Vol. 125, No. 3 Printed in U.SA.
JOURNAL OF BACTERIOLOGY, Mar. 1976, p. 1120-1126 Copyright C) 1976 American Society for Microbiology
Bacteriophage Resistance in Bacillus subtilis 168, W23, and Interstrain Transformants RONALD E. YASBIN,1 * VERNON C. MAINO,2 AND FRANK E. YOUNG Laboratory ofBiochemical Genetics, National Heart and Lung Institute, Bethesda, Maryland 20014, and Department of Microbiology, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
Received for publication 14 October 1975
Strains of Bacillus subtilis 168 deficient in glucosylated teichoic acid vary in their resistance to bacteriophage infection. Although glucosylated teichoic acid is important for bacteriophage attachment, the results demonstrate that alternate receptor sites exist. Non-glucosylated cell wall mutants could be assigned to specific classes (gtaA, gtaB, gtaC) by their pattern of resistance to three closely related bacteriophages (G25, Oe, SP82). In addition to glucosylation, the type of teichoic acid was also important for bacteriophage attachment. B. subtilis strains 168 and W23 have different teichoic acids in their cell walls and have varied susceptibilities to bacteriophage infection. Transfer of bacteriophage resistance from strain W23 into a derivative of strain 168 was accomplished. The resistant bacteria obtained were impaired in their ability to adsorb bacteriophage and in their capacity to be transfected by bacteriophage deoxyribonucleic acid. The importance of teichoic acids as a bacte- phage resistance is strongly affected by the nariophage receptor site in the wall of the bacte- ture of the genetic defect. Glucosylated teichoic acid has also been rium Bacillus subtilis has been demonstrated (2, 8, 24). Specifically, three classes of bacterio- shown to play an essential role in the attachphage-resistant mutants of B. subtilis strain ment of some of the Bacillus bacteriophages to 168 have been shown to lack glucosylated tei- B. subtilis strain W23 (8, 24). B. subtilis strains choic acid in their cell walls (24, 25). The gluco- W23 and 168 differ in the content of their teisylation of cell wall teichoic acids is regulated choic acids (5, 7) and their susceptibility to by the products of the gtaA +, gtaB +, and gtaC + certain bacteriophages (3, 16; B. E. Reilly, genes (24). These three genes have been Ph.D. thesis, Western Reserve Univ., Clevemapped (25), and the proteins coded by two of land, Ohio, 1965). Furthermore, B. subtilis the genes have been characterized (4, 12). Phos- strain W23 is lysogenic for defective bacteriophoglucomutase (EC 2.7.5.1) is the product of phage PBSZ, whereas strain 168 is lysogenic for the gtaC+ gene, whereas uridine 5'-diphos- defective bacteriophage PBSX (14, 18). To exphate (UDP)-glucose:polyglycerol teichoic acid plore the effect of variability of the cell surface glucosyltransferase (EC 1.4.1) is a product of on viral infection, we transferred bacteriophage the gtaA + gene. Although no enzyme has been resistance from strain W23 to strain 168 via identified for the gtaB + gene, it has been sug- deoxyribonucleic acid (DNA)-mediated transgested that this gene codes for an inactive phos- formation. The results of this transfer as well phoglucomutase monomer (13). In the initial as the patterns of bacteriophage resistance in studies, when only a limited number of bacteri- strain W23 indicate the variety and complexity ophages were examined, it appeared that the of mechanisms by which bacteria interfere with bacterial mutants unable to glucosylate tei- bacteriophage development. choic acid were resistant to certain bacterioMATERIALS AND METHODS phages regardless of whether the mutation was Strains and methods of propagation. The bacteat the gtaA, gtaB, or gtaC locus (24). In this report, we have expanded these studies and can rial strains used in this study are listed in Table 1. now demonstrate that the degree of bacterio- The propagation of bacteriophage and the maintePresent address: Department of Biology, Brookhaven National Laboratory, Upton, N.Y. 11973. 2 Present address: Division of Allergy and Clinical Immunology, National Jewish Hospital and Research Center, Denver, Colo. 80206. '
nance of the bacterial cultures were described previously (21, 22). Bacteriophage SP82 was obtained from M. Green. Media and procedures for genetic exchange. Modified M agar and broth were prepared as previously described (22). The procedures for transforma-
120
BACTERIOPHAGE RESISTANCE
VOL. 125, 1976
1121
TABLE 1. Bacterial strains used Strain
BR151 RUB807
Genotypea
lys-3, trpC2, metBlO lys-3, trpC2, metB10, gtaC20
Origin and remarks Competent derivative of B. subtilis 168 Phage-resistant marker transformed into BR151 using donor DNA from a spontaneous )29-resistant strain obtained by V. C. Maino
(168/qb29) Phage-resistant marker transformed into BR151 using donor DNA from a spontaneous +105-resistant strain obtained by S. Zahler (CU548) RUB809 lys-3, trpC2, metB10, gtaA21 Isolated similarly to RUB807 Isolated similarly to RUB807 RUB810 Iys-3, metBlO, gtaB20 Isolated similarly to RUB807 RUB811 lys-3, trpC2, metB10, gtaC21 Isolated similarly to RUB807 RUB812 lys-3, trpC2, metB10, gtaB21 A spontaneous erythromycin-resistant derivaery-1 RUB815 tive of B. subtilis W23 A prototrophic derivative of BR141 RUB818b Lysogenic for bacteriophage SP02 RUB818(SP02) Lysogenic for bacteriophage O105 RUB818(105) Isolated by resistance to' bacteriophages SPP1 RUB821 lys-3, trpC2, metB10, 4Rc and SP02cl2 after transformation of strain BR151 with DNA obtained from strain RUB815 Isolated as was RUB821 RUB822 lys-3, trpC2, metB10, OR RUB823 Isolated as was RUB821 lys-3, trpC2, metB10, OR RUB824 Isolated as was RUB821 lys-3, trpC2, metB10, O)R a Symbols as defined in reference 26 and references 12, 24, 25, and V. C. Maino (Ph.D. thesis), for gta strains. b Obtained after three separate transformations with nonsaturating concentrations of 168 T+ DNA. c Phenotypically phage resistant.
RUB808
lys-3, trpC2, metB10, gtaA20
tion, transfection, and isolation of DNA were identical to those previously used in our laboratory (22, 23). Determination of plating efficiencies. Bacterial cultures were grown at 37 C in modified M broth that had not been supplemented with Ca2 , Mg2+, or Mn2+. After the cells had ceased exponential growth (optical density of 120 Klett units on a Klett-Summerson colorimeter, filter no. 66), 0.2 ml of the bacterial culture was added to 2.0 ml of modified M semisolid agar containing 0.1 ml of the appropriate dilution of the bacteriophage stock, and the mixture was poured onto modified M agar. This procedure was used for all bacteriophage studied except 429. The plating efficiency of bacteriophage 429 was determined by using a modified M overlay on tryptose blood agar base (Difco). Assays for bacteriophage 429 were done at 30 C, whereas all other quantitation of infectious centers were performed at 37 C. Determination of adsorption efficiencies. Bacterial cultures were grown in modified M broth without Mg2+, Ca2+, and Mn2+ at 37 C to an optical density of 120 Klett units, centrifuged (12,000 x g for 20 min at 4 C), and suspended in Spizizen's minimal salts (17) containing 3.7% formaldehyde (Fisher). The suspension was kept at 4 C for 30 min before the cells were washed three times with modified M broth and finally resuspended in half the original volume in modified M broth. The bacteriophage and the formaldehyde-treated bacteria were mixed to yield a multiplicity of infection of 10-4 to 10-s, incu-
bated at 37 C for 20 min, and centrifuged (8,000 x g for 5 min at room temperature), and the number of plaque-forming units in the broth was determined. Enzyme assays. UDP-glucose:pyrophosphorylase (EC 2.7.7.9), UDP-glucose:polyglycerol teichoic acid glucosyltransferase, and phosphoglucomutase were measured as previously specified (4, 12).
RESULTS Isolation and identification of bacteriophage-resistant mutants of B. subtilis strain 168. DNA was isolated from spontaneously arising bacteriophage-resistant mutants of B. subtilis, and the resistance trait was transformed into strain BR151. These bacteriophageresistant mutants were found to lack glucosylated teichoic acid in their cell walls (V. C. Maino, Ph.D. thesis, Univ. of Rochester, Rochester, N.Y., 1972). Previously, we established three classes of bacteriophage-resistant mutants ofB. subtilis that lacked glucosylated cell wall teichoic acid due to their inability to synthesize UDP-glucose:polyglycerol teichoic acid glucosyltransferase or phosphoglucomutase (24, 25). Bacteria lacking phosphoglucomutase are defective in the gtaC gene, whereas those bacteria that lack UDP-glucose:polyglycerol teichoic acid glucosyltransferase have lost a
1122
YASBIN, MAINO, AND YOUNG
J. BACTERIOL.
functional gtaA gene. Bacteria that have both 4)105 was totally unable to infectgtaA mutants. of these enzymes but still lack glucosylated tei- In all remaining cases, the bacteriophages choic acid are deficient in the gtaB gene prod- coul,d infect the non-glucosylated cell wall teiuct. The class for each of the isolated mutants choic acid mutants with reduced efficiency. was determined, and representatives of each Bacteriophages SPOl, SPP1, and 41 appeared class are shown in Table 2. to be least dependent on cell wall-glucosylated Previously our laboratory demonstrated that teichoic acid for infection. It is important to some bacteriophages differed in their ability to note that all of these infections occurred in infect non-glucosylated teichoic acid cell wall semisolid agar. mutants in semisolid agar and in liquid broth Similarly, the ability of these nine bacterio(24). To further investigate this phenomenon, phages to attach to these bacteria in liquid representatives of each bacteriophage-resistant media was determined (Table 4). Although bacclass were infected with nine B. subtilis bacte- teriophage SPOl was unable to infect gtaA muriophages. There were major differences in the tants, it appeared to attach to these bacteria in ability of bacteriophages to infect these mu- liquid media. On the other hand, bacteriophage tants (Table 3). Bacteriophages SPOl and 429 SPP1 infected gtaA, gtaB, and gtaC mutants, were completely dependent on glucosylated cell although in liquid media it did not attach to wall teichoic acid for attachment. Additionally, these same bacteria at wild-type levels. Addibacteriophage 425 was unable to successfully tionally, bacteriophage 4e infected gtaB muinfect gtaB and gtaC mutants, although it tants in agar, but in liquid it attached poorly to could infect gtaA mutants with reduced effi- these cells. These conflicting results will be ciency. Bacteriophages 4e and SP82 were also discussed later. However, it is important to unable to infect gtaC mutants. Bacteriophage note that, generally, the lack of a glucosylated SP82 was severely inhibited in its ability to teichoic acid in the cell wall results in a deinfect cells carrying gtaB, and bacteriophage crease in the ability of bacteriophages to attach. These data confirm the importance of gluTABLE 2. Enzyme activities cosylated teichoic acid as a major bacteriophage receptor. Enzyme activity (nmol/min/mg of protein) TABLE 4. Efficiency of adsorption Tag Strain Genotype Efficiency of adsorption (%P BacterioPGMa GM UDPGBR151 RUB808 RUB810 RUB807 phages PPaseb transferase BR151
gta+
68.5
3.6
(x 10-3)
studied
(gta+)
(gtaA)
(gtaB)
(gtaC)
3.08
SPOl
99 98 94 91 100 100
92 65 21 8 50 100 13 3 74
12 37 18 4 42 8 26 8 20
12 36 18 7 21 9
SPO2 3.8 0.95 RUB807 gtaC20
JOURNAL OF BACTERIOLOGY, Mar. 1976, p. 1120-1126 Copyright C) 1976 American Society for Microbiology
Bacteriophage Resistance in Bacillus subtilis 168, W23, and Interstrain Transformants RONALD E. YASBIN,1 * VERNON C. MAINO,2 AND FRANK E. YOUNG Laboratory ofBiochemical Genetics, National Heart and Lung Institute, Bethesda, Maryland 20014, and Department of Microbiology, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
Received for publication 14 October 1975
Strains of Bacillus subtilis 168 deficient in glucosylated teichoic acid vary in their resistance to bacteriophage infection. Although glucosylated teichoic acid is important for bacteriophage attachment, the results demonstrate that alternate receptor sites exist. Non-glucosylated cell wall mutants could be assigned to specific classes (gtaA, gtaB, gtaC) by their pattern of resistance to three closely related bacteriophages (G25, Oe, SP82). In addition to glucosylation, the type of teichoic acid was also important for bacteriophage attachment. B. subtilis strains 168 and W23 have different teichoic acids in their cell walls and have varied susceptibilities to bacteriophage infection. Transfer of bacteriophage resistance from strain W23 into a derivative of strain 168 was accomplished. The resistant bacteria obtained were impaired in their ability to adsorb bacteriophage and in their capacity to be transfected by bacteriophage deoxyribonucleic acid. The importance of teichoic acids as a bacte- phage resistance is strongly affected by the nariophage receptor site in the wall of the bacte- ture of the genetic defect. Glucosylated teichoic acid has also been rium Bacillus subtilis has been demonstrated (2, 8, 24). Specifically, three classes of bacterio- shown to play an essential role in the attachphage-resistant mutants of B. subtilis strain ment of some of the Bacillus bacteriophages to 168 have been shown to lack glucosylated tei- B. subtilis strain W23 (8, 24). B. subtilis strains choic acid in their cell walls (24, 25). The gluco- W23 and 168 differ in the content of their teisylation of cell wall teichoic acids is regulated choic acids (5, 7) and their susceptibility to by the products of the gtaA +, gtaB +, and gtaC + certain bacteriophages (3, 16; B. E. Reilly, genes (24). These three genes have been Ph.D. thesis, Western Reserve Univ., Clevemapped (25), and the proteins coded by two of land, Ohio, 1965). Furthermore, B. subtilis the genes have been characterized (4, 12). Phos- strain W23 is lysogenic for defective bacteriophoglucomutase (EC 2.7.5.1) is the product of phage PBSZ, whereas strain 168 is lysogenic for the gtaC+ gene, whereas uridine 5'-diphos- defective bacteriophage PBSX (14, 18). To exphate (UDP)-glucose:polyglycerol teichoic acid plore the effect of variability of the cell surface glucosyltransferase (EC 1.4.1) is a product of on viral infection, we transferred bacteriophage the gtaA + gene. Although no enzyme has been resistance from strain W23 to strain 168 via identified for the gtaB + gene, it has been sug- deoxyribonucleic acid (DNA)-mediated transgested that this gene codes for an inactive phos- formation. The results of this transfer as well phoglucomutase monomer (13). In the initial as the patterns of bacteriophage resistance in studies, when only a limited number of bacteri- strain W23 indicate the variety and complexity ophages were examined, it appeared that the of mechanisms by which bacteria interfere with bacterial mutants unable to glucosylate tei- bacteriophage development. choic acid were resistant to certain bacterioMATERIALS AND METHODS phages regardless of whether the mutation was Strains and methods of propagation. The bacteat the gtaA, gtaB, or gtaC locus (24). In this report, we have expanded these studies and can rial strains used in this study are listed in Table 1. now demonstrate that the degree of bacterio- The propagation of bacteriophage and the maintePresent address: Department of Biology, Brookhaven National Laboratory, Upton, N.Y. 11973. 2 Present address: Division of Allergy and Clinical Immunology, National Jewish Hospital and Research Center, Denver, Colo. 80206. '
nance of the bacterial cultures were described previously (21, 22). Bacteriophage SP82 was obtained from M. Green. Media and procedures for genetic exchange. Modified M agar and broth were prepared as previously described (22). The procedures for transforma-
120
BACTERIOPHAGE RESISTANCE
VOL. 125, 1976
1121
TABLE 1. Bacterial strains used Strain
BR151 RUB807
Genotypea
lys-3, trpC2, metBlO lys-3, trpC2, metB10, gtaC20
Origin and remarks Competent derivative of B. subtilis 168 Phage-resistant marker transformed into BR151 using donor DNA from a spontaneous )29-resistant strain obtained by V. C. Maino
(168/qb29) Phage-resistant marker transformed into BR151 using donor DNA from a spontaneous +105-resistant strain obtained by S. Zahler (CU548) RUB809 lys-3, trpC2, metB10, gtaA21 Isolated similarly to RUB807 Isolated similarly to RUB807 RUB810 Iys-3, metBlO, gtaB20 Isolated similarly to RUB807 RUB811 lys-3, trpC2, metB10, gtaC21 Isolated similarly to RUB807 RUB812 lys-3, trpC2, metB10, gtaB21 A spontaneous erythromycin-resistant derivaery-1 RUB815 tive of B. subtilis W23 A prototrophic derivative of BR141 RUB818b Lysogenic for bacteriophage SP02 RUB818(SP02) Lysogenic for bacteriophage O105 RUB818(105) Isolated by resistance to' bacteriophages SPP1 RUB821 lys-3, trpC2, metB10, 4Rc and SP02cl2 after transformation of strain BR151 with DNA obtained from strain RUB815 Isolated as was RUB821 RUB822 lys-3, trpC2, metB10, OR RUB823 Isolated as was RUB821 lys-3, trpC2, metB10, OR RUB824 Isolated as was RUB821 lys-3, trpC2, metB10, O)R a Symbols as defined in reference 26 and references 12, 24, 25, and V. C. Maino (Ph.D. thesis), for gta strains. b Obtained after three separate transformations with nonsaturating concentrations of 168 T+ DNA. c Phenotypically phage resistant.
RUB808
lys-3, trpC2, metB10, gtaA20
tion, transfection, and isolation of DNA were identical to those previously used in our laboratory (22, 23). Determination of plating efficiencies. Bacterial cultures were grown at 37 C in modified M broth that had not been supplemented with Ca2 , Mg2+, or Mn2+. After the cells had ceased exponential growth (optical density of 120 Klett units on a Klett-Summerson colorimeter, filter no. 66), 0.2 ml of the bacterial culture was added to 2.0 ml of modified M semisolid agar containing 0.1 ml of the appropriate dilution of the bacteriophage stock, and the mixture was poured onto modified M agar. This procedure was used for all bacteriophage studied except 429. The plating efficiency of bacteriophage 429 was determined by using a modified M overlay on tryptose blood agar base (Difco). Assays for bacteriophage 429 were done at 30 C, whereas all other quantitation of infectious centers were performed at 37 C. Determination of adsorption efficiencies. Bacterial cultures were grown in modified M broth without Mg2+, Ca2+, and Mn2+ at 37 C to an optical density of 120 Klett units, centrifuged (12,000 x g for 20 min at 4 C), and suspended in Spizizen's minimal salts (17) containing 3.7% formaldehyde (Fisher). The suspension was kept at 4 C for 30 min before the cells were washed three times with modified M broth and finally resuspended in half the original volume in modified M broth. The bacteriophage and the formaldehyde-treated bacteria were mixed to yield a multiplicity of infection of 10-4 to 10-s, incu-
bated at 37 C for 20 min, and centrifuged (8,000 x g for 5 min at room temperature), and the number of plaque-forming units in the broth was determined. Enzyme assays. UDP-glucose:pyrophosphorylase (EC 2.7.7.9), UDP-glucose:polyglycerol teichoic acid glucosyltransferase, and phosphoglucomutase were measured as previously specified (4, 12).
RESULTS Isolation and identification of bacteriophage-resistant mutants of B. subtilis strain 168. DNA was isolated from spontaneously arising bacteriophage-resistant mutants of B. subtilis, and the resistance trait was transformed into strain BR151. These bacteriophageresistant mutants were found to lack glucosylated teichoic acid in their cell walls (V. C. Maino, Ph.D. thesis, Univ. of Rochester, Rochester, N.Y., 1972). Previously, we established three classes of bacteriophage-resistant mutants ofB. subtilis that lacked glucosylated cell wall teichoic acid due to their inability to synthesize UDP-glucose:polyglycerol teichoic acid glucosyltransferase or phosphoglucomutase (24, 25). Bacteria lacking phosphoglucomutase are defective in the gtaC gene, whereas those bacteria that lack UDP-glucose:polyglycerol teichoic acid glucosyltransferase have lost a
1122
YASBIN, MAINO, AND YOUNG
J. BACTERIOL.
functional gtaA gene. Bacteria that have both 4)105 was totally unable to infectgtaA mutants. of these enzymes but still lack glucosylated tei- In all remaining cases, the bacteriophages choic acid are deficient in the gtaB gene prod- coul,d infect the non-glucosylated cell wall teiuct. The class for each of the isolated mutants choic acid mutants with reduced efficiency. was determined, and representatives of each Bacteriophages SPOl, SPP1, and 41 appeared class are shown in Table 2. to be least dependent on cell wall-glucosylated Previously our laboratory demonstrated that teichoic acid for infection. It is important to some bacteriophages differed in their ability to note that all of these infections occurred in infect non-glucosylated teichoic acid cell wall semisolid agar. mutants in semisolid agar and in liquid broth Similarly, the ability of these nine bacterio(24). To further investigate this phenomenon, phages to attach to these bacteria in liquid representatives of each bacteriophage-resistant media was determined (Table 4). Although bacclass were infected with nine B. subtilis bacte- teriophage SPOl was unable to infect gtaA muriophages. There were major differences in the tants, it appeared to attach to these bacteria in ability of bacteriophages to infect these mu- liquid media. On the other hand, bacteriophage tants (Table 3). Bacteriophages SPOl and 429 SPP1 infected gtaA, gtaB, and gtaC mutants, were completely dependent on glucosylated cell although in liquid media it did not attach to wall teichoic acid for attachment. Additionally, these same bacteria at wild-type levels. Addibacteriophage 425 was unable to successfully tionally, bacteriophage 4e infected gtaB muinfect gtaB and gtaC mutants, although it tants in agar, but in liquid it attached poorly to could infect gtaA mutants with reduced effi- these cells. These conflicting results will be ciency. Bacteriophages 4e and SP82 were also discussed later. However, it is important to unable to infect gtaC mutants. Bacteriophage note that, generally, the lack of a glucosylated SP82 was severely inhibited in its ability to teichoic acid in the cell wall results in a deinfect cells carrying gtaB, and bacteriophage crease in the ability of bacteriophages to attach. These data confirm the importance of gluTABLE 2. Enzyme activities cosylated teichoic acid as a major bacteriophage receptor. Enzyme activity (nmol/min/mg of protein) TABLE 4. Efficiency of adsorption Tag Strain Genotype Efficiency of adsorption (%P BacterioPGMa GM UDPGBR151 RUB808 RUB810 RUB807 phages PPaseb transferase BR151
gta+
68.5
3.6
(x 10-3)
studied
(gta+)
(gtaA)
(gtaB)
(gtaC)
3.08
SPOl
99 98 94 91 100 100
92 65 21 8 50 100 13 3 74
12 37 18 4 42 8 26 8 20
12 36 18 7 21 9
SPO2 3.8 0.95 RUB807 gtaC20
