Vol. 28, No. 5

0095-1137/90/051062-04$02.00/0 Copyright (O 1990, American Society for Microbiology

Simple, Efficient Purification of Filamentous Hemagglutinin and Pertussis Toxin from Bordetella pertussis by Hydrophobic and Affinity Interaction STEPHEN K. SKELTON AND K. H. WONG* Center for Infectious Diseases, Centers for Disease Control, Atlanta, Georgia 30333 Received 17 October 1989/Accepted 9 January 1990

Two major antigens of Bordetella pertussis, filamentous hemagglutinin (FHA) and pertussis toxin (PT), were efficiently purified from culture filtrate by exploiting their relative hydrophobicities and differences in affinity to sialic acid-containing protein. High yields of FHA (40 to 80 mg/liter) and PT (8 to 16 mg/liter) were first produced by growing the bacteria in the modified CL medium. The FHA and PT in the culture filtrate were adsorbed onto butyl-Sepharose by hydrophobic interaction at appropriately high ionic strength. Elution of the antigens was effected by decreasing their hydrophobicities with a buffer of low ionic strength. FHA was then separated from PT with an affinity column of fetuin-Sepharose. The fraction passing through the column contained purified FHA, and the fetuin-bound PT was eluted with buffered MgCl2. The FHA and PT purified by these steps were electrophoretically and serologically identical to the reference purified FHA and PT preparations. Approximately 16 to 32 mg of purified FHA and 4 to 8 mg of purified PT were obtained from 1 liter of culture filtrate. The described procedure for making FHA and PT antigens from B. pertussis for serologic and immunologic use is very simple, efficient, and reproducible.

passing it through a filter (pore size, 0.2 nm; Nalge/Sybron Corp., Rochester, N.Y.). Antigen yields in the culture filtrate were estimated by enzyme-linked immunosorbent assay (ELISA) (18, 19) to be approximately 40 to 80 mg of FHA per liter and 8 to 16 mg of PT per liter in reference to purified FHA (FHA-R) and PT (PT-R). FHA-R was obtained from Merieux Institute, Paris, France, and PT-R was obtained from the Michigan Department of Health, Lansing, Mich. Purification of FHA and PT was performed at room temperature. To increase the ionic strength for hydrophobic interaction, ammonium sulfate was added to a 500-ml volume of culture filtrate to a final concentration of 1 M. The filtrate was then passed at a flow rate of 150 ml/h through a column (1.5 by 10 cm) containing 10 ml of butyl-Sepharose 4B (Pharmacia LKB Biotechnology Inc., Piscataway, N.J.) which had been equilibrated with 1 M ammonium sulfate in deionized water. These experimental conditions permitted all of the FHA and PT in the culture filtrate to bind to the hydrophobic sites of butyl-Sepharose. No FHA or PT activity was detected in the fraction that had passed through the column by ELISA and hemagglutination with goose erythrocytes (19). After the column was washed with 50 ml of 3 M NaCI in 0.01 M phosphate buffer, pH 7.4, the bound FHA and PT were eluted with 0.05 M NaCI in 0.005 M phosphate, pH 7.4, in a protein peak as monitored at 280 nm and collected in one fraction of approximately 80 ml. The butylSepharose column was regenerated for use by eluting the remaining bound substances with 6 M urea and then washing the column with phosphate-buffered saline. FHA was separated from PT with an affinity column (1.5 by 10 cm) containing 8 ml of fetuin-Sepharose equilibrated with 0.01 M phosphate-buffered saline, pH 7.4. The immunosorbent was prepared by immobilizing fetuin (GIBCO Laboratories, Grand Island, N.Y.) to Sepharose 4B as previously described (18), using CNBr-activated Sepharose 4B (Pharmacia LKB). The fraction of FHA and PT eluted from the hydrophobic column was passed through the fetuin-

The recent advances in identifying and isolating major antigens from Bordetella pertussis (5, 7, 12, 15, 21) have contributed significantly to the understanding of the biology and pathogenesis of pertussis (11). The filamentous hemagglutinin (FHA) is a surface antigen involved in attachment of the bacteria to host cells in the infection process. As the bacteria multiply among the cilia of epithelial cells along the respiratory tract of the host, they secrete pertussis toxin (PT), which is believed to cause most, if not all, of the specific symptoms as well as immune protection (10). Purified FHA and PT have been under intensive investigation for use as immunogens in experimental acellular pertussis vaccines (1, 14) and as antigens for the serodiagnosis of pertussis (2, 4, 18-20). Because of meager antigen yields in the commonly used growth media (3, 16) and the relatively complicated purification techniques, these antigens have been quite expensive and hard to obtain. Their availability has been limited mostly to specialized research laboratories. In our recent studies to find a more practical solution to antigen production and better serologic assays for FHA and PT antibodies for clinical and public health laboratories, we observed that high yields of FHA and PT were consistently produced by toxigenic strains of B. pertussis in the modified CL medium with relatively little contamination from cell lysis during growth and that the ionic strength of the milieu significantly influenced the hydrophobicity of FHA and PT. In the present study, we exploited these attributes as well as the very different affinities of these two antigens to sialic acid-containing proteins to develop a very simple and efficient procedure for their production and purification. B. pertussis NIH 165 was grown in the CL medium (6) as

modified in our previous studies (18, 19). After incubation in a shaker water bath at 37°C for 48 h, the bacteria were separated from the supernatant by centrifugation (12,000 x g, 30 min, 4°C), and the supernatant was sterilized by *

Corresponding author. 1062

VOL. 28, 1990

NOTES Culture filtrate Add (NH4I)2SS4 to 1 M


A 1


B 1




column (Discard pass-through fraction)

(1) Wash column wlth 3 M NaCI ln 0.01 M phosphate, pH 7.4 (Discard washings)












(2) Elute with 0.05 M NaCI in

0.005 M phosphate, pH 7.4

Collect protein Peak: FHA and PT

Fetuin-Sepharose column


Collect pm-througil fracdone Purified FHA

S4, S5

(1) Wash column with 1 M NaCI ln 0.05 M Tria hydrochloride, pH 7.4 (Discard washings)


FIG. 2. Comparison of sodium dodecyl sulfate-polyacrylamide gel electrophoresis profiles of FHA (A) and PT (B) isolated by hydrophobic and affinity interaction with those of purified reference preparations. (A) Lane 1, FHA-R; lane 2, hydrophobic-affinitypurified FHA. (B) Lane 1, PT-R; lane 2, hydrophobic-affinitypurified PT.

(2) Elute with 4 M MgCl2 in 0.05 M Tris hydrochloride, pH 7.4

Collesc protein peak: Purflsd PT

FIG. 1. Schema for purifying FHA and PT from a culture filtrate of B. pertussis.

Sepharose column at 30 ml/h. Purified FHA was collected in a single fraction of about 80 ml as it passed through the affinity column for PT. The fetuin-Sepharose column was washed with 150 ml of 1 M NaCI in 0.05 M Tris hydrochloride, pH 7.4, at a flow rate of 100 ml/h. The bound PT was then eluted with 4 M MgCl2 in 0.05 M Tris hydrochloride, pH 7.4, and was collected as a protein peak in one fraction of about 6 ml. The complete procedure is outlined in Fig. 1. The yields of FHA and PT in culture filtrate and the purification efficiency of the system derived from a representative experiment are summarized in Table 1. B. pertussis NIH 165 produced about 40 to 80 mg of FHA and 8 to 16 mg of PT in 1 liter of the modified CL medium as estimated by ELISA with specific antiserum to FHA or PT (18). The yields were consistent with those observed in our previous studies (18, 19). The growth conditions and the modified CL medium also enhanced production of FHA and PT by other TABLE 1. Yields and purification efficiencies of FHA and PT FHA PT PT yield purifica(mg/liter) purification tion effiassayacinyfcecef(%e) Filtrate Purified (%) Filtrate Purified FHA yield



40-80 ELISA Hemagglutina- 60 tion (0.5% goose cell) Lowry protein NT

16-32 20

40 33



Reference purified FHA and PT yields. b NT, Not tested. a

were used

8-16 NTb

4.5-9 4.5

57 NT





standards for


toxigenic strains (19). After purification, about 16 to 32 mg of FHA and 4.5 to 9 mg of PT were obtained, indicating isolation efficiencies of about 40% for FHA and 57% for PT. The final yields of FHA and PT as quantified by ELISA were comparable with the results determined by hemagglutination with 0.5% goose erythrocytes (19) and by the Lowry assay (9) for protein, using FHA-R and PT-R as reference preparations. The purified FHA and PT were electrophoretically and serologically identical to the reference FHA-R and PT-R. The electrophoretic patterns as determined by the method of Laemmli (8) with a Protean Il 20-cm Slab Cell (Bio-Rad Laboratories, Richmond, Calif.) are shown in Fig. 2. Serologic specificity and sensitivity of the isolated FHA and PT were ascertained by comparison with FHA-R and PT-R by the ELISA procedures described previously (18, 19). Serum specimens from patients with culture-positive pertussis were selected to cover a wide range of titers for the immunoglobulin A (IgA), IgG, and IgM classes of antibodies to FHA and PT. Results from tests performed simultaneously indicated no difference in sensitivity and specificity between hydrophobic-affinity-purified and reference antigens as indicated by antibody titers (Table 2). This study demonstrates that under appropriate conditions hydrophobic interaction is a simple and effective technique to aid in the purification of FHA and PT from culture filtrates of B. pertussis. It works on the principle that the relative hydrophobicities of FHA and PT change in response to ionic strength, and specific adsorption and desorption of the antigens by hydrophobic resins are achieved by manipulating the ionic strength in the column. The experimental conditions defined in this report appear to be optimal with respect to isolation efficiency and specificity. Two other hydrophobic resins, phenyl-Sepharose CL-4B and octylSepharose CL-4B, were tested in this study along with butyl-Sepharose and showed greatly reduced elution effi-




TABLE 2. Serologic specificities of FHA and PT isolated by hydrophobic and affinity interaction compared with reference purified preparations Serum type and specimen


Antibody titerc to:


Culturepositive clinical sera 1 H/A R 2 H/A R 3 H/A R 4 H/A R 5 H/A R 6 H/A R 7 H/A R 8 H/A R 9 H/A R 10 H/A R







6,400 6,400 800

200 200 400 400 400 400 1,600 1,600 Neg


12,800 12,800 6,400 3,200 200 200 Neg 25,600 6,400 25,600 6,400 12,800 400 12,800 400 25,600 100 25,600 100 NT NT NT NT NT NT NT NT





200 Neg 32,000 16,000 200 Neg 32,000 16,000 8,000 Neg Neg Neg 8,000 Neg Neg Neg 400 Neg 64,000 64,000 400 Neg 64,000 64,000 32,000 Neg Neg 16,000 32,000 Neg Neg 16,000 16,000 Neg Neg 16,000 16,000 Neg Neg 16,000 64,000 Neg 1,600 Neg 64,000 Neg 1,600 Neg NT NT 64,000 NT NT NT 64,000 NT NT NT 16,000 NT NT NT 32,000 NT NT 25,600 12,800 64,000 NT 25,600 12,800 64,000 NT 6,400 1,600 16,000 NT 6,400 1,600 32,000

Negative control

with a simple and efficient purification system which requires only commonly used, inexpensive laboratory chemicals and reagents, the production of purified FHA and PT for serologic and immunologic use should now be well within the reach of most public health and clinical laboratories. In view of the importance of these antigens in immunity against pertussis (10, 11, 13), the described procedures may also be very useful in the development and production of an acellular pertussis vaccine containing these antigens as components (1, 14).

1. 2.


4. 5. 6.







Neg Neg Neg Neg

Neg Neg Neg Neg

Neg Neg Neg Neg

Neg Neg Neg Neg

Neg Neg Neg Neg

Neg Neg Neg Neg

a Serum specimens 1 to 10 from patients with culture-proven pertussis were selected to cover a wide range of titers for the IgA, IgM, and IgG classes of antibodies to FHA and/or PT. Specimens 7 to 10 were tested with only FHA or PT because of limited quantities. Negative control sera were selected from blood donors with no history of pertussis immunization and clinical pertussis. b H/A, Hydrophobic and affinity interaction-purified antigen; R, reference purified antigen. c Titers were determined by ELISA procedures in simultaneous tests. Neg, Negative; NT, not tested.

8. 9. 10.



ciency because of strong hydrophobic interaction with the antigens. Strong chaotropic agents were required to elute the antigens, which resulted in nonspecific desorption of other proteins adsorbed from the culture filtrate (data not shown). Svoboda and co-workers (17) reported purification of PI by consecutive chromatography on Blue Sepharose, phenylSepharose CL-4B, and hydroxylapatite. A solution of 17% glycerol in 0.1 M Tris base was used to elute PTI from phenyl-Sepharose CL-4B before further purification with hydroxylapatite, and the isolation efficiency was reported to be 23%. The procedure described in the present study permits isolation of both FHA and PT antigens from one batch of culture filtrate in two simple chromatographie steps, with isolation efficiencies of about 40% for FHA and 57% for PT. The chromatographie steps can be completed within 6 h, and the columns can then be regenerated for repeated use. With a culture system which yields high concentrations of FHA and PT in a culture filtrate with relatively little contamination of other cellular components (18, 19) combined


14. 15. 16. 17.

18. 19.

LITERATURE CITED Ad Hoc Group for the Study of Pertussis Vaccines. 1988. Placebocontrolled trial of two acellular pertussis vaccines in Swedenprotective efficacy and adverse events. Lancet i:955-960. Burstyn, D. G., L. J. Baraff, M. S. Peppler, R. D. Leake, J. St. Geme, Jr., and C. R. Manclark. 1983. Serological response to filamentous hemagglutinin and lymphocytosis-promoting toxin of Bordetella pertussis. Infect. Immun. 41:1150-1156. Cohen, S. M., and M. W. Wheeler. 1946. Pertussis vaccine prepared with phase 1 cultures grown in fluid medium. Am. J. Public Health 36:371-376. Granstrom, G., B. Wretland, C. R. Salenstedt, and M. Granstrom. 1988. Evaluation of serologic assays for diagnosis of whooping cough. J. Clin. Microbiol. 26:1818-1823. Hewlett, E., and J. Wolff. 1976. Soluble adenylate cyclase from culture medium of Bordetella pertussis: purification and characterization. J. Bacteriol. 127:890-898. Imaizumi, A., Y. Suzuki, S. Ono, H. Sato, and Y. Sato. 1983. Effect of heptakis (2,6-O-dimethyl)p-cyclodextrin on the production of pertussis toxin by Bordetella pertussis. Infect. Immun. 41:1138-1143. Irons, L. I., and A. P. MacLenan. 1979. Isolation of the lymphocytosis promoting factor-hemagglutinin of Bordetella pertussis by affinity chromatography. Biochim. Biophys. Acta 580:175-185. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Pittman, M. 1979. Pertussis toxin: the cause of the harmful effects and prolonged immunity of whooping cough. A hypothesis. Rev. Infect. Dis. 1:401-412. Pittman, M. 1984. The concept of pertussis as a toxin-mediated disease. Pediatr. Infect. Dis. 3:467-486. Sato, Y., J. L. Cowell, H. Sato, D. G. Burstyn, and C. R. Manclark. 1983. Separation and purification of the hemagglutinins from Bordetella pertussis. Infect. Immun. 41:313-320. Sato, Y., K. Izumiya, H. Sato, J. L. Cowell, and C. R. Manclark. 1981. Role of antibody to leukocytosis-promoting factor hemagglutinin and to filamentous hemagglutinin in immunity to pertussis. Infect. Immun. 31:1223-1231. Sato, Y., M. Kimura, and H. Fukumi. 1984. Development of a pertussis component vaccine in Japan. Lancet i:122-126. Sekura, R. D., F. Fish, C. R. Manclark, B. Meade, and Y. L. Zhang. 1983. Pertussis toxin-affinity purification of a new ADP-ribosyltransferase. J. Biol. Chem. 258:14647-14651. Stainer, D. W., and M. J. Scholte. 1971. A simple chemically defined medium for the production of phase 1 Bordetella pertussis. J. Gen. Microbiol. 63:211-220. Svoboda, M., E. Hannecart-Pokorni, M. Borremans, and J. Christophe. 1986. Rapid purification of Bordetella pertussis toxin by alternating affinity and hydrophobic chromatography. Anal. Biochem. 159:402-411. Wong, K. H., and S. K. Skelton. 1988. New, practical approach to detecting antibody to pertussis toxin for public health and clinical laboratories. J. Clin. Microbiol. 26:1316-1320. Wong, K. H., and S. K. Skelton. 1989. Preparation of filamentous hemagglutinin from Bordetella pertussis and assay for

VOL. 28, 1990 serum antibodies to filamentous hemagglutinin and pertussis toxin for clinical and public health laboratories. J. Clin. Microbiol. 27:2805-2810. 20. Zackrisson, G., I. Krantz, T. Legergard, P. Larsson, R. Sekura, N. Sigurs, J. Taranger, and B. Trollfors. 1988. Humoral antibody response to pertussis toxin in patients with clinical pertus-



sis measured by an enzyme-linked immunosorbent assay. Eur. J. Clin. Microbiol. 7:149-154. 21. Zhang, J. M., J. L. Cowell, A. C. Steven, and C. R. Manclark. 1985. Purification of serotype 2 fimbriae of Bordetella pertussis and their identification as a mouse protective antigen. Dev. Biol. Stand. 61:173-185.

Simple, efficient purification of filamentous hemagglutinin and pertussis toxin from Bordetella pertussis by hydrophobic and affinity interaction.

Two major antigens of Bordetella pertussis, filamentous hemagglutinin (FHA) and pertussis toxin (PT), were efficiently purified from culture filtrate ...
719KB Sizes 0 Downloads 0 Views