JOURNAL OF BACTERIOLOGY, JUIY 1990, p. 3905-3908
Vol. 172, No. 7
0021-9193/90/073905-04$02.00/0
Molecular Cloning and Characterization of an Extracellular Protease Gene from Aeromonas hydrophila OCTAVIO RIVERO, JUAN ANGUITA, CARMEN PANIAGUA, AND GERMAN NAHARRO* Departamento de Patologia Animal (Sanidad Animal), Unidad de Microbiologia, Facutad de Veterinaria, Universidad de Leon, 24071 Leon, Spain Received 29 December 1989/Accepted 14 April 1990
A structural gene which codes for an extracellular protease in Aeromonas hydrophila S02/2 and D13 was cloned in Escherichia coli C600-1 by using pBR322 as a vector. The gene codes for a temperature-stable protease with a molecular mass of approximately 38,000 daltons. The protein was secreted to the periplasm of E. coli C600-1 and purified by osmotic shock. Cloned protease (P3) was identical in molecular mass and properties to the one purified from A. hydrophila S02/2 culture supernatant as an extracellular product.
Aeromonas hydrophila and related aeromonads are gramnegative, facultatively anaerobic freshwater bacteria, which are pathogens of humans and fish. In humans they cause soft-tissue wound infections and diarrheal disease, and in fish they cause fatal hemorrhagic septicemia (2, 6, 11, 16, 20, 34). The pathogenicity of the microorganism may involve several extracellular enzymes including two hemolysins, enterotoxins, and proteases (15, 17, 18). Some of the toxins have been isolated and biochemically characterized, but their roles in the pathogenesis of A. hydrophila is still unknown (14). It has been suggested that proteolytic enzymes excreted by Aeromonas spp. play an important role in invasiveness and establishment of infection by overcoming initial host defenses and by providing nutrients for cell proliferation (14, 23). There are many reports describing the number and nature of proteases found in culture supernatants of A. hydrophila. Some of them have reported finding the temperature-stable metalloprotease (TSMP) alone (1, 8, 9, 19). Other reports have described the TSMP to be one of the two proteases produced, the other being temperature-labile serine protease (7, 28). Leung and Stevenson (22) examined 47 A. hydrophila strains and 29 other aeromonads for protease production under uniform cultural conditions. They found two distinct types of proteases in extracellular products: a temperaturestable metalloprotease (TSMP) and a temperature-labile serine protease. In this study we report the molecular cloning and efficient expression in Escherichia coli of an extracellular protease gene from A. hydrophila S02/2 and D13.
selected on Luria agar supplemented with ampicillin (200 ,ug/ml) and 1% skim milk. A. hydrophila strains were grown in Luria broth or on tryptic soy agar (Biolife). For protease production, cultures were grown as described by Allan and Stevenson (1). Preparation of DNA. Chromosomal DNAs from A. hydrophila S02/2 and D13 were obtained by the method of Priefer et al. (31). Plasmid DNA used as a cloning vector (pBR322) was propagated in E. coli C600-1, extracted by an alkaline lysis procedure, and purified by two rounds of ultracentrifugation through CsCl-ethidium bromide gradients followed by extensive dialysis against TE buffer (10 mM Tris hydrochloride [pH 8.0], 1 mM EDTA) (3). Plasmid DNA from transformants was extracted by a rapid-boiling method (12). Genomic library. pBR322 was completely digested with BamHI, and terminal phosphates were removed with calf intestinal alkaline phosphatase as recommended by the manufacturer (Boehringer GmbH, Mannheim, Federal Republic of Germany). A. hydrophila chromosomal DNAs were partially digested with Sau3A and applied to a sucrose density gradient. Fragments between 4 and 10 kilobases (kb) in size were used to construct the genomic libraries. After purification, Sau3A fragments were mixed with BamHIdigested, dephosphorylated pBR322 at a 3:1 ratio and ligated with T4 DNA ligase as suggested by the manufacturer. Ligation mixture was used to transform E. coli C600-1 by the calcium chloride shock procedure (24). Hybridization study. Electroblotting and hybridization were performed by random-primed DNA labeling with digoxigenin-dUTP and detection of hybrids by enzyme immunoassay as specified by the manufacturer (Boehringer GmbH, Mannheim, Federal Republic of Germany). Purification of proteases. Osmotically shocked E. coli C600-1 cells were prepared as described elsewhere (27). Protease P3 from osmotic fluids or extracellular products was purified as follows. Starting materials were fractionated with ammonium sulfate, and the 35 to 65% ammonium sulfate-insoluble material was suspended in 10 mM potassium phosphate buffer (pH 6.5) and applied to a PD-10 column (Pharmacia, Uppsala, Sweden) to remove salts. The desalting materials were applied to a DEAE-Sephadex A25 column (2 by 20 cm) equilibrated with 10 mM potassium phosphate buffer (pH 6.5). The column was washed with 300 ml of the application buffer, and bound material was then eluted with 500 ml of a linear gradient of 0 to 1.0 M KCl in the same buffer. Fractions with protease activity were con-
MATERIALS AND METHODS Bacterial strains and plasmids. E. coli K-12 strain C600-1 was used as the recipient for initial cloning experiments (26). A. hydrophila S02/2, D13, SB7, FP3, SA14, SM6, SM17, SA8, F18, PC5, SA5, and PM10 were isolated in our laboratory from freshwater samples (16, 29, 30) and identified by using the API 20NE system. Other A. hydrophila strains used were ATCC 7966, 67-P-24, and 1.54 (32). Multicopy plasmid pBR322 was used in cloning experiments (4). Plasmids pUC18 and pUC19 (35) were used in subcloning experiments. Media and culture conditions. E. coli was grown in Luria broth or on Luria agar (1.5%) (25). Transformants were *
Corresponding author. 3905
3906
RIVERO ET AL.
centrated on a membrane (PM-10; Amicon Corp., Lexington, Mass.) and then applied to a Sephacryl S-200 column (2.2 by 100 cm) equilibrated with 10 mM potassium phosphate buffer. Fractions with proteolytic activity were used for sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis analysis. Protein determination. Protein levels were determined by using a protein assay kit (Bio-Rad Laboratories, Richmond, Calif.). Bovine serum albumin was used as the standard. Electrophoresis. Denatured samples of enzymes were run on SDS-polyacrylamide gels by using the Laemmli procedure (21) and staining with Coomassie brilliant blue R. Proteolytic activity was visualized in the SDS-polyacrylamide gel by incorporating 0.2% sodium caseinate (10).
J. BACTERIOL. OWE
PLASMID *) PsK P P
I
II
pGN003 E
pN012
EK BSj1
I I 11
II
11
1
1
Prt. +
11 I
pGN04 A
8 P,
pNOO5 EK
A
f
I111
RESULTS AND DISCUSSION Molecular cloning of A. hydrophila protease gene. Genomic libraries of A. hydrophila S02/2 and D13 were constructed in E. coli C600-1 as indicated in Materials and Methods. Approximately 2,500 ampicillin-resistant (Ap') transformants of each strain were selected on Luria agar supplemented with ampicillin (200 ,ug/ml) and skim milk (1%). Ninety percent of the transformants were tetracycline sensitive (Tcs), indicating a high percentage of A. hydrophila DNA insertion into the BamHI site on pBR322. A clear zone, representing the degradation of milk proteins, surrounded three transformants from the genomic library of strain S02/2 and one transformant from the genomic library of strain D13, after 72 h at 37°C. Plasmid DNA was extracted from these four isolates and used to transform E. coli C600-1 again. When these cells were grown on a Luria agar medium supplemented with ampicillin and skim milk, as indicated above, 100% of the colonies became Apr, Tcs, and protease positive (Prt+) with all four plasmids, pGN002, pGN012, and pGN008, from the genomic library of the strain S02/2, and pGN003, from the genomic library of the strain D13. As control, strain C600-1 untransformed and transformed with pBR322 was used. Neither of the controls became Prt+. For further studies, plasmids pGN012 and pGN003 were used (Fig. 1). Analysis and expression of protease genes in E. coli C600-1. Restriction endonuclease analysis of plasmid DNAs, isolated from transformants with proteolytic activity, demonstrated that pGN002 was 20 kb long with a 15.7-kb insert, pGN012 was 9.8 kb long with a 5.3-kb insert, and pGN003 was 9.5 kb long with a 5-kb insert. All three plasmids showed identical restriction endonuclease maps in the region near the tetracycline promoter. Similarly, the 1.4-kb BamHI fragment cloned from pGN012 hybridized to a 2.2-kb EcoRI fragment in pGN003 (data not shown) as well as to a 6.1-kb fragment from A. hydrophila chromosomal DNA (Fig. 2, lanes 4 and 5). The protease gene was localized, within a 1.9-kb BamHII Sau3A junction-EcoRI fragment near the tetracycline promoter, by subcloning experiments with the following constructions. The 1.4-kb BamHI fragment from pGN012 was subcloned into the BamHI site of plasmids pUC18 and pUC19 in either orientation (Fig. 1, plasmids pGNOO4 and pGN005). When E. coli C600-1 cells were transformed with plasmids pGN004 and pGN005, no proteolytic activity was detected. We further constructed plasmids pGN006 and pGN007 by subcloning the 0.7-kb BglII-EcoRI fragment from pGN012 into the BglII-EcoRI site of pGN004 and pGN005 (Fig. 1). Only E. coli C600-1 transformed with pGN006 showed the Prt+ phenotype. All culture media were
11
EK BF PsBE
BE PsK PvPv
1
Ps) E
+
PR-t.
-
Prt
-
P, Bo B A
11
PsBA
11
BOh &gB KE IlI I 11II
A
PGN007
Prt,
Prt +
A
Irt -
lKb
FIG. 1. Restriction endonuclease maps of plasmid clones carrying the prt gene (pGN003 and pGN012). pGN004 and pGN006 are pUC18 derivatives; pGN005 and pGN007 are pUC19 derivatives. , pBR322, pUC18, or pUC19 DNA. -, A. hydrophila DNA; Abbreviations: A, AflIII; B, BamHI; (B), BamHV/Sau3A junction; Bg, BglII; E, EcoRI; K, KpnI; Ps, PstI; Pv, PvuII. Arrows indicate tetracycline promoter (pGN003 and pGN012) or ,B-galactosidase promoter (pGN004, pGN005, pGN006, and pGN007).
supplemented with isopropyl-j3-D-thiogalactopyranoside (IPTG) when used to grow E. coli C600-1 transformed with pUC18, pUC19, or derivative plasmids. These results suggest that the P3 prt gene is being read from the pUC18 lac promoter in plasmid pGN006 and from the pBR322 tet promoter in the original clones. Proteolytic activity, when determined in E. coli carrying pGN003 or pGN012, was not detected in the culture supernatant but was present in osmotic-shock fluids. When genes encoding exotoxins are cloned in E. coli K-12, they are neither processed nor excreted efficiently, but are accumulated in the periplasm (5, 13). It has been previously shown (33) that some E. coli strains contain several distinct soluble enzymes capable of degrading proteins to acid-soluble ma-
- _
FIG. 2. Southern hybridization analysis of genomic DNA from A. hydrophila digested with BamHI and probed with the 1.4-kb BamHI fragment from pGN004. Lanes: 1, ATCC 7966; 2, 67-P-24; 3, 1.54; 4, S02/2; 5, D13; 6, SB7; 7, FP3; 8, SA14; 9, SM6; 10, SA17; 11, lambda bacteriophage cut with HindIII; 12, SA8; 13, F18; 14, PC5; 15, SA5; 16, PM10.
CLONING OF A PROTEASE GENE FROM A. HYDROPHILA
VOL. 172, 1990
Kd 94
A
B
C
I)
E
TABLE 1. Summary of purification of P3 protease from osmoticshock fluid of E. coli C600-1 transformed with pGN003
mo
Stage
.4, -im
2f). |i
14
3907
Total protein
(p.g) 8,960.6 Osmotic shock (NH4)2SO4 35 to 65% 307.0 83.2 DEAE-Sephadex 20.0 Sephacryl S-200 a b
Total tb Relative % Reatity covery activity Sp act aciiyaciiycvr
312.8 256.0 156.0 47.0
34.9 833.8
1,875.0 2,350.0
1 23.6 53.7 67.3
100 81.8 49.8 15.0
Data are absorbance units per hour. Data are absorbance units per hour per milligram of protein.
FIG. 3. SDS-polyacrylamide gel electrophoresis of purified P3 protease from the culture supernatant of A. hydrophila S02/2 (lane
B) and from osmotically shocked fluid from E. coli C600-1 cells transformed with pGN003 (lane C) and P3 proteolytic activity in an SDS-polyacrylamide gel containing 0.2% sodium caseinate from either source (lanes D and E). Molecular mass markers (A) are indicated in kilodaltons (kd); from top to bottom, phosphorylase b, bovine serum albumin, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor, and a-lactalbumin.
terial. E. coli C600-1 is a highly mutagenized strain, and we could not detect any proteolytic activity capable of degrading azocasein, either in the periplasm or in culture supernatants, under our assay conditions. Southern blot analysis. Fifteen high-molecular-weight DNAs from A. hydrophila strains were BamHI digested, electrophoresed in 0.8% agarose, and hybridized with the 1.4-kb BamHI fragment from pGN012, containing most of the P3 prt gene. Only two strains (67-P-24 and 1.54, considered highly virulent) (32), did not hybridize with the probe (Fig. 2). The other strains gave three patterns of hybridization, since three sizes of fragments were obtained. DNA from A. hydrophila ATCC 7966 hybridized to a 3.8-kb fragment. Strains S02/2 and D13 (the sources of DNA for our cloning experiments), and other virulent and avirulent strains, gave a 6.1-kb hybridization fragment. The third hybridization band corresponds to a 1.2-kb fragment found in strains SM6, SA8, and F18. These results demonstrate that the P3 prt gene is not present in the genome of all A. hydrophila strains analyzed and would suggest that minor differences also occur within the protease gene nucleotide sequences, giving rise to different behavior after treatment with inhibitors. Purification and properties of protease P3. Protease P3 was purified from the culture supernatant of A. hydrophila S02/ 2, as indicated in Materials and Methods, as a single band on SDS-polyacrylamide gel electrophoresis, with a molecular mass of 38,000 daltons (Fig. 3, lane B). A protein with an identical molecular mass was purified from osmotically shocked E. coli C600-1 cells transformed with pGN003 (Fig. 3, lane C). Following the same protocol for osmotically shocked E. coli C600-1 cells transformed with pBR322, neither proteolytic activity nor a protein band on SDSpolyacrylamide gel electrophoresis could be detected. Table 1 shows the purification steps for protease P3 from either source.
The proteolytic activity of P3 protein was visualized in situ by electrophoresis in a sodium caseinate-containing polyacrylamide gel (Fig. 3, lanes D and E). P3 protein retained proteolytic activity once SDS was removed. Purified P3 protease, from either A. hydrophila S02/2 or E. coli C600-1 transformed with pGN003, presented quite identical properties concerning inhibitors and heat stability (Ta-
ble 2). Enzyme activity was 40% inhibited by the metal-ion chelator EDTA, 92% inhibited by the thiol component dithiothreitol, and 25% inhibited by phenylmethylsulfonyl fluoride, a serine protease inhibitor. Ninety-five percent of the proteolytic activity remained after heating at 56°C for 30 min. These results suggest that purified P3 protease acts differently from the TSMP purified from A. hydrophila NCR 505 and Ba5 (22), since P3 protease is not affected by the presence of EDTA (an inhibitor of metalloproteases) to the same degree that the TSMP mentioned above. Also, P3 protease is about 25% inhibited by phenylmethylsulfonyl fluoride, whereas the TSMP from strains NCR 505 and Ba5 is not. These results would also support the suggestions of Nieto and Ellis (28) that significant differences can be found in the characteristics of extracellular products and extracellular proteases isolated from different strains of A. hydrophila. When purified P3 protease from both sources, A. hydrophila culture supernatant, and E. coli C600-1 osmoticshock fluids transformed with plasmid pGN003 was incubated at 37°C overnight, no self-proteolytic activity was observed. When extracellular products from A. hydrophila S02/2 were treated at 56°C for 30 min, 25% of the proteolytic activity remained. This activity was still inhibited by dithiothreitol. These results indicate that two types of proteolytic activities are present in the culture supernatant of A. hydrophila S02/2 and that 25% of the total proteolytic activity corresponds to a thermostable proteolytic activity. Isolation of genes encoding extracellular proteases will provide more detailed information on the number and nature of these extracellular enzymes. TABLE 2. Effect of inhibitors and heat treatment on proteolytic activity of purified P3 protease from culture supernatant of
A. hydrophila S02/2 and from osmotically shocked E. coli cells transformed with pGN003a Treatment
None EDTA (10 mM) DTTC (10 mM) Cysteine (5 mM) PMSFd (1 mM) 56°C for 30 min
Remaining proteolytic activity (%) of P3 fromb: E. coli A. hydrophila
100 62.4 9.3 6.8 74.2 94.8
100 60.2 8.2 7.2 75.6 95.2
a P3, from either source, was preincubated with inhibitors at 20°C for 30 min before azocasein substrate was added. b The remaining proteolytic activity is expressed as a percentage of the activity of an untreated sample on azocasein substrate. Data are means of duplicate samples. cDTT, Dithiothreitol. d PMSF, Phenylmethylsulfonyl fluoride.
3908
RIVERO ET AL.
ACKNOWLEDGMENTS This work was supported by grant PB86-0077 from the Comision Asesora de Investigaci6n Cientffica y Tecnica, Ministerio de Educaci6n y Ciencia, Spain. LITERATURE CITED 1. Allan, B., and R. M. W. Stevenson. 1981. Extracellular virulence factors of Aeromonas hydrophila in fish infections. Can. J. Microbiol. 27:1114-1122. 2. Barghouthi, S., R. Young, M. 0. J. Olson, J. E. L. Arceneaux, and B. R. Byers. 1989. Amonabactin, a novel tryptophan- or phenylalanine-containing phenolate siderophore in Aeromonas hydrophila. J. Bacteriol. 171:1811-1816. 3. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1523. 4. Bolivar, F., R. Rodriguez, P. J. Greene, M. C. Betlack, H. L. Heyneker, H. W. Boyer, J. H. Crora, and S. Falkow. 1977. Construction of new cloning vehicles. II. A multiple cloning system. Gene 2:95-113. 5. Chakraborty, T., B. Huhle, H. Bergbauer, and W. Goebel. 1986. Cloning, expression, and mapping of the Aeromonas hydrophila aerolysin gene determinant in Escherichia coli K-12. J. Bacteriol. 167:363-374. 6. Chakraborty, T., B. Huhle, H. Hof, H. Bergbauer, and W. Goebel. 1987. Marker exchange mutagenesis of the aerolysin determinant in Aeromonas hydrophila demonstrates the role of aerolysin in A. hydrophila-associated systemic infection. Infect. Immun. 55:2274-2280. 7. Dahle, H. K. 1971. The purification and some properties of two Aeromonas proteinases. Acta Pathol. Microbiol. Immunol. Scand. Sect. B 79:726-738. 8. Denis, F., and L. Veillet-Poncet. 1984. Purification partielle du systeme proteolytique extracellulaire de Aeromonas hydrophila LP50: etude comparative chromatographique et electrophoretique. Ann. Inst. Pasteur Microbiol. 135A:219-227. 9. Gross, R., and N. W. Coles. 1969. A proteinase produced by Aeromonas hydrophila. Aust. J. Sci. 31:330-331. 10. Hensen, C., and E. B. Dowdle. 1980. Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates. Anal. Biochem. 102:196-202. 11. Hickman-Brenner, F. W., K. L. McDonald, A. G. Steiferwalt, F. R. Faning, D. J. Brenner, and J. J. Farmer III. 1987. Aeromonas veronii, a new ornithine decarboxylase-positive species that may cause diarrhea. J. Clin. Microbiol. 25:900-906. 12. Holmes, D. S., and M. Quigley. 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114: 193-200. 13. Howard, S. P., and J. T. Buckley. 1986. Molecular cloning and expression in Escherichia coli of the structural gene for the hemolytic toxin aerolysin from Aeromonas hydrophila. Mol. Gen. Genet. 204:289-295. 14. Hsu, T. C., W. D. Waltman, and E. B. Shots. 1981. Correlation of extracellular enzymatic activity and biochemical characteristics with regard to virulence of Aeromonas hydrophila. Dev. Biol. Stand. 49:101-111. 15. Janda, J. M. 1985. Biochemical and exoenzymatic properties of Aeromonas species. Diagn. Microbiol. Infect. Dis. 3:223-232. 16. Janda, J. M., L. S. Oshiro, S. L. Abbot, and P. S. Duffey. 1987. Virulence markers of mesophilic aeromonads: association of the autoagglutination phenomenon with mouse pathogenicity and the presence of a peripheral cell-associated layer. Infect. Immun. 55:3070-3077.
J. BACTERIOL.
17. Joanne, M. R., C. W. Houston, D. H. Coppenhaver, J. D. Dixon, and A. Kurosky. 1989. Purification and chemical characterization of a cholera toxin-cross-reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila. Infect. Immun. 57:1165-1169. 18. Joanne, M. R., C. W. Houston, and A. Kurosky. 1989. Bioactivity and immunological characterization of a cholera toxincross-reactive cytolytic enterotoxin from Aeromonas hydrophila. Infect. Immun. 57:1170-1179. 19. Kanai, K., and H. Wakabayashi. 1984. Purification and some properties of proteases from Aeromonas hydrophila. Bull. Jpn. Soc. Sci. Fish. 40:1367-1374. 20. Kindschuh, M., L. K. Pickering, T. G. Clearly, and G. RuizPalacios. 1987. Clinical and biochemical significance of toxin production by Aeromonas hydrophila. J. Clin. Microbiol. 25: 916-921. 21. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. 22. Leung, K. Y., and R. M. Stevenson. 1988. Characteristics and distribution of extracellular proteases from Aeromonas hydrophila. J. Gen. Microbiol. 134:151-160. 23. Leung, K. Y., and R. M. W. Stevenson. 1988. TnS-induced protease-deficient strains of Aeromonas hydrophila with reduced virulence for fish. Infect. Immun. 56:2639-2644. 24. Mandel, M., and A. Higa. 1970. Calcium dependent bacteriophage DNA infection. J. Mol. Biol. 53:156-162. 25. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 26. Nagahare, K., T. Tanaka, F. Hishimuna, M. Kuroda, and K. Sakaguchi. 1977. Control of tryptophan synthetase amplified by varying the number of composite plasmids in E. coli cells. Gene 1:141-152. 27. Neu, H. E., and H. A. Heppel. 1965. The release of enzymes from E. coli by osmotic shock and during formation of spheroplasts. J. Biol. Chem. 240:3685-3692. 28. Nieto, T. P., and E. Ellis. 1986. Characterization of extracellular metallo- and serine-proteases of Aeromonas hydrophila strain B51. J. Gen. Microbiol. 132:1975-1979. 29. Palumbo, S. A., F. Maxino, A. C. Wiliams, R. L. Buchanan, and D. W. Thayer. 1985. Starch-ampicillin agar for the quantitative detection of Aeromonas hydrophila. Appl. Environ. Microbiol. 50:1027-1030. 30. Popoff, M. 1984. Genus III. Aeromonas Kluyver and Van Niel 1936, 398AL, p. 545-548. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. The Williams & Wilkins Co., Baltimore. 31. Priefer, U., R. Simon, and A. Puhler. 1984. Cloning with cosmid, p. 190-201. In A. Puhler and K. N. Timmis (ed.), Advanced molecular genetics. Springer-Verlag KG, Berlin. 32. Santos, Y., A. E. Toranzo, J. L. Barja, T. P. Nieto, and T. G. Villa. 1988. Virulence properties and enterotoxin production of Aeromonas strains isolated from fish. Infect. Immun. 56:32853293. 33. Sreedhara Swamy, K. H., and A. L. Goldberg. 1982. Subcellular distribution of various proteases in Escherichia coli. J. Bacteriol. 149:1027-1033. 34. Thune, R. L., M. C. Johnson, T. E. Graham, and R. L. Amborski. 1986. Aeromonas hydrophila ,-hemolysin: purification and examination of its role in virulence of 0-group channel catfish, Ictalburus punctatus (Rafinesque). J. Fish Dis. 9:55-61. 35. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 9:103-119.
Vol. 172, No. 7
0021-9193/90/073905-04$02.00/0
Molecular Cloning and Characterization of an Extracellular Protease Gene from Aeromonas hydrophila OCTAVIO RIVERO, JUAN ANGUITA, CARMEN PANIAGUA, AND GERMAN NAHARRO* Departamento de Patologia Animal (Sanidad Animal), Unidad de Microbiologia, Facutad de Veterinaria, Universidad de Leon, 24071 Leon, Spain Received 29 December 1989/Accepted 14 April 1990
A structural gene which codes for an extracellular protease in Aeromonas hydrophila S02/2 and D13 was cloned in Escherichia coli C600-1 by using pBR322 as a vector. The gene codes for a temperature-stable protease with a molecular mass of approximately 38,000 daltons. The protein was secreted to the periplasm of E. coli C600-1 and purified by osmotic shock. Cloned protease (P3) was identical in molecular mass and properties to the one purified from A. hydrophila S02/2 culture supernatant as an extracellular product.
Aeromonas hydrophila and related aeromonads are gramnegative, facultatively anaerobic freshwater bacteria, which are pathogens of humans and fish. In humans they cause soft-tissue wound infections and diarrheal disease, and in fish they cause fatal hemorrhagic septicemia (2, 6, 11, 16, 20, 34). The pathogenicity of the microorganism may involve several extracellular enzymes including two hemolysins, enterotoxins, and proteases (15, 17, 18). Some of the toxins have been isolated and biochemically characterized, but their roles in the pathogenesis of A. hydrophila is still unknown (14). It has been suggested that proteolytic enzymes excreted by Aeromonas spp. play an important role in invasiveness and establishment of infection by overcoming initial host defenses and by providing nutrients for cell proliferation (14, 23). There are many reports describing the number and nature of proteases found in culture supernatants of A. hydrophila. Some of them have reported finding the temperature-stable metalloprotease (TSMP) alone (1, 8, 9, 19). Other reports have described the TSMP to be one of the two proteases produced, the other being temperature-labile serine protease (7, 28). Leung and Stevenson (22) examined 47 A. hydrophila strains and 29 other aeromonads for protease production under uniform cultural conditions. They found two distinct types of proteases in extracellular products: a temperaturestable metalloprotease (TSMP) and a temperature-labile serine protease. In this study we report the molecular cloning and efficient expression in Escherichia coli of an extracellular protease gene from A. hydrophila S02/2 and D13.
selected on Luria agar supplemented with ampicillin (200 ,ug/ml) and 1% skim milk. A. hydrophila strains were grown in Luria broth or on tryptic soy agar (Biolife). For protease production, cultures were grown as described by Allan and Stevenson (1). Preparation of DNA. Chromosomal DNAs from A. hydrophila S02/2 and D13 were obtained by the method of Priefer et al. (31). Plasmid DNA used as a cloning vector (pBR322) was propagated in E. coli C600-1, extracted by an alkaline lysis procedure, and purified by two rounds of ultracentrifugation through CsCl-ethidium bromide gradients followed by extensive dialysis against TE buffer (10 mM Tris hydrochloride [pH 8.0], 1 mM EDTA) (3). Plasmid DNA from transformants was extracted by a rapid-boiling method (12). Genomic library. pBR322 was completely digested with BamHI, and terminal phosphates were removed with calf intestinal alkaline phosphatase as recommended by the manufacturer (Boehringer GmbH, Mannheim, Federal Republic of Germany). A. hydrophila chromosomal DNAs were partially digested with Sau3A and applied to a sucrose density gradient. Fragments between 4 and 10 kilobases (kb) in size were used to construct the genomic libraries. After purification, Sau3A fragments were mixed with BamHIdigested, dephosphorylated pBR322 at a 3:1 ratio and ligated with T4 DNA ligase as suggested by the manufacturer. Ligation mixture was used to transform E. coli C600-1 by the calcium chloride shock procedure (24). Hybridization study. Electroblotting and hybridization were performed by random-primed DNA labeling with digoxigenin-dUTP and detection of hybrids by enzyme immunoassay as specified by the manufacturer (Boehringer GmbH, Mannheim, Federal Republic of Germany). Purification of proteases. Osmotically shocked E. coli C600-1 cells were prepared as described elsewhere (27). Protease P3 from osmotic fluids or extracellular products was purified as follows. Starting materials were fractionated with ammonium sulfate, and the 35 to 65% ammonium sulfate-insoluble material was suspended in 10 mM potassium phosphate buffer (pH 6.5) and applied to a PD-10 column (Pharmacia, Uppsala, Sweden) to remove salts. The desalting materials were applied to a DEAE-Sephadex A25 column (2 by 20 cm) equilibrated with 10 mM potassium phosphate buffer (pH 6.5). The column was washed with 300 ml of the application buffer, and bound material was then eluted with 500 ml of a linear gradient of 0 to 1.0 M KCl in the same buffer. Fractions with protease activity were con-
MATERIALS AND METHODS Bacterial strains and plasmids. E. coli K-12 strain C600-1 was used as the recipient for initial cloning experiments (26). A. hydrophila S02/2, D13, SB7, FP3, SA14, SM6, SM17, SA8, F18, PC5, SA5, and PM10 were isolated in our laboratory from freshwater samples (16, 29, 30) and identified by using the API 20NE system. Other A. hydrophila strains used were ATCC 7966, 67-P-24, and 1.54 (32). Multicopy plasmid pBR322 was used in cloning experiments (4). Plasmids pUC18 and pUC19 (35) were used in subcloning experiments. Media and culture conditions. E. coli was grown in Luria broth or on Luria agar (1.5%) (25). Transformants were *
Corresponding author. 3905
3906
RIVERO ET AL.
centrated on a membrane (PM-10; Amicon Corp., Lexington, Mass.) and then applied to a Sephacryl S-200 column (2.2 by 100 cm) equilibrated with 10 mM potassium phosphate buffer. Fractions with proteolytic activity were used for sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis analysis. Protein determination. Protein levels were determined by using a protein assay kit (Bio-Rad Laboratories, Richmond, Calif.). Bovine serum albumin was used as the standard. Electrophoresis. Denatured samples of enzymes were run on SDS-polyacrylamide gels by using the Laemmli procedure (21) and staining with Coomassie brilliant blue R. Proteolytic activity was visualized in the SDS-polyacrylamide gel by incorporating 0.2% sodium caseinate (10).
J. BACTERIOL. OWE
PLASMID *) PsK P P
I
II
pGN003 E
pN012
EK BSj1
I I 11
II
11
1
1
Prt. +
11 I
pGN04 A
8 P,
pNOO5 EK
A
f
I111
RESULTS AND DISCUSSION Molecular cloning of A. hydrophila protease gene. Genomic libraries of A. hydrophila S02/2 and D13 were constructed in E. coli C600-1 as indicated in Materials and Methods. Approximately 2,500 ampicillin-resistant (Ap') transformants of each strain were selected on Luria agar supplemented with ampicillin (200 ,ug/ml) and skim milk (1%). Ninety percent of the transformants were tetracycline sensitive (Tcs), indicating a high percentage of A. hydrophila DNA insertion into the BamHI site on pBR322. A clear zone, representing the degradation of milk proteins, surrounded three transformants from the genomic library of strain S02/2 and one transformant from the genomic library of strain D13, after 72 h at 37°C. Plasmid DNA was extracted from these four isolates and used to transform E. coli C600-1 again. When these cells were grown on a Luria agar medium supplemented with ampicillin and skim milk, as indicated above, 100% of the colonies became Apr, Tcs, and protease positive (Prt+) with all four plasmids, pGN002, pGN012, and pGN008, from the genomic library of the strain S02/2, and pGN003, from the genomic library of the strain D13. As control, strain C600-1 untransformed and transformed with pBR322 was used. Neither of the controls became Prt+. For further studies, plasmids pGN012 and pGN003 were used (Fig. 1). Analysis and expression of protease genes in E. coli C600-1. Restriction endonuclease analysis of plasmid DNAs, isolated from transformants with proteolytic activity, demonstrated that pGN002 was 20 kb long with a 15.7-kb insert, pGN012 was 9.8 kb long with a 5.3-kb insert, and pGN003 was 9.5 kb long with a 5-kb insert. All three plasmids showed identical restriction endonuclease maps in the region near the tetracycline promoter. Similarly, the 1.4-kb BamHI fragment cloned from pGN012 hybridized to a 2.2-kb EcoRI fragment in pGN003 (data not shown) as well as to a 6.1-kb fragment from A. hydrophila chromosomal DNA (Fig. 2, lanes 4 and 5). The protease gene was localized, within a 1.9-kb BamHII Sau3A junction-EcoRI fragment near the tetracycline promoter, by subcloning experiments with the following constructions. The 1.4-kb BamHI fragment from pGN012 was subcloned into the BamHI site of plasmids pUC18 and pUC19 in either orientation (Fig. 1, plasmids pGNOO4 and pGN005). When E. coli C600-1 cells were transformed with plasmids pGN004 and pGN005, no proteolytic activity was detected. We further constructed plasmids pGN006 and pGN007 by subcloning the 0.7-kb BglII-EcoRI fragment from pGN012 into the BglII-EcoRI site of pGN004 and pGN005 (Fig. 1). Only E. coli C600-1 transformed with pGN006 showed the Prt+ phenotype. All culture media were
11
EK BF PsBE
BE PsK PvPv
1
Ps) E
+
PR-t.
-
Prt
-
P, Bo B A
11
PsBA
11
BOh &gB KE IlI I 11II
A
PGN007
Prt,
Prt +
A
Irt -
lKb
FIG. 1. Restriction endonuclease maps of plasmid clones carrying the prt gene (pGN003 and pGN012). pGN004 and pGN006 are pUC18 derivatives; pGN005 and pGN007 are pUC19 derivatives. , pBR322, pUC18, or pUC19 DNA. -, A. hydrophila DNA; Abbreviations: A, AflIII; B, BamHI; (B), BamHV/Sau3A junction; Bg, BglII; E, EcoRI; K, KpnI; Ps, PstI; Pv, PvuII. Arrows indicate tetracycline promoter (pGN003 and pGN012) or ,B-galactosidase promoter (pGN004, pGN005, pGN006, and pGN007).
supplemented with isopropyl-j3-D-thiogalactopyranoside (IPTG) when used to grow E. coli C600-1 transformed with pUC18, pUC19, or derivative plasmids. These results suggest that the P3 prt gene is being read from the pUC18 lac promoter in plasmid pGN006 and from the pBR322 tet promoter in the original clones. Proteolytic activity, when determined in E. coli carrying pGN003 or pGN012, was not detected in the culture supernatant but was present in osmotic-shock fluids. When genes encoding exotoxins are cloned in E. coli K-12, they are neither processed nor excreted efficiently, but are accumulated in the periplasm (5, 13). It has been previously shown (33) that some E. coli strains contain several distinct soluble enzymes capable of degrading proteins to acid-soluble ma-
- _
FIG. 2. Southern hybridization analysis of genomic DNA from A. hydrophila digested with BamHI and probed with the 1.4-kb BamHI fragment from pGN004. Lanes: 1, ATCC 7966; 2, 67-P-24; 3, 1.54; 4, S02/2; 5, D13; 6, SB7; 7, FP3; 8, SA14; 9, SM6; 10, SA17; 11, lambda bacteriophage cut with HindIII; 12, SA8; 13, F18; 14, PC5; 15, SA5; 16, PM10.
CLONING OF A PROTEASE GENE FROM A. HYDROPHILA
VOL. 172, 1990
Kd 94
A
B
C
I)
E
TABLE 1. Summary of purification of P3 protease from osmoticshock fluid of E. coli C600-1 transformed with pGN003
mo
Stage
.4, -im
2f). |i
14
3907
Total protein
(p.g) 8,960.6 Osmotic shock (NH4)2SO4 35 to 65% 307.0 83.2 DEAE-Sephadex 20.0 Sephacryl S-200 a b
Total tb Relative % Reatity covery activity Sp act aciiyaciiycvr
312.8 256.0 156.0 47.0
34.9 833.8
1,875.0 2,350.0
1 23.6 53.7 67.3
100 81.8 49.8 15.0
Data are absorbance units per hour. Data are absorbance units per hour per milligram of protein.
FIG. 3. SDS-polyacrylamide gel electrophoresis of purified P3 protease from the culture supernatant of A. hydrophila S02/2 (lane
B) and from osmotically shocked fluid from E. coli C600-1 cells transformed with pGN003 (lane C) and P3 proteolytic activity in an SDS-polyacrylamide gel containing 0.2% sodium caseinate from either source (lanes D and E). Molecular mass markers (A) are indicated in kilodaltons (kd); from top to bottom, phosphorylase b, bovine serum albumin, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor, and a-lactalbumin.
terial. E. coli C600-1 is a highly mutagenized strain, and we could not detect any proteolytic activity capable of degrading azocasein, either in the periplasm or in culture supernatants, under our assay conditions. Southern blot analysis. Fifteen high-molecular-weight DNAs from A. hydrophila strains were BamHI digested, electrophoresed in 0.8% agarose, and hybridized with the 1.4-kb BamHI fragment from pGN012, containing most of the P3 prt gene. Only two strains (67-P-24 and 1.54, considered highly virulent) (32), did not hybridize with the probe (Fig. 2). The other strains gave three patterns of hybridization, since three sizes of fragments were obtained. DNA from A. hydrophila ATCC 7966 hybridized to a 3.8-kb fragment. Strains S02/2 and D13 (the sources of DNA for our cloning experiments), and other virulent and avirulent strains, gave a 6.1-kb hybridization fragment. The third hybridization band corresponds to a 1.2-kb fragment found in strains SM6, SA8, and F18. These results demonstrate that the P3 prt gene is not present in the genome of all A. hydrophila strains analyzed and would suggest that minor differences also occur within the protease gene nucleotide sequences, giving rise to different behavior after treatment with inhibitors. Purification and properties of protease P3. Protease P3 was purified from the culture supernatant of A. hydrophila S02/ 2, as indicated in Materials and Methods, as a single band on SDS-polyacrylamide gel electrophoresis, with a molecular mass of 38,000 daltons (Fig. 3, lane B). A protein with an identical molecular mass was purified from osmotically shocked E. coli C600-1 cells transformed with pGN003 (Fig. 3, lane C). Following the same protocol for osmotically shocked E. coli C600-1 cells transformed with pBR322, neither proteolytic activity nor a protein band on SDSpolyacrylamide gel electrophoresis could be detected. Table 1 shows the purification steps for protease P3 from either source.
The proteolytic activity of P3 protein was visualized in situ by electrophoresis in a sodium caseinate-containing polyacrylamide gel (Fig. 3, lanes D and E). P3 protein retained proteolytic activity once SDS was removed. Purified P3 protease, from either A. hydrophila S02/2 or E. coli C600-1 transformed with pGN003, presented quite identical properties concerning inhibitors and heat stability (Ta-
ble 2). Enzyme activity was 40% inhibited by the metal-ion chelator EDTA, 92% inhibited by the thiol component dithiothreitol, and 25% inhibited by phenylmethylsulfonyl fluoride, a serine protease inhibitor. Ninety-five percent of the proteolytic activity remained after heating at 56°C for 30 min. These results suggest that purified P3 protease acts differently from the TSMP purified from A. hydrophila NCR 505 and Ba5 (22), since P3 protease is not affected by the presence of EDTA (an inhibitor of metalloproteases) to the same degree that the TSMP mentioned above. Also, P3 protease is about 25% inhibited by phenylmethylsulfonyl fluoride, whereas the TSMP from strains NCR 505 and Ba5 is not. These results would also support the suggestions of Nieto and Ellis (28) that significant differences can be found in the characteristics of extracellular products and extracellular proteases isolated from different strains of A. hydrophila. When purified P3 protease from both sources, A. hydrophila culture supernatant, and E. coli C600-1 osmoticshock fluids transformed with plasmid pGN003 was incubated at 37°C overnight, no self-proteolytic activity was observed. When extracellular products from A. hydrophila S02/2 were treated at 56°C for 30 min, 25% of the proteolytic activity remained. This activity was still inhibited by dithiothreitol. These results indicate that two types of proteolytic activities are present in the culture supernatant of A. hydrophila S02/2 and that 25% of the total proteolytic activity corresponds to a thermostable proteolytic activity. Isolation of genes encoding extracellular proteases will provide more detailed information on the number and nature of these extracellular enzymes. TABLE 2. Effect of inhibitors and heat treatment on proteolytic activity of purified P3 protease from culture supernatant of
A. hydrophila S02/2 and from osmotically shocked E. coli cells transformed with pGN003a Treatment
None EDTA (10 mM) DTTC (10 mM) Cysteine (5 mM) PMSFd (1 mM) 56°C for 30 min
Remaining proteolytic activity (%) of P3 fromb: E. coli A. hydrophila
100 62.4 9.3 6.8 74.2 94.8
100 60.2 8.2 7.2 75.6 95.2
a P3, from either source, was preincubated with inhibitors at 20°C for 30 min before azocasein substrate was added. b The remaining proteolytic activity is expressed as a percentage of the activity of an untreated sample on azocasein substrate. Data are means of duplicate samples. cDTT, Dithiothreitol. d PMSF, Phenylmethylsulfonyl fluoride.
3908
RIVERO ET AL.
ACKNOWLEDGMENTS This work was supported by grant PB86-0077 from the Comision Asesora de Investigaci6n Cientffica y Tecnica, Ministerio de Educaci6n y Ciencia, Spain. LITERATURE CITED 1. Allan, B., and R. M. W. Stevenson. 1981. Extracellular virulence factors of Aeromonas hydrophila in fish infections. Can. J. Microbiol. 27:1114-1122. 2. Barghouthi, S., R. Young, M. 0. J. Olson, J. E. L. Arceneaux, and B. R. Byers. 1989. Amonabactin, a novel tryptophan- or phenylalanine-containing phenolate siderophore in Aeromonas hydrophila. J. Bacteriol. 171:1811-1816. 3. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1523. 4. Bolivar, F., R. Rodriguez, P. J. Greene, M. C. Betlack, H. L. Heyneker, H. W. Boyer, J. H. Crora, and S. Falkow. 1977. Construction of new cloning vehicles. II. A multiple cloning system. Gene 2:95-113. 5. Chakraborty, T., B. Huhle, H. Bergbauer, and W. Goebel. 1986. Cloning, expression, and mapping of the Aeromonas hydrophila aerolysin gene determinant in Escherichia coli K-12. J. Bacteriol. 167:363-374. 6. Chakraborty, T., B. Huhle, H. Hof, H. Bergbauer, and W. Goebel. 1987. Marker exchange mutagenesis of the aerolysin determinant in Aeromonas hydrophila demonstrates the role of aerolysin in A. hydrophila-associated systemic infection. Infect. Immun. 55:2274-2280. 7. Dahle, H. K. 1971. The purification and some properties of two Aeromonas proteinases. Acta Pathol. Microbiol. Immunol. Scand. Sect. B 79:726-738. 8. Denis, F., and L. Veillet-Poncet. 1984. Purification partielle du systeme proteolytique extracellulaire de Aeromonas hydrophila LP50: etude comparative chromatographique et electrophoretique. Ann. Inst. Pasteur Microbiol. 135A:219-227. 9. Gross, R., and N. W. Coles. 1969. A proteinase produced by Aeromonas hydrophila. Aust. J. Sci. 31:330-331. 10. Hensen, C., and E. B. Dowdle. 1980. Electrophoretic analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulfate and copolymerized substrates. Anal. Biochem. 102:196-202. 11. Hickman-Brenner, F. W., K. L. McDonald, A. G. Steiferwalt, F. R. Faning, D. J. Brenner, and J. J. Farmer III. 1987. Aeromonas veronii, a new ornithine decarboxylase-positive species that may cause diarrhea. J. Clin. Microbiol. 25:900-906. 12. Holmes, D. S., and M. Quigley. 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114: 193-200. 13. Howard, S. P., and J. T. Buckley. 1986. Molecular cloning and expression in Escherichia coli of the structural gene for the hemolytic toxin aerolysin from Aeromonas hydrophila. Mol. Gen. Genet. 204:289-295. 14. Hsu, T. C., W. D. Waltman, and E. B. Shots. 1981. Correlation of extracellular enzymatic activity and biochemical characteristics with regard to virulence of Aeromonas hydrophila. Dev. Biol. Stand. 49:101-111. 15. Janda, J. M. 1985. Biochemical and exoenzymatic properties of Aeromonas species. Diagn. Microbiol. Infect. Dis. 3:223-232. 16. Janda, J. M., L. S. Oshiro, S. L. Abbot, and P. S. Duffey. 1987. Virulence markers of mesophilic aeromonads: association of the autoagglutination phenomenon with mouse pathogenicity and the presence of a peripheral cell-associated layer. Infect. Immun. 55:3070-3077.
J. BACTERIOL.
17. Joanne, M. R., C. W. Houston, D. H. Coppenhaver, J. D. Dixon, and A. Kurosky. 1989. Purification and chemical characterization of a cholera toxin-cross-reactive cytolytic enterotoxin produced by a human isolate of Aeromonas hydrophila. Infect. Immun. 57:1165-1169. 18. Joanne, M. R., C. W. Houston, and A. Kurosky. 1989. Bioactivity and immunological characterization of a cholera toxincross-reactive cytolytic enterotoxin from Aeromonas hydrophila. Infect. Immun. 57:1170-1179. 19. Kanai, K., and H. Wakabayashi. 1984. Purification and some properties of proteases from Aeromonas hydrophila. Bull. Jpn. Soc. Sci. Fish. 40:1367-1374. 20. Kindschuh, M., L. K. Pickering, T. G. Clearly, and G. RuizPalacios. 1987. Clinical and biochemical significance of toxin production by Aeromonas hydrophila. J. Clin. Microbiol. 25: 916-921. 21. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. 22. Leung, K. Y., and R. M. Stevenson. 1988. Characteristics and distribution of extracellular proteases from Aeromonas hydrophila. J. Gen. Microbiol. 134:151-160. 23. Leung, K. Y., and R. M. W. Stevenson. 1988. TnS-induced protease-deficient strains of Aeromonas hydrophila with reduced virulence for fish. Infect. Immun. 56:2639-2644. 24. Mandel, M., and A. Higa. 1970. Calcium dependent bacteriophage DNA infection. J. Mol. Biol. 53:156-162. 25. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 26. Nagahare, K., T. Tanaka, F. Hishimuna, M. Kuroda, and K. Sakaguchi. 1977. Control of tryptophan synthetase amplified by varying the number of composite plasmids in E. coli cells. Gene 1:141-152. 27. Neu, H. E., and H. A. Heppel. 1965. The release of enzymes from E. coli by osmotic shock and during formation of spheroplasts. J. Biol. Chem. 240:3685-3692. 28. Nieto, T. P., and E. Ellis. 1986. Characterization of extracellular metallo- and serine-proteases of Aeromonas hydrophila strain B51. J. Gen. Microbiol. 132:1975-1979. 29. Palumbo, S. A., F. Maxino, A. C. Wiliams, R. L. Buchanan, and D. W. Thayer. 1985. Starch-ampicillin agar for the quantitative detection of Aeromonas hydrophila. Appl. Environ. Microbiol. 50:1027-1030. 30. Popoff, M. 1984. Genus III. Aeromonas Kluyver and Van Niel 1936, 398AL, p. 545-548. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1. The Williams & Wilkins Co., Baltimore. 31. Priefer, U., R. Simon, and A. Puhler. 1984. Cloning with cosmid, p. 190-201. In A. Puhler and K. N. Timmis (ed.), Advanced molecular genetics. Springer-Verlag KG, Berlin. 32. Santos, Y., A. E. Toranzo, J. L. Barja, T. P. Nieto, and T. G. Villa. 1988. Virulence properties and enterotoxin production of Aeromonas strains isolated from fish. Infect. Immun. 56:32853293. 33. Sreedhara Swamy, K. H., and A. L. Goldberg. 1982. Subcellular distribution of various proteases in Escherichia coli. J. Bacteriol. 149:1027-1033. 34. Thune, R. L., M. C. Johnson, T. E. Graham, and R. L. Amborski. 1986. Aeromonas hydrophila ,-hemolysin: purification and examination of its role in virulence of 0-group channel catfish, Ictalburus punctatus (Rafinesque). J. Fish Dis. 9:55-61. 35. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 9:103-119.
