled MAb was separated by gel chromatography on a PD-10 Sephadex G-25M column (Pharmacia, Uppsala, Sweden) in phosphate-buffered saline (PBS). Bacterial strains. P. aeruginosa IIDlOOl (serotype A) was obtained from the Institute of Medical Science, University of Tokyo, Tokyo, Japan. Clinical isolates used for this study were from our laboratory collection. The serotypes of strains were determined by a serotype grouping kit, Mei-assay (Meiji Seika Co., Tokyo, Japan). All strains were grown on heart infusion agar (Nissui Pharmaceuticals, Tokyo, Japan) at 370C. Binding assay. The binding specificity of MH-4H7 to these clinical isolates was tested by an enzyme-linked immunosorbent assay previously described in detail (27). Briefly, P. aeruginosa cell suspensions in PBS were dispensed into 96-well plates and fixed with glutaraldehyde. After the plates were washed, 1-,ug/ml MAb solutions were dispensed and incubated at 37°C for 2 h. After several washings, the bound

* Corresponding author. 1

2

TERASHIMA ET AL.

human MAb MH-4H7 was colorimetrically determined by incubating the plates with alkaline phosphatase-conjugated goat anti-human IgM antibody, washing the plates, and then adding the substrate p-nitrophenylphosphate. Agglutination assay. The agglutination assay was carried out with viable P. aeruginosa cells in U-bottomed microtiter plates (Costar, Cambridge, Mass.). Bacterial cell suspension (25 ,ul) in PBS (pH 7.2) was prepared to Klett constant 200 and mixed with an equal volume of a twofold serial dilution of the MAb. The mixture was incubated for 18 h at room temperature. Agglutination was then monitored. The agglutination titer was determined as the minimum concentration of MAb required for positive agglutination. Phagocytosis. Macrophages (M+) were prepared by washing out peritoneal cavities of normal 6-week-old ICR male mice (Japan SLC Inc., Shizuoka, Japan) with 3 ml of RPMI 1640 containing 0.1% (wt/vol) bovine serum albumin (BSA). Cell suspensions were washed and suspended in RPMI 1640 at 2 x 106 cells per ml. Approximately 90% of cells were confirmed to be M4) by Giemsa staining, and the viability was determined to be over 95% by trypan blue exclusion. Polymorphonuclear leukocyte-rich suspensions were obtained as described above, except that mice were injected intraperitoneally (i.p.) with 0.5% (wt/vol) oyster glycogen (Nacalai Tesque, Inc., Kyoto, Japan) and peritoneal fluid was recovered 3 h after the injection. The purity of the polymorphonuclear leukocytes was about 90%. Syngenic mouse serum, stored at -80°C, was absorbed for 1 h at 4°C with each of the P. aeruginosa strain tested. The absorbed mouse serum was used as a source of complement. Heat-inactivated serum was prepared by heating the serum for 30 min at 56°C. The bacterial suspension (0.1 ml; approximately 106 CFU/ ml), MH-4H7 (0.2 ml), mouse serum (0.2 ml), and the cell suspension (0.5 ml) were mixed in a petri dish (Falcon 3001). Approximately 105 CFU of bacteria, 106 M( or polymorphonuclear leukocytes, and 10% (wt/vol) mouse serum were contained in final reaction mixtures. The mixtures were incubated for 2 h under continuous rolling (100 rpm) at 37°C. A small portion was removed, serially diluted with sterile water or saline, and inoculated on heart infusion agar medium. The number of viable bacteria was calculated by colony counting 16 h after inoculation. Opsonophagocytic activity was expressed as bacterial killing (percent), defined as (1 - N/NO) x 100, where Ni and No represent the number of viable bacteria after the incubation in the presence of the MAb at a concentration of i (micrograms per milliliter) and in the absence of the MAb, respectively. Because serum absorbed with homologous bacteria at 4°C had no opsonophagocytic activity in our assay, the initial number of inoculated bacteria was almost the same as No. These experiments were run in quadruplicate, and the means were determined. Student's t test was used to determine the significance of log1o reductions of bacteria in the opsonoph-

agocytic

assay.

Protection experiments in mice. In the following three experimental infection models with P. aeruginosa, male ICR mice (Japan SLC Inc.) 4 weeks old (body weight, 22 to 24 g) were used. Each of the 10 mice per group was infected with four different inoculum sizes of fivefold serial dilutions. Serial 10-fold-diluted MH-4H7 (0.01 to 10 ,ug per mouse) was i.p. or intravenously (i.v.) administered. The animals were observed for 6 days. The lethal dosage causing death in 50% of mice (LD50) was calculated by probit analysis. The significance of differences from control group was indicated by nonoverlapping 95% confidence intervals.

INFECT. IMMUN. TABLE 1. Binding, agglutinating, and opsonophagocytic activities of MH-4H7 P. aeruginosa

Binding

Agglutinating

Opsonophagocytic activity (% killed)' 94 90 93 76 60 71 86 77 50 0 0 0 0 0 0

Serotype

Strain

activity"

activityb

G G G

SP9792 SP6788 SP9785 SP6783 SP6818 IID1001 SP7514 SP9751 SP6864 SP6764 SP6765 SP10067 SP6897 SP10043 SP10046

1.9 >2.0 1.3 1.1 >2.0 1.9 >2.0 2.0 1.5 >2.0 >2.0 >2.0 0 0 0

0.94 0.94 0.47 0.47 0.47 0.47 0.94 0.94 3.8 15 15 0.12 >120 >120 >120

A A A H K F M M M B E I

"The binding activity of MH-4H7 was tested at a concentration of 1 ,ug/ml. This activity was expressed as A405 for an enzyme-linked immunosorbent assay in which plates coated with glutaraldehyde-fixed P. aeruginosa cells were used. b The minimum concentration (micrograms per milliliter) of MH-4H7 required for agglutinating viable P. aeruginosa cells (approximately 5 x 108 CFU/ml). ' The ability of MH-4H7 to promote the uptake and killing of P. aeruginosa by mouse resident peritoneal macrophages in the presence of complement was measured. The opsonophagocytic activity was expressed as the highest percentage of bacterial cells killed.

There was no significant difference in LD50 between BSAtreated groups and untreated groups. (i) Normal mouse model. A cell suspension of each bacterial strain was serially diluted in saline and mixed with bacteriological mucin (Difco Laboratories, Detroit, Mich.). The bacterial suspension (0.2 ml) containing 5% mucin was injected into mouse peritoneum. One hour later, 0.2 ml of MAb in PBS was administered to the mice either i.p. or i.v. Negative control groups of mice were either untreated or injected with BSA instead of MAb. (ii) Leukopenic mouse model. Leukopenic mice were prepared by i.p. administration of 250 mg of cyclophosphamide (Shionogi Pharmaceuticals Co. Ltd., Osaka, Japan) per kg of body weight. Four days after the treatment, when the total leukocyte number was less than 500 per mm3, mice were i.p. challenged with the bacterial suspension in the presence of 5% mucin; 1 h later, MAb was administered i.p. or i.v. (iii) Burned mouse model. Burned mice were prepared by the procedure of Stieritz and Holder (24) with a slight modification. After anesthesia, an 8-cm2 burn was created on the shaved backs of mice with an ethanol flame. Immediately after 10 s of burning, saline (0.3 ml) was injected i.p. for fluid replacement, and the bacterial suspension in saline (0.2 ml) was inoculated subcutaneously at the burn site. One hour later, MAb in PBS (0.2 ml) was i.v. administered. RESULTS In vitro activities. Human MAb MH-4H7 specifically bound to standard strains and clinical isolates of P. aeruginosa Homma serotype A, F, G, H, K, and M strains in the enzyme-linked immunosorbent assay, but it did not bind to serotype B, E, or I strains at all (Table 1) (27). The binding specificity (spectrum) with glutaraldehydefixed cells in the enzyme-linked immunosorbent assay was

VOL. 59,

1991

PROTECTIVE HUMAN MAb TO P. AERUGINOSA LPS CORE of

c

3.8 3

MAb

viable

4.1

bacteria

4.3

4.7

log( CFU/ml)

5.0

5.3

(mig/ml)

A

0 + +

+

0.0 01 0.0 1 0.1 1

10 0

TABLE 2. Protective activities of MH-4H7 in experimental infection models of normal micea

P. aeruginosa

--

0.001 0.0

I_

~~~~~B

1

0.1

-I

a

1

FIG. 1. In vitro opsonophagocytic activity of MH-4H7 against P. aeruginosa serotype A strain SP6783 (A) and serotype G strain SP9792 (B) by mouse peritoneal macrophages. Bacteria (A, 1.3 x 105 CFU/ml; B, 3.5 x 105 CFU/ml) were incubated at various concentrations of MH-4H7 with mouse absorbed normal serum (complement) or heat-inactivated serum (no complement) for 2 h at 37°C. Aliquots of the mixture were serially diluted with sterile saline and inoculated on heart infusion agar medium. Opsonophagocytic activity was determined by calculating the number of viable bacteria. Values represent the means + standard errors of two independent experiments determined from the number of colonies on the four culture plates. A similar number of viable bacteria was also obtained after aliquots of the mixture were diluted with sterile water to lyse macrophages and recover intracellular viable bacteria. Significant differences (Student t test) were observed from the number of viable bacteria in the absence of MH-4H7 (**, P < 0.01; ***, P < 0.001).

qualitatively confirmed by staining viable cells of several different serotypes with fluorescein isothiocyanate-labeled MH-4H7. Strains of Homma serotypes A, F, G, H, K, and M were positively stained by fluorescein isothiocyanate-labeled MH-4H7, whereas serotype B, E, and I strains were not stained (data not shown). MH-4H7 agglutinated viable bacteria of serotypes A, F, G, H, and K at a concentration of more than 0.47 to 3.8 jig/ml (Table 1). In contrast, MH-4H7 did not agglutinate strains of serotype B, E, or I even at a high concentration of 120 jig/ml. For serotype M, more than 15 jig of MH-4H7 per ml was required for agglutination, except that strain SP10067 (one of the three M strains) was agglutinated in the presence of 0.12 jig of the MAb per ml. Mouse peritoneal M4X efficiently killed bacteria of P. aeruginosa serotypes A and G in the presence of mouse complement (mouse normal serum preabsorbed with bacteria tested) and MH-4H7 (Fig. 1). The opsonophagocytic killing activity was dependent upon the concentration of MH-4H7. The absence of either MX or MAb resulted in no killing effect. Serotype G strains were more sensitive to killing by the MAb than were serotype A, H, K, and F strains (Fig. 1 and Table 1). In contrast, serotype M strains and serotype B, E, and I strains were not killed by the MAb even in the complete reaction mixture. MH-4H7 also promoted the opsonophagocytic activity of mouse peritoneal polymorphonuclear leukocytes killing bacteria of the specific serotypes in the presence of complement (data not shown).

LD50b in mice

Dosage of MH4H7 administered

Serotype

Strain

(,ug/mouse)

G G G

SP9792 SP6788 SP9785

G G A A A H K F M M M B E I

SP9755 SP9701 SP6783 SP6818 1ID1001 SP7514 SP9751 SP6864 SP6764 SP6765 SP10067 SP6897 SP10043 SP10046

0.1 0.1 0.1 1.0 0.1 0.1 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

10 ~~~~~~~~~~~~I-

3

treated with:

MH4H7

1.6 3.2 4.4 1.4 2.7 7.3 3.2 1.3 2.5 6.6 7.3 1.8 2.8 7.2 3.9 4.6 4.2 3.2

x

Protective

BSA

107 5.0 x 105 x 104

x 105 3.0 x 106 3.3 x 107 1.1 x 106 1.5

x 106 x 106

x 104

x 106 1.6 x x 106 4.0 x x 106 1.7 x x 106 4.3 x x 106 1.5 x x 106 2.3 x x 106 7.1 x x 106 2.7 x x 105 7.0 x x 105 3.8 x x 104 5.8 x x 104 4.6 x x 105 2.8 x

106

105 105 105 106 106

105 106

105 105 104 104 105

activityc

32d lld 1.3 13d 180d 4.6d 8.0d

7.6d 5.8d 4.4d

3.2d 2.5d 1.0 1.0 1.0 0.8 0.9 1.1

"Mice received MH-4H7 i.p. in 0.2 ml of PBS 1 h after i.p. challenge of bacteria with 5% mucin. Control mice received BSA instead of MH-4H7. b Expressed as the number of colony-forming units of bacteria that killed 50% of mice. ' Expressed as a relative protection rate: (LD50 for MH-4H7-treated group)/(LD50 for BSA-treated group). " Significantly different (P < 0.05) from BSA-treated group.

In vivo activities. The protective activities of MH-4H7 were evaluated in three kinds of experimental mouse models. In normal mice i.p. challenged with P. aeruginosa clinical isolates, MH-4H7 i.p. administered 1 h after the bacterial challenge at a dosage of 0.1 jig per mouse significantly (P < 0.05) protected mice against all strains of serotype G tested except for strain SP9785 (Table 2). For example, with MH-4H7 (0.1 jig per mouse) the LD50 of strain SP9792 was 32-fold higher (1.6 x 107 CFU) than that of the control. MH-4H7 (1.0 jig per mouse) also significantly (P < 0.05) protected mice against infections by strains of serotypes A, H, K, and F. In contrast, MH-4H7 (1.0 jig per mouse) had no efficacy against serotype M strains or against serotype B, E, and I strains. The protective activity was dependent upon the concentration of MH-4H7 (Fig. 2). More than 0.01 jig of the MAb per mouse significantly (P < 0.05) protected against challenge with SP9792 (7 to 12 LD50s) (Fig. 2B). The 50% protective dose (PD50) was calculated to be 0.013 jig per mouse (0.52 jig/kg). The PD50s of serotype A strains ranged from 3 to 8 jig per mouse (150 to 400 jig/kg) against a challenge of 7 to 12 LD50s, whereas the PD50s were 0.04 to 0.1 jig per mouse (2 to 5 jig/kg) against a challenge of 2 to 4 LD50s. MH-4H7, however, had no efficacy against serotype M even at the dosage of 10 jig per mouse. Furthermore, compared with conventional immunoglobulin IVGG, which possessed 103-fold lower binding activity to SP9792 cells in the enzyme-linked immunosorbent assay, MH-4H7 demonstrated 104- to 105-fold more potent efficacy (Fig. 3). The i.v. administered MAb showed the same magnitude of efficacy as did the i.p. administered MAb against strains of sensitive serotypes (data not shown). In leukopenic mice, 50% of which were killed by challenge

4

INFECT. IMMUN.

TERASHIMA ET AL.

A

8

-

0

*| C

B

n 0

E 7

-

T

,,

o 6

*

0

_~~~~~~~~~~~~~~

i _ 0

5

-J

I"I

-O

I

II

-2 -1 0 1

I

.i ..L.,A

iI a

I

-o-3 -2 -1 0

I

I

1

2

I

0

I

I

-OD-2 -1 0

1

Dose of MAb

log (jig mouse) FIG. 2. Protective activity of MH-4H7 in normal mice infected i.p. with P. aeruginosa serotype A strain SP6783 (A), serotype G strain SP9792 (B), and serotype M strain SP6764 (C). Mice received MH-4H7 (0) at the dosages indicated or were treated with BSA (A) at a dosage of 10 ,ug per mouse (controls). Asterisks (*) indicate results that are significantly different (P < 0.05) from those of controls.

with only 5 CFU of bacteria, MH-4H7 i.p. administered at a dosage of 0.1 ,ug per mouse protected mice significantly against challenge with serotype G strain SP9792. MH-4H7treated mice survived when challenged with 360 times the number of bacteria that was lethal for the BSA-treated group (Table 3). More than 90% of burned mice treated i.v. with 0.1 ,Ig of MH-4H7 survived challenge with the highest inoculum tested (5.0 x 104 CFU per mouse), whereas the LD50 for the control group was calculated to be 77 CFU of serotype G strain SP6788 per mouse (Table 3). To address the question of how MH-4H7 displays its protective activity, we quantitated the P. aeruginosa in the peritoneum and blood at various times after the infection with SP9792 organisms in the normal mouse model (Fig. 4). In the peritoneums of mice that did not receive MH-4H7, the number of viable bacteria was constant until 4 to 6 h after infection and then bacteria multiplied exponentially. In the case of mice that received MH-4H7 (0.1 ,ug per mouse),

however, the number of viable bacteria rapidly decreased within 1 h after the administration of MAb and was under the limit of detection 7 h after the administration of MAb (Fig. 4A). Approximately 104 CFU of bacteria per ml were detected in the blood 2 h after infection in both MAb-treated and untreated mice, but no bacteria could be detected 4 h after infection in MAb-treated mice, in contrast to rapid proliferation of bacteria in untreated mice (Fig. 4B).

DISCUSSION Antisera and MAbs reactive with the 0 polysaccharide of P. aeruginosa LPS have been reported to act as opsonins and effectively protect mice from death caused by experimental bacteremia (11, 21, 22, 25, 29). Almost all of the antibodies so far reported were directed to serotype-specific epitopes on LPS and were only active against a specific serotype out of 17 serotypes. Lang et al. (11) recently reported MAbs that recognized and had protective activity against two or more Fisher immunotypes of P. aeruginosa. These MAbs are likely to recognize the epitopes in the

10 8 -

6

;ol kot 10

TABLE 3. Protective activity of MH-4H7 against experimental infections with P. aeruginosa serotype G strains in burned and leukopenic mouse modelsa 0

0

co 4

Infection model

,

LD50 in mice

;o 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ C,,

2

MH4H7b

C)

Leukopenic Burned

W

-2

0

2

4

log(,ug/mouse) FIG. 3. Protective activity of MH-4H7 and a normal gamma globulin preparation, IVGG, in normal mice infected with P. aeruginosa serotype G. Mice were infected with 8 LD50s of P. aeruginosa serotype G strain SP9792. MH-4H7 (0) and IVGG (M) were i.p. administered. All mice that received 10 ,ug of BSA per mouse instead of antibody died on day 6 after infection. Dose of Ab

SP9792 SP6788

Protective activity'

treated with:

Strain

1.8 >5.0

x x

103 104

BSA

5.0 77

360 >649

In a leukopenic mouse model, mice were administered with cyclophosphamide (250 mg/kg) and challenged with bacteria 4 days later; then 0.1 pg of MH4H7 was administered i.p. 1 h after bacterial challenge. In the burned mouse model, mice received 0.1 ,ug of MH4H7 i.v. 1 h after subcutaneous bacterial challenge. b Mice received MH4H7 i.p. for the leukopenic model and i.v. for the burned model, administered at a dosage of 0.1 pig per mouse, 1 h after bacterial challenge. ' Expressed as a relative protection rate: (LD50 for MH4H7-treated group)/(LD50 for BSA-treated group). Results were significantly different (P < 0.05) from those of the BSA-treated group.

PROTECTIVE HUMAN MAb TO P. AERUGINOSA LPS CORE

VOL. 59, 1991 8

7 m

4J 0

6

1-. LL 5

u 0)

0 -j

4

3 1 0

._v_ 2

4

6

8

I i 0 2

Time after

. x_

4

6

1 3 8

infection ( h)

FIG. 4. Number of bacteria in total peritoneal fluid (A) and in 1 ml of blood (B) of mice challenged with 4 x 106 CFU (5 LD50s) of P. aeruginosa SP9792 at time 0. Data are expressed as mean values + standard errors of the numbers of viable bacteria in six mice. Symbols: 0, mice that received MH-4H7 (0.1 p.g per mouse) i.p. 1 h after infection; 0, control mice treated with BSA instead of MH-4H7. Fewer than 103 CFU in the total peritoneal fluid was not detectable in this assay. Significant differences (Student t test) were observed from the number of viable bacteria in the absence of MH-4H7 at each time (***, P < 0.001).

O-polysaccharide region of LPS shared by several immunoof P. aeruginosa. In contrast, there have been no reports on protective MAbs that have been directed to the outer core region of P. aeruginosa LPS. Coughlin and Bogard only reported that murine MAbs specific for the outer-core polysaccharide types

were

immunoprotective against infections by Escherichia

coli (5). We previously described (27) the production and characterization of human MAb MH-4H7, directed to the outer core region of P. aeruginosa LPS. MH-4H7 bound in vitro to several serotypes of P. aeruginosa, such

as

serotypes A, F,

G, H, K, and M. The epitope on LPS recognized by the MAb supposed to be L-rhamnose and its neighboring residue in the outer core region. In this paper we have first demonstrated that MH-4H7 protected mice in vivo significantly against a challenge with clinical isolates of Homma serotypes A, F, G, H, and K but not with serotype M strains (Table 2). What kinds of factors determine the immunoprotective activity of MAbs against bacterial infections? To address the question, we examined the binding, agglutinating, and opwas

sonophagocytic activities of MH-4H7 against 15 to 17 strains nine serotypes compared with its in vivo protective activity. MH-4H7 was more effective in vivo against serotype G strains than against serotype A, F, H, and K strains, in accordance with the observation that opsonophagocytic activity, expressed as the reduction rate of viable bacteria in the presence of the MAb, complement, and macrophages, was higher against serotype G strains (more than 90%) than against serotype A strains (60 to 80%) and serotype F, H, and K strains (50 to 86%). The protective activity of the MAb was more related with the opsonophagocytic activity rather than its binding and agglutinating activities (Table 1 and 2), suggesting that the opsonophagocytic activity would greatly contribute to the protection of MH-4H7 against pseudomonal infections. over

5

Although MH-4H7 agglutinated all three strains of serotype M tested, it showed neither protective nor opsonophagocytic activity. Neither mucoid products nor elastase secreted by P. aeruginosa was involved in these low MAb activities (data not shown). Therefore we do not yet know the real reason why serotype M strains were resistant to the opsonophagocytic activity of MH-4H7. Opsonophagocytosis of MH-4H7 was complement dependent. Pier et al. (19, 20) have also reported that IgM requires complement for phagocytic killing. Recently, Schiller and Joiner (23) and Engels et al. (7, 8) reported that some strains of P. aeruginosa could not fix complement 3b to their surface although they activated complement 3. This phenomenon might answer the question of why serotype M strains were not well phagocytized. Wild strains of P. aeruginosa have a smooth type of LPS consisting of both high- and low-molecular-weight LPS. High-molecular-weight LPS seems to have a long 0-polysaccharide chain. Hence we were afraid that HMW LPS would interfere with the access of the MAb to the inner target, resulting in low protective activity. However, we have demonstrated that this is not the case, because MH-4H7 bound to viable cells of susceptible strains in vitro and was also protective against them in vivo. Interestingly, Kelly et al. (10) recently reported that cells grown in vivo appeared to lack a series of high-molecular-weight LPSs compared with cells grown in vitro and to have gained a new series of low-molecular-weight LPSs. This may mean that under some conditions, cells in vivo become enriched in the rough-natured cell population that is highly sensitive to the MAb. Human MAbs to the 0 polysaccharide of P. aeruginosa have been shown to have potent efficacy against experimental infection models in normal mice by several laboratories. For example, the PD50s were reported to range from 0.2 to 10 ,ug per mouse (10 to 500 ,ug/kg) for IgG (21) and 0.5 to 3 ,ug per mouse (25 to 150 ,ug/kg) for IgM under a relatively mild challenge of 5 LD50s (26). 0 polysaccharide-specific human MAbs of IgG and IgM isolated by us also showed PD50s ranging from 0.8 to 2.5 ,ug per mouse (data not shown). The PD50s in the neutropenic mouse model were also reported to range from 0.3 to 0.5 ,ug per mouse for IgM and IgG (20). Compared with 0 polysaccharide-specific human MAbs, MH-4H7 showed a more potent protective activity against infections with serotype G strains (their PD50s were 0.01 to 0.1 ,ug per mouse against a challenge of 8 to 22 LD50s) and a comparable activity against serotype A strains (their PD50s ranged from 3 to 8 ,ug per mouse against a challenge of 7 to 12 LD50s and 0.04 to 0.1 p,g per mouse against a challenge of 2 to 5 LD50s). MH-4H7 strongly promoted the clearance of challenge bacteria from the peritoneal cavity and blood of mice, as shown in Fig. 4. A preliminary pharmacokinetic study demonstrated that the level (in serum) of MH-4H7 administered to mice at the effective dose reached and maintained a sufficient concentration to promote phagocytosis for 30 h (data not shown), which may explain the potent protective activity of MH-4H7. In this study we showed the efficacy of MH-4H7 in experimental infection models with several serotypes of P. aeruginosa. These results suggest its potential usefulness for immunotherapy in clinics. REFERENCES 1. Ames, P., D. Desjardins, and G. B. Pier. 1985. Opsonophagocytic killing activity of rabbit antibody to Pseudomonas aeruginosa mucoid exopolysaccharide. Infect. Immun. 49:281-285. 2. Barclay, G. R., P. L. Yap, D. B. L. McClelland, R. J. Jones,

TERASHIMA ET AL.

6

3. 4.

5.

6.

7.

8.

9. 10. 11.

12.

13.

14.

15.

E. A. Roe, M. C. McCann, L. R. Micklem, and K. James. 1986. Characterization of mouse monoclonal antibodies produced by immunisation with a single serotype component of a polyvalent Pseudomonas aeruginosa vaccine. J. Med. Microbiol. 21:87-90. Bodey, G. P., R. Bolivar, V. Fainstein, and L. Jadeja. 1983. Infections caused by Pseudomonas aeruginosa. Rev. Infect. Dis. 5:279-313. Collins, M. S., and R. E. Roby. 1976. Protective activity of a intravenous immune globulin (human) enriched in antibody against lipopolysaccharide antigens of Pseudomonas aeruginosa. Am. J. Med. 76(Suppl. 3A):168-174. Coughlin, R. T., and W. C. Bogard, Jr. 1987. Immunoprotective murine monoclonal antibodies specific for the outer-core polysaccharide and for the 0-antigen of Escherichia coli 0111:B4 lipopolysaccharide (LPS). J. Immunol. 139:557-561. Cryz, S. J. Jr., E. Furer, and R. Germanier. 1983. Protection against Pseudomonas aeruginosa infection in a murine burn wound sepsis model by passive transfer of antitoxin A, antielastase, and antilipopolysaccharide. Infect. Immun. 39:10721079. Engels, W., J. Endert, M. A. F. Kamps, and C. P. A. van Boven. 1985. Role of lipopolysaccharide in opsonization and phagocytosis of Pseudomonas aeruginosa. Infect. Immun. 49: 182-189. Engels, W., J. Endert, and C. P. A. van Boven. 1985. A quantitative method for assessing the third complement factor (C3) attached to the surface of opsonized Pseudomonas aeruginosa: interrelationship between C3 fixation, phagocytosis, and complement consumption. J. Immunol. Methods 81:43-53. Holder, I. A., R. Wheeler, and T. C. Montie. 1982. Flagellar preparations from Pseudomonas aeruginosa: animal protection studies. Infect. Immun. 35:276-280. Kelly, N. M., A. Bell, and R. E. W. Hancock. 1989. Surface characteristics of Pseudomonas aeruginosa grown in a chamber implant model in mice and rats. Infect. Immun. 57:344-350. Lang, A. B., J. W. Larrick, and S. J. Cryz, Jr. 1989. Isolation and characterization of a human monoclonal antibody that recognizes epitopes shared by Pseudomonas aeruginosa immunotype 1, 3, 4, and 6 lipopolysaccharides. Infect. Immun. 57:3851-3855. Lowbury, E. J. L., and R. J. Jones. 1975. Treatment and prophylaxis for Pseudomonas infections, p. 237-269. In M. R. W. Brown (ed.), Resistance of Pseudomonas aeruginosa. John Wiley & Sons, Inc., New York. Maclntyre, S., R. Lucken, and P. Owen. 1986. Smooth lipopolysaccharide is the major protective antigen for mice in the surface extract from IATS serotype 6 contributing to the polyvalent Pseudomonas aeruginosa vaccine PEV. Infect. Immun. 52:7684. Matthews-Greer, J. M., and H. E. Gilleland, Jr. 1987. Outer membrane protein F (porin) preparation of Pseudomonas aeruginosa as a protective vaccine against heterologous immunotype strains in a burned mouse model. J. Infect. Dis. 155:1282-1291. Pennington, J. E. 1979. Lipopolysaccharide pseudomonas vaccine: efficacy against pulmonary infection with Pseudomonas aeruginosa. J. Infect. Dis. 140:73-80.

INFECT. IMMUN. 16. Pennington, J. E., H. Y. Reynolds, and P. P. Carbone. 1973. Pseudomonas pneumonia. A retrospective study of 36 cases. Am. J. Med. 55:155-160. 17. Pennington, J. E., and G. J. Small. 1987. Passive immune therapy for experimental Pseudomonas aeruginosa pneumonia in the neutropenic host. J. Infect. Dis. 155:973-978. 18. Pennington, J. E., G. J. Small, M. E. Lostrom, and G. B. Pier. 1986. Polyclonal and monoclonal antibody therapy for experimental Pseudomonas aeruginosa pneumonia. Infect. Immun. 54:239-244. 19. Pier, G. B., and D. M. Thomas. 1983. Characterization of the human immune response to a polysaccharide vaccine from Pseudomonas aeruginosa. J. Infect. Dis. 148:206-213. 20. Pier, G. B., D. Thomas, G. Small, A. Siadak, and H. Zweerink. 1989. In vitro and in vivo activity of polyclonal and monoclonal human immunoglobulins G, M, and A against Pseudomonas aeruginosa lipopolysaccharide. Infect. Immun. 57:174-179. 21. Sawada, S., T. Kawamura, and Y. Masuho. 1987. Immunoprotective human monoclonal antibodies against five major serotypes of Pseudomonas aeruginosa. J. Gen. Microbiol. 133: 3581-3590. 22. Sawada, S., M. Suzuki, T. Kawamura, S. Fujinaga, Y. Masuho, and K. Tomibe. 1984. Protection against infection with Pseudomonas aeruginosa by passive transfer of monoclonal antibodies to lipopolysaccharides and outer membrane proteins. J. Infect. Dis. 150:570-576. 23. Schiller, N. L., and K. A. Joiner. 1986. Interaction of complement with serum-sensitive and serum-resistant strains of Pseudomonas aeruginosa. Infect. Immun. 54:689-694. 24. Stieritz, D. D., and I. A. Holder. 1975. Experimental studies of the pathogenesis of infections due to Pseudomonas aeruginosa: description of a burned mouse model. J. Infect. Dis. 131:688691. 25. Stoll, B. J., M. Pollack, L. S. Young, N. Koles, R. Gascon, and G. B. Pier. 1986. Functionally active monoclonal antibody that recognizes an epitope on the 0 side chain of Pseudomonas aeruginosa immunotype-1 lipopolysaccharide. Infect. Immun. 53:656-662. 26. Suzuki, H., Y. Okubo, M. Moriyama, M. Sasaki, Y. Matsumoto, and T. Hozumi. 1987. Human monoclonal antibodies to Pseudomonas aeruginosa produced by EBV-transformed cells. Microbiol. Immunol. 31:959-966. 27. Yokota, S., H. Ochi, H. Ohtsuka, M. Kato, and H. Noguchi. 1989. Heterogeneity of the L-rhamnose residue in the outer core of Pseudomonas aeruginosa lipopolysaccharide, characterized by using human monoclonal antibodies. Infect. Immun. 57: 1691-1696. 28. Young, L. S. 1972. Human immunity to Pseudomonas aeruginosa. II. Relationship between heat-stable opsonins and typespecific lipopolysaccharides. J. Infect. Dis. 126:277-287. 29. Zweerink, H. J., M. C. Gammon, C. F. Hutchison, J. J. Jackson, D. Lombardo, K. M. Miner, J. M. Puckett, T. J. Sewell, and N. H. Sigal. 1988. Human monoclonal antibodies that protect mice against challenge with Pseudomonas aeruginosa. Infect. Immun. 56:1873-1879.