Microbial Pathogenesis 1989; 6: 287-295

Different sensitivity of Pseudomonas aeruginosa exotoxin A and diphtheria toxin to enzymes from polymorphonuclear leukocytes*

Gerd DGringt and Evelyn Miiller

Hygiene- institute, University of Ttibingen, Tiibingen, F. R. G

(Received December 1, 1988; accepted in revised form February 2, 1989)

Doring, G. (Hygiene-Institute, University of Tubingen, Tubingen, F.R.G.) and E. Miiller. Different sensitivity of Pseudomonas aeruginosa exotoxin A and diphtheria toxin enzymes from poly- morphonuclear leukocytes. Microbial Pathogenesis 1969; 6: 287-295.

We demonstrate that exotoxin A (ExoA) of Pseudomonas aeruginosa is one to two orders of magnitude more sensitive than diphtheria toxin (DT) of Corynebacterium diphtheriae to lysosomal enzymes from polymorphonuclear leukocytes (PMN). It is especially sensitive to PMN elastase which inactivates its cell free enzymatic activity and its cytotoxicity as measured with the Chinese hamster ovary cell assay and the rabbit skin test. Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) showed a rapid fragmentation of ExoA into small peptides at low PMN elastase concentrations, whereas DT remained largely uncleaved at PMN elastase concentrations 10 times higher. PMN elastase also removed the cell surface receptors for ExoA and DT on Chinese hamster ovary cells, suggesting that both toxins may be ineffective at local sites of severe inflammation. A comparison of fibroblasts from cystic fibrosis patients and normal healthy individuals revealed no differences in susceptibility to either DT or ExoA; this tends to exclude a genetic defect as an explanation for the absence of ExoA effects in cystic fibrosis patients.

Key words: Pseudomonas aeruginosa exotoxin A; diphtheria toxin; clinical diphtheria; PMN enzymes; cystic fibrosis.

Introduction

Diphtheria toxin (DT) of Corynebacterium diphtheriae and exotoxin A (ExoA) of Pseudomonas aeruginosa belong to a family of bacterial toxins which inactivate or modify specific target proteins in mammalian cells.‘,* Minute amounts of DT and ExoA cause death of most eukaryotic cells, including human cells.* The intoxication of both enzymes is thought to proceed by (i) receptor-mediated endocytosis of the pro- enzymes,3-5 (ii) generation of the enzymatically active fragment A which is translocated to the cytosol and (iii) covalent linkage of the adenosine 5’-diphosphate (ADP)- ribosyl moiety of oxidized nicotine-amide adenine dinucleotide (NAD) to elongation factor 2 (EF-2), terminating protein synthesis6,’

Despite this striking similarity in the mode of action on a molecular level, DT and ExoA differ in their nucleotide sequences,‘,’ and show significant homology only in

* In honour of the 50th birthday of Prof. Dr Konrad Botzenhart. t Author to whom correspondence should be addressed: Prof. Gerd Doring, Abteilung Allgemeine Hygiene

und Umwelthygiene, Hygiene-lnstitut, Silcherstrasse 7, D-7400 Tubingen, F.R.G.

0882-401 O/89/040287+09 $03.00/O @ 1989 Academic Press Limited

288 G. Dbrmg and E. Miiller

the enzymatic centre.” The structural differences may explain the binding of the toxins to different cell receptors’,” and consequently different cell susceptibilities as well as different modes of toxin activation.‘,”

The contribution of DT to the disease has been well established, while that of ExoA is unclear. Diphtheria, like tetanus and botulism, is a toxin-mediated disease.” C. diphtheriae is a non-invasive pathogen which commonly forms pseudomembranes in the upper respiratory tract. At this site the organisms secrete DT which is spread via the bloodstream; the resulting toxicity to internal organs is believed to contribute significantly to rapid death after infection.‘* In the localized chronic P. aeruginosa lung infection in patients with cystic fibrosis, on the other hand, toxic effects are not seen in parts of the body other than the Iungs,‘3,‘4 and it is uncertain whether ExoA represents a major virulence factor in this disease. Cystic fibrosis patients may be colonized for more than 10 years with high numbers of P. aeruginosa organisms.‘5 ExoA production by most P. aeruginosa strains in vivo is indirectly proven by the detection of toxin-specific serum antibodies in chronically infected patients.‘“,”

Several things could account for the clinical differences between ExoA and DT: firstly, cells from cystic fibrosis patients may be more resistant to ExoA than cells from normal individuals; such a difference has been demonstrated for a number of steroid hormones.‘8~‘g Secondly, DT and ExoA may differ in their ability to withstand the proteolytic enzymes present at the infection site.”

Recently, we reported that ExoA is cleaved in vitro by the lyosomal elastase from polymorphonuclear leukocytes into small fragments and loses its ADP-ribosylating activity.” Here, we show that ExoA is significantly more sensitive than DT to lysosomal- derived PMN-enzymes. We further show that fibroblasts of cystic fibrosis patients are no more resistant to either ExoA or DT than fibroblasts of normal individuals.

Results

influence of PMN enzymes on ADP-ribosyl transferase activity of ExoA and DT Figure 1 shows that a 5 min incubation of ExoA with PMN elastase at a 1 : 1 (w/w) ratio was sufficient to inactivate the ADP-ribosyl transferase activity of the toxin. In contrast, DT retained 90% of its enzymatic activity up to a PMN elastase: DT ratio of 50: 1. The myeloperoxidase (MPO)/hydrogen peroxide/halide system” affected ExoA but not DT at a 1 : 1 ratio (Fig. 1). No loss of enzymatic activity was noticed in preliminary experiments when ExoA or DT were incubated at a 1 : 1 ratio with cathepsin

Influence of P/MN-enzymes on cytotoxicity of ExoA and DT The Chinese hamster ovary (CHO) cell assay is a more sensitive measure of toxin activity than the cell-free ADP-ribosyl transferase assay, meaning that lower tox- in: PMN-enzyme ratios can be used. Various toxin-PMN-enzyme mixtures were added to CHO cell monolayers after various preincubation times (Fig. 2). The inactivation of 16 ng ExoA by PMN elastase was time dependent. Complete inactivation was seen by 60 min with PMN elastase: ExoA ratios of 1 : 1 and 10 : 1 and by 10 min at a 70 : 1 ratio (Fig. 2). DT has a 10 times higher cytotoxic effect on CHO cells than ExoA. Therefore, we incubated only 1.6 ng DT with the CHO cells. This amount of DT was stable at PMN elastase: DT ratios of 10: 1 and 100: 1 for a period of 120 min and was only slightly affected at a ratio of 700: 1 after 100 min incubation (Fig. 2). No cytotoxic effects were seen in control experiments when CHO cells were treated with PMN elastase/phenylmethylsulphonyl fluoride (PMSF) alone.

In contrast to PMN elastase, cathepsin G had no effect on a cytotoxic concentration

PMN enzymes: toxin interactions 289

I IO 25 50 75 100

PMN enzyme toxrn rotlo (w/w)

Fig. 1. Influence of PMN enzymes on the enzymatic activity of diphtheria toxin (DT) and exotoxin A (ExoA) in a cell-free assay. DT and ExoA were incubated for 5 min at 37°C at various PMN elastase: toxin or myeloperoxidase (MPO)/hydrogen peroxide/halide: toxin ratios (w/w). The reaction mixtures were assayed for ADP-ribosyl transferase activity. ‘* The lower limit of sensitivity of the assay was 310 ng for ExoA or DT. Enzymatic activity of PMN-enzyme: toxin mixtures was expressed as the percentage of cpm to toxin controls. Values represent means of quadruplicate values in two independent assays. SD was ~20%. n -m = DT+PMN elastase; m---m = DT+MPO/hydrogen peroxide/halide; l -. = ExoA+PMN elastase; o---o = ExoA+MPO/hydrogen peroxide/halide.

of ExoA at ratios of 150 : 1 or 300 : 1, or on DT at ratios of 300 : 1 or 1500 : 1, even after 1 h at 37°C. When 8.7 ng ExoA were incubated with the MPO system for 5 min at 37°C at a 1 : 1 ratio of MPO : ExoA, 50% cytotoxicity was measured, compared to 100% cytotoxicity in the control. The same inactivating effect was seen when 870 pg DT were incubated at a 1 : 1 ratio with the MPO system.

ExoA DT

1009 = ; = = , I \ I 70 I ‘,700 I

‘;; 5

I : \ , x 1 :

c $ 50-j

‘\ &---,

i; I

2 I 0 I

I ,

0 - ~-t---t---t----1----+---1 - I I I 1 0 20 40 60 80 100 120 0 20 40 60 00 100 120

Preincubotion time (mln)

Fig. 2. Influence of PMN enzymes on cytotoxicity of diphtheria toxin (DT) and exotoxin A (ExoA). ExoA (16 ng) or DT (1.6 ng) were incubated with PMN elastase at the indicated ratios (w/w) for various periods and the reaction mixtures then added to 2~10~ Chinese hamster ovary (CHO) cells. Cytotoxicity was assessed after 34 days by medium colour change,3’ and by cell counting after trypsin treatment in the Neubauer cell chamber. Values represent means of six values in three independent assays. SD was ~20%. The lower limit of sensitivity of the CHO cell assay was about 10 ng for ExoA and about 1 ng for DT. ~-- = Toxin: PMN elastase incubations; __ = toxin controls.

290 G. Doring and E. Mtiller

Cleavage of toxin receptors on CHO ceils by PMN elastase The cellular uptake of ExoA and DT requires toxin-specific cell surface receptors.‘,” Proteinases may remove such receptors and thus inhibit toxin uptake. Therefore, we incubated various concentrations of PMN elastase with 1 .5x105 CHO cells for 1 h and subsequently challenged the cells with cytotoxic concentrations of ExoA or DT. In order to prevent rapid resynthesis of toxin receptors and subsequent cell uptake, the incubation was performed at 4”C, since previous studies22,23 have shown that at this temperature ExoA binds to receptors but is not internalized. In contrast, DT did not bind to CHO cells at 4°C. Therefore we incubated DT for 5 min at 37°C with the proteinase-pretreated cells. Preincubation of CHO cells with PMN elastase at a concentration of 140 pg/ml was found to completely protect them from the cytotoxic effects of both toxins (data not shown).

Effect of toxins on normal fibroblast cultures derived from a healthy individual and a patient with cystic fibrosis When fibroblasts derived from either a healthy individual or a cystic fibrosis patient were incubated with 100 ng of either ExoA or DT, 100% cytotoxicity was noticed (data not shown). As the toxin concentrations were decreased, there was a continuous increase in cell survival for both fibroblast cell lines. 10 ng of ExoA caused only 50% cytotoxicity, and the same amount of DT caused no cytotoxic effect. Thus, fibroblasts from both sources were equally sensitive to DT or to ExoA. In contrast to CHO cells, the fibroblasts were less sensitive to DT than to ExoA.

Effect of untreated and PMN elastase-treated toxins on rabbit skin Incubation of ExoA with an equal amount of PMN elastase resulted in a complete loss of toxicity in the rabbit skin assay, whereas a similar incubation mixture of DT with PMN elastase retained full toxicity (not shown).

Analysis of PMN elastase effect on ExoA and DT by SDS-PAGE Incubation of ExoA with PMN elastase at a 0.5: 1 ratio for 5 min generated a 50000 Da fragment [Fig. 3(a), lane 11. Intact ExoA was only marginally visible with silver staining at a PMN elastase : ExoA ratio of 10 : 1 (lane 3). PMN elastase partially cleaved DT at ratios of 1 : 1 and 10: 1, resulting in a 52000 Da fragment [Fig. 3(b), lanes 1, 21; even at a 10 : 1 ratio, the majority of the DT was not cleaved.

Discussion and conclusions

The present study addresses the intriguing question of the difference between DT and ExoA in clinical infections. The fact that ExoA is not a major virulence factor in chronic P. aeruginosa lung infections in patients with cystic fibrosis, in contrast to the role of DT in patients with diphtheria can be largely explained by the striking difference in the stability of the two toxins towards PMN enzymes, especially PMN elastase. The importance of PMN elastase in these infectious processes is suggested by the fact that the bacterial environment at both infection sites, the pseudomembrane in diphtheria and the purulent sputum in patients with cystic fibrosis, contains high numbers of PMN and bacteria,‘2,‘4 in addition to fibrin clots, necrotic epithelium, lymphocytes and some erythrocytes.

The cell-free ADP-ribosyl transferase assay showed DT to be about 50 times more resistant than ExoA to the proteolytic activity of PMN elastase. A similar difference in stability of the two toxins was seen with the more sensitive CHO cell assay: a 10 min preincubation of ExoA with PMN elastase at a 70: 1 ratio totally eliminated the

PMN enzymes: toxin interactions

a

PMN- ElA

b

DT

1 2 3 4

0 -30k

-20k

291

-67k

-43k

5

PMN- EU

12

-Sk

Fig. 3. Toxin cleavage by PMN elastase analysed by SDS-PAGE. Exotoxin A (ExoA) and diphtheria toxin (DT) were incubated with PMN elastase (PMN-ELA) for 5 min at 37°C. the reaction stopped by adding phenylmethyl-sulphonyl fluoride, and the samples subjected to 1 O-22% gradient gels under non-reducing conditions after heating. ” Gels were stained with silver. (a) PMN elastase: ExoA ratios were 0.5: 1 (lane 1). 1 :l (lane 2) and 1O:l (lane 3). ExoA alone (lane 4), PMN elastase alone (lane 5). Arrow indicates small amounts of intact ExoA. (b) PMN elastase: DT ratios were 1 : 1 (lane 1) and 10: 1 (lane 2). DT alone (lane 3), PMN elastase alone (lane 4). lsoenzyme bands of PMN elastase due to different sugar moities of the glycoprotein are visible between 30000 Da and 25000 Da, as well as autodigestion bands of the PMN elastase below 20 000 Da.

292 G. Dbring and E. Muller

enzymatic activity of ExoA, whereas a 10 times higher ratio decreased DT activity by only 50% during 100 min of incubation. These results were confirmed by using the rabbit skin assay, where intracutaneously injected DT pretreated with PMN elastase retained its full necrotizing effect but similarly pretreated ExoA was totally inactive. In contrast to PMN elastase, cathepsin G, the other major lysosomal neutral proteinase of PMN, did not inactivate either ExoA or DT. Also, oxidation by the MPO system may inactivate the bacterial toxins; our results from the cell free assay (Fig. 1) suggest that again ExoA is more susceptible to this damage than DT. Interestingly, pertussis toxin, cholera toxin, and shiga toxin are significantly more stable than ExoA towards PMN elastase (unpublished data). Thus, ExoA seems exceptional in this family of intracellular bacterial toxins.

The fact that skin fibroblasts from a patient with cystic fibrosis are as susceptible to toxin as fibroblasts derived from a normal healthy donor suggests that toxin uptake is not impaired in these patients. Direct and indirect evidence for the participation of ExoA in the pathogenesis of P. aeruginosa infections in animals as well as in humans24,25 derives from the infections in which the immune system is impaired and the infection is not localized but bacteraemic. Typically, PMN numbers in these situations are low. Thus, active or passive immunization against ExoA may have potential clinical benefit in immunocompromised patients but not in patients with cystic fibrosis.

In addition to a direct inactivation of toxins by PMN elastase, host cells may also be protected in viva by a second mechanism, the cleavage by PMN elastase of toxin- specific cell surface receptors which are essential for toxin uptake.‘,4,5 In support of this, we demonstrated that pretreatment of CHO cells with PMN elastase protected them from a lethal concentration of both DT and ExoA.

Previous studies on the effects of both toxins on PMN reported that PMN were resistant to ExoA, deduced form lack of effect on uptake and killing of bacteria,26 but were sensitive to DT, deduced from reduced incorporation of [3H]leucine at low DT concentrations.27 These results agree with the present findings that ExoA but not DT is sensitive to the proteolytic activity of PMN elastase. However, interestingly, human macrophages have been shown to be sensitive to both ExoA’* and DT.27

Materials and methods

Patients. Sputa from 50 patients with cystic fibrosis and P. aeruginosa lung infection were obtained for the isolation of P. aeruginosa strains. The diagnosis of cystic fibrosis was based on accepted criteria, 2g including a typical history of cystic fibrosis with markedly elevated sweat electrolyte levels in repeated tests and altered lung function. Aseptic skin biopsies were taken from the left forearms of one cystic fibrosis patient and one healthy individual for fibroblast cell cultures. Informed consent was obtained from all patients or their parents.

Materials. Purified DT and ExoA were purchased from the Swiss Serum and Vaccine Institute (Bern). The lyophilized toxins were dissolved in Dulbecco’s modified Eagle medium (DMEM) and aliquots were stored at -70°C until use. Human PMN elastase was purified from the pooled sputum of cystic fibrosis patients suffering from chronic P. aeruginosa lung infections by affinity chromatography on Trasylol Sepharose and by ion exchange chromatography on carboxymethyl- Sepharose (Pharmacia GmbH, Freiburg) as described previously.20 The purity of the enzyme was checked with SDS-PAGE using the method developed by Laemmli.30 Only isoenzyme bands of the PMN elastase were seen between 30000 Da and 25000 Da due to the different sugar moities of the glycoprotein. The proteinase had a specific activity of 204 U/mg using the chromogenic peptide substrate methoxysuccinyl-~-alanyl-~-alanyl-~-prolyl-~-valine-p-nitroan- ilide (Bachem Feinbiochemika, Bubendorf.” Cathepsin G had a specific activity of 23.3 U/mg using succinyl-~-phenylalanyl-~-leucyl-~-phenyl-alanine-~-nitroanalide (Bachem) as a substrate.” Using these enzyme-specific substrates, no contamination of PMN elastase with cathepsin G or vice versa was noticed. Human MPO was purified from pooled sputum material

PMN enzymes: toxin interactions 293

from patients with cystic fibrosis as described previously.” The enzyme had a specific activity of 93 U/mg measured with o-dianisidine (Sigma, Munich) as a substrate.” No proteolytic activity was found in the enzyme preparation using the above mentioned chromogenic substrates. [U-14C]NAD was obtained from Amersham Buchler (Braunschweig). EF-2 was purified from untreated wheat germ (Sigma).3’

Cell-free toxin assay. Toxin: PMN enzyme mixtures were incubated for 5 min at 37°C and then subjected to the ADP-ribosyl transferase assay. 32 The reaction mixture contained 25 ~1 partly purified EF-2, 25 ~1 50 m M Tris/HCI buffer, pH 8.2, 1 mM EDTA, 40 mM dithiothreithol, 10 ~1 toxin dilutions and 5 ~1 [U-14C]NAD. ExoA was preactivated with 8 M urea and 2% dithiothreithol, DT was not preactivated (DT was approximately 10 to 20% nicked, according to the manufacturer’s information). The PMN elastase: DT mixtures were 1 : 1, 10: 1, 25: 1, 50 : 1, 75 : 1 and 100 : 1 (w/w); the PMN elastase : ExoA mixtures were 0.1 : 1, 1 : 1 and 10 : 1 (w/w). The MPO: ExoA and MPO: DT mixtures were 0.1 : 1 and 1 : 1 (w/w). Incubations were carried out for 20 min at 37°C. Protein was precipitated with 10% trichloracetic acid, collected, washed, and counted in a p-counter (Zinsser Analytic, Frankfurt). The ADP-ribosyl transferase activity of the toxin : PMN enzyme mixtures was expressed as a percentage of the cpm of toxin controls.

Cell-toxin interactions. For interactions of cells with ExoA and DT CHO cells were used. CHO cells were propagated in DMEM, supplemented with 10% fetal calf serum, 1% pen- rcillin/streptomycin, 1% L-glutamine, 1% sodium pyruvate and 1% non-essential amino acids. 2x 1 O4 cells were incubated in flat-bottom microtiter plates (Greiner, Nurtingen) at 37°C in 5% CO, atmosphere. ExoA and DT were preincubated with PMN elastase, cathepsin G or MPO. Then the incubation mixtures were added to the cell suspensions. Toxins were dissolved in DMEM; DT (1.6 ng) was incubated with PMN elastase at proteinase: toxin ratios of 10: 1, 100: 1 and 700: 1 (w/w); ExoA (16 ng) was incubated with PMN elastase at proteinase : toxin ratios of 1 : 1, 10 : 1 and 70 : 1. Incubations were carried out at 37°C for 5, 10, 20, 40, 60, 80, 100 and 120 min. DT was incubated with cathepsin G at proteinase: toxin ratios of 300: 1 and 1500 : 1; ExoA was incubated with cathepsin G at proteinase : toxin ratios of 150 : 1 and 300 : 1. Incubations were carried out for 1 h at 37°C. DT or ExoA were incubated with MPO, 0.1 M

NaCl and 0.05% hydrogen peroxide” at a MPO-toxin ratio of 1 : 1 for 5 min at 37°C. The reactions were stopped by adding a molar surplus of PMSF (Sigma, Munich) for the incubations with the proteinases, or catalase (10 ~1; 42 mg/ml) (Sigma) for the incubation with MPO. Cytotoxicity of ExoA and DT for CHO cells was assessed by the characteristic colour change of the medium from yellow to violet in the case of cell death after 34 days3’ and by cell counting after trypsin treatment in the Neubauer cell chamber.

In another experiment CHO cells were incubated with PMN elastase prior to incubation with toxins. 8x 1 O5 CHO cells were preincubated with 1.4 pg, 14 pg and 140 pg PMN elastase for 1 h at 37”C, centrifuged and washed with DMEM at 800 rpm, incubated at 4°C for 10 min with 2 pg ExoA or at 37°C for 5 min with 2 pg DT in DMEM, centrifuged and washed two times with cold DMEM. Then 1 .5x105 cells were incubated in microtiter plates for 4 days at 37°C. Cytotoxicity was assessed as described above.

For the fibroblast cell cultures, skin biopsies were taken from the left forearms of one cystic fibrosis patient and one healthy individual using a 3 mm biopsy punch (Stiefel, Offenbach). The biopsies were cut into small pieces and cultivated in minimal essential medium, Eagle (MEM) (Seromed), supplemented with 10% fetal calf serum, 1% L-glutamine, 1% pen- icillin/streptomycin and 2% UltroserG (BF Biochemics, France). 2x 1 O4 subcultivated fibroblasts from between the fifth and tenth population doublings were challenged with serial dilutions of ExoA or DT and incubated for 4 days at 37°C and 5% COP. Cell death was evaluated by light microscopy. Cells were counted after trypsin treatment in the Neubauer cell chamber.

Rabbit skin test. DT and ExoA were each incubated with PMN elastase at a ratio of 1 : 1 (w/w) for 1 h at 37°C. The reactions were stopped by adding a molar surplus of CI, -antiproteinase (Sigma). 100 ~1 of the reaction mixtures and toxin controls containing 1 pg ExoA or 100 pg DT, respectively, were injected intracutaneously into the shaved skin of a rabbit. The amount of necrosis was assessed by light photography after three days.

Analysis of toxin-PMN elastase incubations by SDS-PAGE. PMN elastase: DT mixtures of 1 : 1 and 10: 1 and PMN elastase: ExoA mixtures of 0.5 : I,1 : 1 and 10 : 1 (w/w) were incubated for 5 min at 37°C in DMEM. After the addition of 5 ~1 PMSF (1.3 mg/ml), the samples were

294 G. Doring and E. Muller

heated to 100°C for 5 min and subjected to a IO-22% gradient polyacrylamide gel under nonreducing conditions. Gels were silver stained (silver stain kit, Biorad, Munich).

We are indebted to Dr Erich Schnabel, Bayer AG, Wuppertal, Federal Republic of Germany, for a kind gift of cathepsin G. We thank Dr Stan Cryz, Jr., Swiss Serum and Vaccine Institute, Berne, Switzerland, and Dr Catherine B. Saelinger, University of Cincinnati, Cincinnati, U.S.A., for valuable suggestions with the manuscript. This work was supported by a grant from the Ministerium ftir Wissenschaft und Kunst, Baden-Wiirttemberg, F.R.G.

References

1. Middlebrook JL, Dorland RB. Bacterial toxins: cellular mechanisms of action. Microbial Rev 1984; 48: 199-221.

2. Gill DM. Bacterial toxins: a table of lethal amounts. Microbial Rev 1982; 46: 86-94. 3. Goldstein JL, Anderson RGW, Brown MS. Coated pits, coated vesicles, and receptor-mediated

endocytosis. Nature 1979; 279: 679-85. 4. Morris RE, Gernstein AS, Bonventre PF, Saelinger CB. Receptor-mediated entry of diphtheria toxin

into monkey kidney (Vero) cells: electron microscopic evaluation. Infect lmmun 1985; 50: 721-7. 5. Morris RE, Manhart MD, Saelinger CB. Receptor-mediated entry of pseudomonas toxin: methylamine

blocks clustering step. Infect lmmun 1983; 48: 806-l 1. 6. lglewski BH, Kabat D. NAD-dependent inhibitions of protein synthesis by Pseudomonas aeruginosa

toxin. Proc Nat1 Acad Sci USA 1975; 72: 2284-8. 7. Collier RJ. Diphtheria toxin: mode of action and structure. Bacterial Rev 1975; 39: 54-85. 8. Gray GL, Smith OH, Baldridge JS, Hoskins RN, Vasil ML, Chen EY, Heynecker HL. Cloning, nucleotide

sequence and expression in Escherichia co/i of the exotoxin A structural gene of Pseudomonas aeruginosa. Proc Natl Acad Sci USA 1984; 81: 2645-9.

9. Greenfield L, Bjorn MJ, Horn G, Fong D, Buck GA, Collier RJ, Kaplan DA. Nucleotid sequence of the structural gene for diphtheria toxin carried by corynebacteriophage p. Proc Natl Acad Sci USA 1983; 80: 6853-7.

10. Carroll SF, Collier RJ. Amino acid sequence homology between the enzymic domains of diphtheria toxin and Pseudomonas aeruginosa exotoxin A. Mol Microbial 1988; 2: 293-6.

11. Vasil ML, lglewski BH. Comparative toxicities of diphtheria toxin and Pseudomonas aeruginosa exotoxin A: evidence for different cell receptors. J Gen Microbial 1978; 108: 333-7.

12. Goldstein E, Hoeprich PD. Diphtheria. In: Hoeprich PD, ed. Infectious diseases. Hagerstown: Harper and Row, 1977; 247-58.

13. Heiby N. Microbiology of lung infections in cystic fibrosis patients. Acta Paediatr Stand [Suppl.] 1982; 301: 34-54.

14. Myers MG, Koontz FP, Weinberger M. Lower respiratory Infections in patients with cystic fibrosis. In: Lloyd-Still JD, ed. Textbook of cystic fibrosis. Boston: John Wright, 1983; 91-107.

15. Heiby N, Doring G, Schiotz PO. The role of immune complexes in the pathogenesis of bacterial infections. Ann Rev Microbial 1986; 40: 29-53.

16. Klinger JD, Straus DC, Hilton CB, Bass JA. Antibody to proteases and exotoxin A of Pseudomonas aeruginosa in patients with cystic fibrosis. J Infect Dis 1978; 138: 49-58.

17. Doring G, Goldstein W, Roll W, Schietz PO, Hmiby N, Botzenhart K. Role of Pseudomonas aeruginosa exoenzymes in lung infections of patients with cystic fibrosis. Infect lmmun 1985; 49: 557-62.

18. Epstein J, Breslow JL, Davidson RL. Increased dexamethasone resistance of cystic fibrosis fibroblasts. Proc Natl Acad Sci USA 1977; 74: 5642-6.

19. Breslow JL, Epstein J, Fontaine JH. Dexamethasone-resistant cystic fibrosis fibroblasts show cross- resistance to sex steroids. Cell 1978; 13: 663-9.

20. Goldstein W, Doring G. Lysosomal enzymes from polymorphonuclear leukocytes and proteinase inhibitors in patients with cystic fibrosis. Am Rev Respir Dis 1986; 134: 49-56.

21. Klebanoff SJ. Myeloperoxidase-halide-hydrogen peroxide antibacterial system. J Bacterial 1968; 95: 2131-8.

22. FitzGerald D, Morris RE, Saelinger CB. Receptor-mediated internalization of Pseudomonas toxin by mouse fibroblasts. Cell 1980; 21: 867-73.

23. Manhart MD, Morris RE, Bonventre PF, Leppla S, Saelinger CB. Evidence for Pseudomonas exotoxin A receptors on plasma membrane of toxin-sensitive LM fibroblasts. Infect lmmun 1984; 45: 596-603.

24. Saelinger CB, Snell K, Holder IA. Experimental studies on the pathogenesis of infections due to Pseudomonas aeruginosa: direct evidence for toxin production during Pseudomonas infection of burned skin tissues. J Infect Dis 1977; 136: 55561.

25. Pollack M. The role of exotoxin A in Pseudomonas disease and immunity. Rev Infect Dis 1983; 5 [Suppl]: 979-84.

26. Pavlovskis OL, Callahan LT Ill, Meyer RD. Characterization of exotoxin of Pseudomonas aeruginosa. J Infect Dis 1974; 130 [Suppl]: 100-2.

PMN enzvmes : toxin interactions 295

27. Saelinger CB, Bonventre PF, lmhoff J. interaction of toxin of Corynebacterium diphtherae with phagocytes from susceptible and resistant species. J Infect Dis 1975; 131: 431-8.

28. Pollack M, Anderson SE Jr. Toxicity of Pseudomonas aeruginosa exotoxin A for human macrophages. Infect lmmun 1978; 19: 10926.

29. Wood RE, Boat TF, Doershuk CF. Cystic fibrosis: state of the art. Am Rev Respir Dis 1976; 113: 833- 78.

30. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 1970; 227: 680-5.

31. lglewski BH, Sadoff JC. Toxin inhibitors of protein synthesis: production, purification, and assay of Pseudomonas aeruginosa toxin A. Methods Enzymol 1979; 60: 780-93.

32. Collier RJ, Kandel J. Structure and activity of diphtheria toxin I. Thiol dependent dissociation of a fragment of toxin into enzymatically active and inactive fragments. J Biol Chem 1971; 246: 1496-503.