Masahiko Endoh

Adenylate Cyclase Activity of Bordetella Organisms

Abundant adenylate cyclase activity was found in the phase I cultures not only of Bordetella pertussis but also of B. parapertussis and B. bronchiseptica. The enzyme activity in the culture fluid increased rapidly and reached a peak during the logarithmic growth phase. B. parapertussis and B. bronchiseptica especially produced a high activity of the enzyme in the culture fluid during the logarithmic phase, but little or no activity was detected in the cells throughout the growth period. In the culture of B. pertussis, the intracellular activity was higher than that in the culture fluid. Phase III cultures of these species lacked both the extracellular and intracellular enzyme activities throughout their growth.

Purification and characterization of heat-labile toxin from Bordetella bronchiseptica

The heat‐labile toxin (HLT) of Bordetella bronchiseptica was purified successively from sonic extracts of phase I organisms grown in Stainer‐Scholte medium, by partition in hydrophobic interaction, sucrose density gradient centrifugation, gel filtration through Sepharose 4B and 6B, isoelectric precipitation and isoelectric focusing. The purified HLT was homogeneous by disc Polyacrylamide gel electrophoresis and the gel diffusion‐test, and free of detectable hemagglutinin and endotoxin activity. A 386‐fold purification over the crude extract was obtained at a yield of about 28%, and a minimum dose of 0.9 ng was dermonecrotizing with a lesion 5 mm in diameter in guinea pigs and induced splenoatrophy. The mouse LD50 was 200 ng (intraperitoneal) or 70 ng (intravenous). The HLT was found to be a simple protein with an isoelectric point of pi 6.9. It has a molecular weight of 102,000 estimated by Sepharose 6B gel filtration and was found to consist of two different types of polypeptide by SDS‐polyacrylamide gel electrophoresis, their molecular weights being 30,000 and 20,000. Amino acid analysis showed 15 common amino acid residues, and methionine, cysteine and tryptophan were undetectable. The HLT crystallized by methylpentanediol showed a block form. The HLT was inactivated at 56 C when heated for 10 min, and at above pH 9 and below pH 4.

Efficacy of chemically cross-linked antigens for acellular pertussis vaccine.

The efficacy of chemically cross-linked antigens for acellular pertussis vaccine was investigated by a murine respiratory infection model. Filamentous hemagglutinin (FHA), pertussis toxin (PT) and pertactin, prepared from a supernatant fluid of Bordetella pertussis culture or the bacterial cells, were chemically cross-linked by the addition of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysulfosuccinimide (NHSS). The biological activities of FHA and PT were inactivated by the cross-linking reaction. The cross-linked antigens induced potent protective immunity against B. pertussis more than the vaccine prepared by formalin treatment (0.2 M). These data suggest a possibility that the vaccine prepared by using EDC and NHSS is more effective than the formalin-treated vaccine.

Effect of Purified Heat-Labile Toxin of Bordetella bronchiseptica on the Peripheral Blood Vessels in Guinea Pigs or Suckling Mice

The heat-labile toxin (HLT) produced by Bordetella species induces a hemorrhagic dermonecrosis with its surrounding ischemic lesions and a splenic atrophy when injected into animals (2, 10). Since its discovery, this toxin has been generally regarded as an important virulence factor (4) in the pathogenesis of the disease. However the mechanism by which HLT induces these lesions is poorly understood. Previously we described the purification and characterization of HLT from B. bronchiseptica (1). In the present work, we examined the in vivo effect of the purified HLT on the peripheral blood vessels to clarify its mode of action in guinea pigs and mice, and found that HLT affects the peripheral blood vessels to induce vasoconstriction. The purified HLT was obtained from B. bronchiseptica strain H-16 by the method described previously (1). Its minimum dermonecrotizing dose (MND) was approximately 0.75 ng in the shaved back of female Hartley guinea pigs, weighing 300 to 350 g. The ddY strain female or male mice, 2-day-old, were also used throughout the studies. Under pentobarbiturate anesthesis administrated intraperitoneally, a part of the dorsal skin was surgically reflected and fixed under a microscope so as to be able to observe the exposed peripheral blood vessels on the inner surface of skin. Before and after giving a 10 ,ƒÊl dose (10 MNDs) of HLT by instillation or intracutaneous injection, the vessels were microscopically observed continuously for 3 hr, and afterward at 30-min-intervals up to 12 hr. In the guinea pigs the peripheral arterioles exposed to HLT constricted remarkably after a lag period of about 15 min

In Vitro Assay for Histamine-Sensitizing Factor of Bordetella pertussis

(HLT). Purified samples of HA, ET, KA, and HLT were obtained by the methods of Nakase et al (5-8). In vivo assays for HSF and LPF were performed in 5to 6-week-old , SPF female ICR mice, by the method of Morse and Morse (3). Anti-HSF sera were obtained by injecting rabbits subcutaneously at one-week intervals with 20-50 ƒÊg of purified HSF. The serum was collected after 9 or 10 injections. The in vitro assay for HSF was performed as follows . Paired epididymal adipose tissues were excised from male Wistar rats weighing 100-300 g, and digested for 45 min with collagenase at a concentration of 10 mg per 1 g of tissue in KrebsRinger bicarbonate (KRB) buffer (pH 7.4) containing 2% bovine serum albumin , under a gas phase of 95% O2 and 5% CO2 . The adipocytes were obtained by filtering the digested tissue through a nylon filter and washing it three times with the KRB buffer.

Cytopathic Effect of Heat‐Labile Toxin of Bordetella parapertussis on Aortic Smooth Muscle Cells from Pigs or Guinea Pigs

The in vitro effect of the heat‐labile toxin (HLT) of Bordetella parapertussis on HeLa, baby hamster kidney (BHK), chinese hamster ovary (CHO), myeloma (P3‐NS‐1), human embryonic lung (HEL‐R66) cells, erythrocytes, adipocytes and lymphocytes from guinea pigs, mice or rats, or aortic smooth muscle cells from pigs or guinea pigs was examined. Within 8, 6, 4, 2, and 2 hr after the exposure to 1, 3, 10, 30, and 100 MNDs/ml of HLT, respectively, the cultured smooth muscle cells only showed a cytopathic change. When the cells exposed to HLT were washed out within 60 min post‐exposure, the change could be induced with an extend period of lag. Histamine, KCl or norepinephrine caused similar change in the cells, but the period of lag was within 30 min. The HLT activity was neutralized by an anti‐B. parapertussis‐ or B. bronchiseptica‐HLT guinea pig IgG. HLT had no effects on any other cells tested.

Contractile Action of Heat-Labile Toxin of Bordetella parapertussis on Aortic Smooth Muscles of Pigs

Using both vascular smooth muscle strips (VSMS) and cultured cells (VSMC) from aortas of pigs, the contractile action of Bordetella heat‐labile toxin (HLT) purified from B. parapertussis was studied in an attempt to elucidate the mechanisms of its action. HLT induced contractions in VSMC in parallel with the increase of Ca2+‐influx. The HLT‐induced Ca2+‐influx and contraction were not influenced by verapamil or diltiazem, though a certain extension of the lag period was seen. The contractile action of HLT on VSMS and VSMC was not influenced either by diltiazem or quinacrine; that on VSMC was not influenced by prednisolone, indomethacin, aspirin, CV‐3988, FPL‐55712, ruthenium red, or TEAC. On VSMS, prednisolone caused the extension of lag period following HLT exposure. The action of HLT on VSMS was inhibited by TMB‐8, whereas that on VSMC was not though the extension of lag period was seen. The HLT‐induced contraction in both VSMS and VSMC was completely inhibited by H‐7. The contraction in VSMS, but not in VSMC, was inhibited by H‐8. HLT did not induce specific activation of the protein kinases in VSMC. The addition of cGMP or cAMP brought about relaxation in the HLT‐exposed VSMS contracting in maximum. HLT caused a significant increase in permeability of VSMC membrane to trypan blue, accompanied with contraction. Both HLT‐induced contraction and increase in permeability were inhibited by dextran of M.W. 8,000, but not of M.W. 5,000. These results suggested that HLT acted on vascular smooth muscle cells by damaging the membrane permeability, but not by disturbing the known cascades or systems for physiological contractions, resulting in the increase in Ca2+‐influx and then contractions.

Effect of Bordetella Heat‐Labile Toxin on Perfused Lung Preparations of Guinea Pigs

In vitro effects of the Bordetella HLT on the isolated perfused lung and some other tissue preparations from guinea pigs were examined. When HLT (30 to 300 MNDs/ml) was administered, an increase of the perfusion pressure was induced in the perfused lungs, indicating vasoconstriction. When 100 or 300 MNDs/ml of HLT was given, the pressure increase appeared after a lag period of 3.5 to 4 min, reached a maximum within 8 to 13 min, and then slowly decreased by 60 to 80% 25 min after exposure. In calcium‐free medium, the pressure increase dur to HLT did not occur, but these HLT‐treated lungs manifested an increase without any lag period immediately after the calcium‐free medium was replaced by normal medium containing calcium. No difference in the response of the perfused lungs to histamine was observed before and after exposure to HLT. The arterial strip did not respond to HLT, but after predigestion with a collagenase and elastase solution the contractive response to 100 MNDs/ml of HLT appeared with a lag period of 1 min. HLT had no effect on the pharmacological responses of the isolated atria, deferent canal or intestinal preparations, or on the ciliary movement of cultured tracheal rings.

Glycerol-Releasing Activity of Histamine-Sensitizing Factor of Bordetella pertussis for Rat Adipocytes In Vitro

Histamine‐sensitizing factor (HSF) purified from Bordetella pertussis induced specifically the release of glycerol from rat epididymal adipocytes in vitro. The most sensitive and reproducible results were obtained by using 1 to 2 × 105 adipocytes/tube from rats weighing 150 to 200 g, and by incubation at 37 C for 180 min. After a lag period of about 60 min, HSF‐treated adipocytes released glycerol in increasing amounts between 60 and 240 min, depending on the dose of HSF. A close correlation between the glycerol‐releasing (GR) activity of HSF for adipocytes and histamine‐sensitizing or leukocytosis‐promoting activity in mice was observed. The GR activity was inactivated by heating at 56 C for 60 min, 63 C for 30 min or 96 C for 10 min. The adipocytes washed out with a Krebs‐Ringer bicarbonate buffer immediately after being exposed to HSF for 1 to 3 min manifested about 75% of the total GR activity induced by HSF, and those washed out after being exposed for 30 min or longer had full activity. Anti‐HSF serum neutralized the activity when it was added to adipocytes simultaneously with HSF, but did not when it was added 30 min after being exposed to HSF. By using both native and 125I‐labeled HSF, the ratio of binding of HSF to adipocytes was estimated to be 10 to 15% of the total HSF per 2 × 105 cells/tube, and to be about 1,000 molecules of HSF per cell to induce the release of glycerol. The GR activity induced with 10 ng of HSF was inhibited by addition of insulin at a dose of over 1 μIU/tube, but not by concanavalin A.

Efficacy of pertussis vaccines consisted of antigens detoxified with tea-leaf catechins

The ability of tea-leaf catechins to detoxify agents was examined. Filamentous hemagglutinin (FHA) and pertussis toxin (PT) were detoxified by the catechins at an extraordinarily lower concentration compared with that of formalin. The sera from the mice immunized by the catechin-treated antigens recognized, not only catechin-treated, but also untreated antigens. Furthermore, catechin-treated PT induced the antibody to neutralize PT activity in the sera of the immunized mice. Pertussis vaccines were prepared including antigens detoxified by the treatment of catechins and intraperitoneally injected into mice. Protection against Bordetella pertussis infection was shown in mice immunized with the vaccines prepared by treatment with catechins. These data suggest that catechins are effective toxoiding agents for preparing a pertussis vaccine.