A leucine repeat motif in AbiA is required for resistance of Lactococcus lactis to phages representing three species.

Gene 212 (1998) 5–11

A leucine repeat motif in AbiA is required for resistance of Lactococcus lactis to phages representing three species. Polly K. Dinsmore a, Daniel J. O’Sullivan a, Todd R. Klaenhammer a,b,* a Department of Food Science, Southeast Dairy Foods Research Center, North Carolina State University, Box 7624, Raleigh, NC 27695, USA b Department of Microbiology, North Carolina State University, Box 7615, Raleigh, NC 27695, USA Received 12 September 1997; received in revised form 4 February 1998; accepted 5 February 1998; Received by A.M. Campbell

Abstract The abiA gene encodes an abortive bacteriophage infection mechanism that can protect Lactococcus species from infection by a variety of bacteriophages including three unrelated phage species. Five heptad leucine repeats suggestive of a leucine zipper motif were identified between residues 232 and 266 in the predicted amino acid sequence of the AbiA protein. The biological role of residues in the repeats was investigated by incorporating amino acid substitutions via site-directed mutagenesis. Each mutant was tested for phage resistance against three phages, w31, sk1, and c2, belonging to species P335, 936, and c2, respectively. The five residues that comprise the heptad repeats were designated L234, L242, A249, L256, and L263. Three single conservative mutations of leucine to valine in positions L235, L242, and L263 and a double mutation of two leucines (L235 and L242) to valines did not affect AbiA activity on any phages tested. Non-conservative single substitutions of charged amino acids for three of the leucines (L235, L242, and L256) virtually eliminated AbiA activity on all phages tested. Substitution of the alanine residue in the third repeat (A249) with a charged residue did not affect AbiA activity. Replacement of L242 with an alanine elimination phage resistance against w31, but partial resistance to sk1 and c2 remained. Two single proline substitutions for leucines L242 and L263 virtually eliminated AbiA activity against all phages, indicating that the predicted alpha-helical structure of this region is important. Mutations in an adjacent region of basic amino acids had various effects on phage resistance, suggesting that these basic residues are also important for AbiA activity. This directed mutagenesis analysis of AbiA indicated that the leucine repeat structure is essential for conferring phage resistance against three species of lactococcal bacteriophages. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Lactic acid bacteria; Phage resistance; Leucine zipper; Dairy starter cultures

1. Introduction Lactococcus lactis subsp. lactis strains are used frequently as starter cultures for dairy fermentations. Upon repeated use of starter strains, it is common for bacteriophages virulent against these strains to appear in the fermentation environment. Phage infection of starter cultures is a major cause of failed or slow fermentation in the dairy industry. Therefore, bacteriophage resis-

* Corresponding author. Tel: +1 919 515 2971; Fax: +1 919 515 7124; e-mail: klaenhammer@ncsu,edu Abbreviations: A, alanine; Abi, abortive bacteriophage infection; D, aspartic acid; E, glutamic acid; EOP, efficiency of plaquing; H, histidine; K, lysine; L, leucine; P, proline; PCR, polymerase chain reaction; R, arginine; V, valine. 0378-1119/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 9 8 ) 0 0 13 2 - 2

tance is a highly sought and researched trait in these bacteria. Abortive bacteriophage infection (Abi) mechanisms are widespread in lactococci and can be very effective in protecting cultures from phage infection. Although many Abi mechanisms have been identified, only recently have efforts begun to address the molecular basis of how these mechanisms interfere with phage proliferation. The first report on this topic suggested that AbiD1 retarded phage development by limiting an essential phage product (Bidnenko et al., 1995). Also, it has been reported that AbiB promotes degradation of phage transcripts after 10–15 min of infection (Parreira et al., 1996). A thorough understanding of the modes of action of the Abi mechanisms will be extremely useful in designing phage-resistant Lactococcus strains in the future.

6

P.K. Dinsmore et al. / Gene 212 (1998) 5–11

The abiA gene encodes an Abi mechanism (AbiA) that provides resistance to a variety of lactococcal phages (Hill et al., 1989; Dinsmore and Klaenhammer, 1994). Phage DNA replication is not observed in strains harboring AbiA (Hill et al., 1991), and therefore, AbiA inhibits phages early in their infection cycle. The DNA sequence of the abiA gene (1887-bp) was reported by Hill et al. (1990) from a native L. lactis plasmid, pTR203, and independently by Coffey et al. (1991) from a different L. lactis plasmid (GenBank Accession No. M30192) A leucine repeat structure was identified within the deduced AbiA amino acid sequence (Fig. 1). This leucine repeat structure resembles the leucine zipper motif that consists of heptad repeats of leucine residues over a span of 20–30 amino acids (Landschulz et al., 1988). When the residues in each repeat are designated abcdefg, the leucine residue is in position d (Hu et al., 1990), and a hydrophobic residue is in position a, forming a 4–3 hydrophobic repeat (O’Shea et al., 1989). Leucine zipper domains are alpha-helical in structure (O’Shea et al., 1989), and therefore, the repeats in positions a and d form a hydrophobic surface along one side of the helix. The interaction of the hydrophobic surfaces of two helices to form a coiled coil can mediate dimerization of proteins containing these motifs (O’Shea et al., 1989), resulting in homo- or heterodimers. Leucine zippers can be associated with a region rich in basic amino acids that promotes DNA binding by the protein dimer (Gentz et al., 1989; Landschulz et al., 1989). In this study, we investigated the biological role of the leucine repeat structure, and an adjacent basic region, in AbiA using site-directed mutagenesis.

Fig. 1. Diagram and restriction enzyme map of the abiA gene after introduction of the PvuI and SstI sites. The line above the diagram of abiA indicates the number of base pairs along the coding sequence of the gene. Box with diagonal lines, heptad leucine repeats. Shaded box, basic region. Asterisks indicate that these sites were introduced by sitedirected mutagenesis. Small arrows depict primers used for sequencing all of the plasmid clones containing mutated PvuI/SstI cassettes. The primers correspond to nucleotides 581–599 and to the complement of nucleotides 1021–1043 of the AbiA coding sequence. Amino acids 232–281 of the AbiA protein are shown indicating the five heptad repeats and the region of basic amino acids. Each repeat is numbered, and the abcdefg designation is shown for the first repeat. Residues in positions a and d are underlined.

2. Materials and methods 2.1. Propagation of bacteria and bacteriophages The bacterial strains, bacteriophages, and plasmids used in this study are listed in Table 1. L. lactis strains were propagated at 30°C in M17 in broth ( Terzaghi and Sandine, 1975) supplemented with 0.5% glucose (GM17). Erythromycin (1.5 mg/ml ) was added when necessary for plasmid maintenance. Lactococcal bacteriophages were propagated and titrated by the method of Terzaghi and Sandine (1975). Efficiencies of plaquing ( EOP) were calculated by dividing the number of plaque-forming units per milliliter (pfu/ml ) observed with the experimental strain by the pfu/ml observed with the phage-sensitive strain. All EOPs reported are the average of at least three independent experiments. E. coli JM110 was used as transformation hosts for cloning and plasmid isolation. E. coli transformants were propagated in LB medium (Sambrook et al., 1989) containing 100 mg/ml of erythromycin. The phage M13 clones used to generate single-stranded DNA for sitedirected mutagenesis were propagated using E. coli strain TG1 or JM110. 2.2. Bacterial transformation L. lactis strains were grown in GM17 containing 1% glycine and prepared for electroporation by the method of Holo and Nes (1989) except that the sucrose was omitted from the growth and resuspension media. After electroporation and non-selective growth for 2 h, cells were plated on GM17 agar plates containing erythromycin (1.5 mg/ml ). RbCl-competent E. coli were prepared by the following procedure: 100 ml of cells in mid-log phase were centrifuged and resuspended in 30 ml of transformation buffer I [30 mM potassium acetate, 50 mM MnCl , 100 mM RbCl, 10 mM CaCl , and 15% 2 2 (wt/vol ) glycerol; pH 5.8]. Cells were incubated on ice in transformation buffer I for 2 h, then centrifuged and resuspended in 4 ml of transformation buffer II [10 mM NaMOPS (pH 7.0), 75 mM CaCl , 10 mM RbCl, 15% 2 glycerol ]. Cells were frozen at −70°C in aliquots. For transformation, the procedure was the same as that described for CaCl -competent cells (Sambrook et al., 2 1989). Transformants were selected on BHI agar plates containing erythromycin (100 mg/ml ). 2.3. DNA isolation and cloning Rapid plasmid DNA isolation from E. coli was performed by alkaline lysis (Sambrook et al., 1989). DNA was purified from agarose gels with the Qiaex II Gel Extraction Kit (Qiagen, Chatsworth, CA). Singlestranded DNA was isolated from phage M13 clones as described by Sambrook et al. (1989). Plasmid DNA was

7

P.K. Dinsmore et al. / Gene 212 (1998) 5–11 Table 1 Bacterial strains, bacteriophages, and plasmids Strain, phage, plasmid Bacterial strains Lactococcus lactis NCK203 MG1363 E. coli JM110 TG1 Bacteriophages w31 sk1 c2 M13amp18 M13/AbiA(XbaI ) M13/AbiA(XbaI+PS) Plasmids pTRKH2 pTRK376 pTRK376.APS

Description

References

Phage w31-sensitive host Phages sk1- and c2-sensitive host

Hill et al. (1989) Gasson (1983)

Transformation host and host for propagation of M13 clones Host for propagation of M13 clones

Yanisch-Perron et al. (1985) Sambrook et al. (1989)

Small isometric phage, species P335 Small isometric phage, species 936 Prolate phage, species c2 Cloning vector M13amp18 containing the XbaI fragment from the abiA gene with the leucine repeat domain M13/AbiA(XbaI ) mutated to contain a PvuI and an SstI site flanking the leucine repeat domain

Alatossava and Klaenhammer (1991) Powell et al. (1989) Sanders and Klaenhammer (1983) Yanisch-Perron et al. (1985) This study

Cloning vector, high copy number in Lactococcus, erythromycin resistance pTRKH2 containing the subcloned abiA gene pTRK376 derivative in which a PvuI site and an SstI site have been introduced, by silent mutation, into the abiA gene flanking the putative leucine zipper domain

isolated from L. lactis by the method of O’Sullivan and Klaenhammer (1993a). General procedures for DNA manipulations were as previously described (Sambrook et al., 1989). Restriction endonucleases, T4 DNA ligase, and their corresponding buffers were purchased from Boehringer Mannheim Biochemicals (Indianapolis, IN ) with the exception of SstI, which was purchased from GIBCOBRL (Gaithersburg, MD). 2.4. Polymerase chain reaction The abiA structural gene and promoter were amplified from pTK6 (Hill et al., 1989) using 30 cycles of the following steps: a 15-s denaturation at 94°C, a 15-s annealing step at 50°C, and a 2-min extension at 68°C. To facilitate cloning of the PCR product, the upstream primer incorporated a BamHI site (5∞-GAGGGATCCGCAGGGAGCTGTATCTGC-3∞), and the downstream primer incorporated a PstI site (5∞-AGACTGCAGGCGGATGAGGAGATAACAGG-3∞). Taq polymerase and other PCR reagents were purchased from Boehringer Mannheim Biochemicals. 2.5. DNA sequencing Double-stranded plasmid DNA and single-stranded DNA were sequenced by the dideoxy chain termination method with the Sequenase 2.0 enzyme ( United States Biochemical Corp.).

This study

O’Sullivan and Klaenhammer (1993b) This study This study

2.6. Site-directed mutagenesis Site-directed mutations in the abiA gene were constructed using one of the following two methods. Mutations to introduce the PvuI and SstI sites were made using the Sculptor in-vitro Mutagenesis System (Amersham Life Science, Arlington Heights, IL) and the following mutagenic oligonucleotides: PvuI (5∞-CCTAAAAACGATCGTTGAAATCG-3∞) and SstI (5∞-CCCACAAGTGAGCTCAGCACTA-3∞). Singlestranded DNA from phage M13/AbiA (XbaI ) was used as the template for mutagenesis. The resulting phage clone containing the PvuI and SstI sites was designated M13/Abi (XbaI+PS ). Mutations L235V, L242V, L242P, L242R, L242A, and A249R were also generated using the Sculptor kit. The [L235V, L242V ] double mutant was made by sequential mutagenesis with the L235V and L242V oligonucleotides. Single-stranded DNA from the phage M13/Abi (XbaI+PS) was used as the template for mutagenesis. The resulting phage DNAs were screened for the desired mutation by sequencing. The wild-type 183-bp PvuI/SstI fragment in pTRK376.APS (Fig. 1) was replaced with the mutated PvuI/SstI fragment from each of the phage clones. The resulting plasmid contained the abiA gene with the desired mutation. The plasmid clones were confirmed by sequencing from nucleotide 630 to 980 of the abiA coding sequence with the primers illustrated in Fig. 1. Mutations L235R, L242R, [ K270A, K271A],

8

P.K. Dinsmore et al. / Gene 212 (1998) 5–11

[ K270E, K271D, K272D], and [ K270R, K271R, K272R] were constructed using the three-step PCR procedure of Picard et al. (1994) except that Taq polymerase was substituted for Pfu polymerase. Plasmid pTRK376.APS was used as the template for mutagenesis. The upstream and downstream primers required for this method flank the PvuI/SstI cassette and are illustrated in Fig. 1. The final PCR product was cut with PvuI and SstI and cloned into the pTRK376.APS cut with the same enzymes resulting in replacement of the wild-type PvuI/SstI fragment with the mutated fragment. DNA sequencing of the resulting plasmid from nucleotides 630 to 980 of the abiA coding sequence confirmed the presence of the desired mutation in each clone.

3. Results 3.1. The AbiA deduced amino acid sequence contains a heptad repeat sequence similar to the leucine zipper motif Examination of the AbiA deduced amino acid sequence (Hill et al., 1990; Coffey et al., 1991) revealed five 4–3 hydrophobic repeats characteristic of the leucine zipper motif. This motif appears from amino acid 232 to 266 in the AbiA protein ( Fig. 1). Four out of five residues in position d are leucines, and four out of five residues in position a are hydrophobic. Residues 232–266 of the AbiA protein are predicted to form a alpha-helix. The predicted helix is amphipathic with the hydrophobic residues lining up on one face of the helix and many charged residues in the other positions on the helix. In several DNA binding proteins, a region of basic amino acids is located N-terminal to a leucine zipper motif and is essential for the DNA binding property of the protein (Gentz et al., 1989; Landschulz et al., 1989). In the AbiA protein, a short region of basic amino acids (six basic out of 15 amino acids) lies immediately C-terminal to the putative leucine zipper ( Fig. 1). Although this region varies in position and length from those studied previously, its role in AbiA activity was also investigated by site-directed mutagenesis.

the conditions described in Materials and Methods. The PCR primers incorporated a BamHI site upstream of AbiA and PstI site downstream. The 2.2-kb PCR product was cloned into BamHI and PstI sites of pTRKH2 (O’Sullivan and Klaenhammer, 1993b). The resulting plasmid, pTRK376, conferred resistance to w31 at an EOP of 5.1×10−4. These phenotypic results are comparable to those with pTRK363, a high copy plasmid containing AbiA and 4.8-kb of flanking pTR2030 DNA (Dinsmore and Klaenhammer, 1994). 3.3. Construction of a PvuI/SstI cassette containing the heptad leucine repeats To facilitate sequencing of each mutagenesis substrate, a small cassette containing the leucine zipper region was constructed, which could be removed and mutated and then reinserted into the abiA gene. The wild-type abiA sequence does not contain any unique restriction enzyme sites flanking the putative leucine zipper. Therefore, two restriction enzyme sites were introduced into abiA via site-directed mutagenesis. Inspection of the abiA nucleotide sequence indicated that a PvuI (upstream) and an SstI site (downstream) could be introduced flanking the region without altering the amino sequence of the encoded protein. No PvuI or SstI sites existed in the wild-type abiA sequence. Before introduction of the PvuI and SstI sites, the PvuI and SstI sites were removed from the vector portion of pTRK376 by deleting most of the polylinker on each side of the abiA insert. The resulting plasmid was designated pTRK376D. To introduce the PvuI and SstI sites, the XbaI fragment containing the leucine repeats (Fig. 1A) was cloned into M13mp18. Oligonucleotide-directed mutagenesis was performed as described in Materials and Methods. The XbaI fragment was cloned from the M13 clone into pTRK376D, which had been partially digested with XbaI. The proper orientation of the inserted fragment was confirmed by restriction enzyme analysis and DNA sequencing. The final plasmid was designated pTRK376.APS and confers the same level of resistance to phage w31 as pTRK376 ( EOP with pTRK376.APS was 4.5×10−4). Plasmid pTRK376.APS was used to create all of the subsequent abiA mutations.

3.2. PCR and cloning of the abiA gene Previously, the AbiA phenotype had been studied on plasmids containing the abiA gene flanked by other sequences derived from pTR2030, the lactococcal plasmid from which abiA was originally cloned (Hill et al., 1989; Dinsmore and Klaenhammer, 1994). The downstream flanking region has not been sequenced, and the restriction enzyme sites have not been mapped. To facilitate site-directed mutagenesis of the abiA gene, it was necessary to subclone this gene into an appropriate plasmid vector. PCR was used to amplify abiA using

3.4. Site-directed mutagenesis of the heptad leucine repeat domain A mutation analysis was conducted on residues in position d of the hydrophobic repeats to define the requirements in this position for biological activity of AbiA. The residues in position d of the repeats are referred to as L235, L242, L249, L256, and L263. Conservative changes to valine residues were made to determine whether another hydrophobic residue can be substituted for leucine. Residues L235, L242, and L263

9

P.K. Dinsmore et al. / Gene 212 (1998) 5–11

were each individually changed to valine. A double mutant with both L235 and L242 changed to valines was also made. None of these conservative mutations significantly altered the phage resistance phenotype conferred by AbiA against phages w31, sk1, or c2 ( Table 2), indicating that another hydrophobic amino acid could replace leucine at positions L235, L242, and L263. Conservative substitutions were not made for A249 and L256. Non-conservative changes to polar or neutral residues were made to determine whether hydrophobicity is required in the d position. Individual changes of L235, L242, or L256 to an arginine residue and L242 to aspartic acid resulted in the loss of the AbiA phage resistance phenotype (Table 2). However, changing A249 to arginine had no significant effect on the AbiA phenotype. Substitution of L242 with alanine decreased the hydrophobicity of the repeats, virtually eliminated AbiA activity on w31 and significantly reduced resistance to phages sk1 and c2 ( Table 2). These substitutions indicated that hydrophobicity was required at three of the four positions tested. Leucine was also substituted with proline at positions L242 and L263 to disrupt the predicted alpha-helix. Both of these mutations virtually eliminated the phage resistance conferred by AbiA, indicating that the alphahelical structure is necessary for phage resistance. The effects of the leucine zipper mutations on the AbiA phage resistance phenotype described in Table 2 suggested that both the hydrophobicity of the heptad leucine residues and the predicted alpha-helical structure were required for AbiA activity against phage w31.

3.5. Site-directed mutagenesis of the basic region adjacent to the heptad leucine repeats The role of the short basic region adjacent to the heptad leucine repeat domain in AbiA activity was also investigated by mutagenesis. The three basic residues, K270, H271, H272, were the targets for alterations. Changing all three residues to negatively charged amino acids ( K270E, H271D, H272D mutations) virtually eliminated AbiA activity against w31 and significantly reduced resistance to phages sk1 and c2 ( Table 3). Changing both K270 and H271 to alanine ( K270A, H271A) decreased the number of positively charged residues in the region from six to four but did not alter phage resistance. Mutation of all three basic residues to another basic residue, arginine, ( K270R, H271R, H272R mutation), decreased the level of phage resistance conferred to AbiA but did not completely eliminate it. These data indicated that the basic residues studied are important in AbiA function, but not simply because of their positive charge.

4. Discussion This is the first report of a functional leucine repeat structure in any lactic acid bacterium. The data presented demonstrate that a leucine repeat structure in the AbiA amino acid sequence is critical for the general phage resistance phenotype conferred by this protein. Single conservative mutations that preserve the hydrophobic character of the repeats did not alter the level of phage resistance against the three phages tested. Also,

Table 2 Efficiencies of plaquing ( EOP) of phages w31, sk1, and c2 on strains containing mutations in the leucine repeat motif of abiA abiA construction

Residue in ABiA amino acid sequence:

EOP(w31)

EOP (sk1)

EOP (c2)

235

242

249

256

263

Absence of abiA Wild-type abiA L235V L242V L235V,L242V L263V

L V L V L

L L V V L

A A A A A

L L L L L

L L L L V

1.0 5.1×10−4 6.8×10−4 2.4×10−4 5.0×10−4 2.1×10−4

1.0 <1.6×10−8 NTa <1.6×10−8 <1.6×10−8 <1.6×10−8

1.0 5.3×10−6 NT 4.4×10−6 6.1×10−6 7.3×10−6

L235R L242A L242D L242R L249R L256R

R L L L L L

L A D R L L

A A A A R A

L L L L L R

L L L L L L

0.6 0.8 0.8 1.0 3.0×10−4 0.8

1.0 0.35 NT 0.8 <1.6×−8 1.0

1.0 0.16 NT 1.0 4.5×10−6 1.0

L242P L263P

L L

P L

A A

L L

L P

0.6 1.0

aNT, not tested.

1.0 0.9

0.9 1.0

10

P.K. Dinsmore et al. / Gene 212 (1998) 5–11

Table 3 Efficiencies of plaquing ( EOP) of phages w31,sk1, c2 on strains containing mutations in the basic region of abiA abiA construction

Wild-type abiA K270A, H271A K270E, H271D, H272D K270R, H271R, H272R

Residue in AbiA amino acid sequence 270

271

272

K A E R

H A D R

H H D R

a mutant with two conservative substitutions had the wild-type level of phage resistance. A reduction or loss of AbiA phage resistance was associated with four individual charged amino acid substitutions, except in the case of mutation A249R. A substitution of alanine for leucine (L242A) resulted in only slight resistance against the three phages tested. Since the AbiA leucine repeat motif already has an alanine at position A249, a second alanine in the preceding repeat (L24A) may decrease the hydrophobicity of the non-polar face of the helix enough to partially inactivate the protein. In reported cases of a leucine zipper residue substituted with a proline, protein dimerization and biological activity were disrupted due to the distortion of the alphahelical structure of the region (Gentz et al., 1989; Hu et al., 1990; Turner and Tjian, 1989). AbiA activity against all three phages was virtually eliminated with the two proline substitution mutations tested. Overall, mutations that disrupt the hydrophobicity of the leucines or the predicted alpha-helical nature of the domain reduced or eliminated inhibition of phages by AbiA. These data are consistent with previous mutational studies of leucine repeat structures (Gentz et al., 1989; Landschulz et al., 1989; Turner and Tjian, 1989; Hu et al., 1990). The short region of basic amino acids immediately following the leucine repeat domain was also studied because such basic domains are often associated with leucine zippers in DNA binding proteins. Substitution experiments within this region gave varying results, including no change in phage resistance ( K270A, H271A), partial resistance ( K270R, H271R, H272R), and complete loss of phage resistance ( K270E, H271D, H272D). Further mutagenesis is required to fully understand the role of the basic amino acids in AbiA activity. It is clear from the data presented that the leucine repeat domain in AbiA is essential for phage resistance. We can speculate that the role of the leucine repeats in AbiA activity could be in the formation of homodimers that are required for its function or to form heterodimers with a phage protein. A gene (ORF245) encoding an early phage w31 protein has recently been implicated in the sensitivity of the phage to AbiA (Dinsmore and Klaenhammer, 1997). This protein may be a candidate for heterodimer formation with the AbiA protein. A

EOP (w31)

EOP (sk1)

EOP (c2)

5.1×10−4 3.1×10−4 0.6 2.2×10−2

<1.6×10−8 <1.6×10−8 0.2 8.0×10−4

5.3×10−6 8.1×10−6 9.7×10−2 1.7×10−2

detailed study of the AbiA protein is necessary to determine whether or not this domain mediates dimer formation. AbiA is a large protein (73.8 kDa) that is active against a variety of phages, but different phages vary in their degree of susceptibility to AbiA. The phages sk1 (species 936) and c2 (species c2) are more susceptible to AbiA than w31 (species P335). (Dinsmore and Klaenhammer, 1994; Table 2). Three of the mutations in the present study affected the three phages tested differently (L242A; K270E; H271D, H272D; K270R, H271R, H272R). It is unclear as to whether AbiA has different modes of action on different phages or whether it acts in the same manner to different degrees. Further studies of this region may clarify these differences and reveal what makes phages more or less sensitive to AbiA. Future studies of this region may also reveal ways to improve the AbiA activity on various phages. For example, the protein may be more effective if the alanine present in the d position of the third repeat was changed to a leucine. Changes that make the putative alpha-helix or coiled-coil structures more stable might also improve activity.

Acknowledgement This research was supported by the North Carolina Agricultural Research Service under project NCO2168 and in part by Marschall Products, Rhoˆne-Poulenc. The authors thank Gordana Djordjevic, Evelyn Durmaz, John McCormick, and Shirley Walker for helpful discussions and critical review of this work.

References Alatossava, T., Klaenhammer, T.R., 1991. Molecular characterization of three small isometric-headed bacteriophages which vary in their sensitivity to the lactococcal phage resistance plasmid pTR2030. Appl. Environ. Microbiol. 57, 1346–1353. Bidnenko, E., Ehrlich, D., Chopin, M.C., 1995. Phage operon involved in sensitivity to the Lactococcus lactis abortive infection mechanism AbiD1. J. Bacteriol. 177, 3824–3829. Coffey, A.G., Fitzgerald, G.F., Daly, C., 1991. Cloning and characterization of the determinant for abortive infection of bacteriophage

P.K. Dinsmore et al. / Gene 212 (1998) 5–11 from lactococcal plasmid pCI829. J. Gen. Microbiol. 137, 1355–1362. Disnmore, P.K., Klaenhammer, T.R., 1994. Phenotypic consequences of altering the copy number of abiA, a gene responsible for aborting bacteriophage infections in Lactococcus lactis. Appl. Environ. Microbiol. 60, 1129–1136. Dinsmore, P.K., Klaenhammer, T.R., 1997. Molecular characterization of a genomic region in a Lactococcus bacteriophage that is involved in its sensitivity to the phage defense mechanism AbiA. J. Bacteriol. 179, 2949–2957. Gasson, M.J., 1983. Plasmid complements of Streptococcus lactis NCDO712 and other lactic streptococci after protoplast-induced curing. J. Bacteriol. 154, 1–9. Gentz, R., Rauscher, F.J., Abate, C., Curran, T., 1989. Parallel association of Fos and Jun leucine zippers juxtaposes DNA binding domains. Science 243, 1695–1699. Hill, C., Pierce, K., Klaenhammer, T.R., 1989. The conjugative plasmid pTR2030 encodes two bacteriophage defense mechanisms, restriction/modification (R+/M+) and abortive infection (Hsp+). Appl. Environ. Microbiol. 55, 2416–2419. Hill, C., Miller, L.A., Klaenhammer, T.R., 1990. Nucleotide sequence and distribution of the pTR2030 resistance determinant (hsp) which aborts bacteriophage infection in lactococci. Appl. Environ. Microbiol. 56, 2255–2258. Hill, C., Massey, K.J., Klaenhammer, T.R., 1991. Rapid method to characterize lactococcal bacteriophage genomes. Appl. Environ. Microbiol. 57, 283–288. Holo, H., Nes, I.F., 1989. High-frequency transformation, by electroporation, of Lactococcus lactis subsp. cremoris grown with glycine in osmotically stabilized media. Appl. Environ. Microbiol. 55, 3119–3123. Hu, J.C., O’Shea, E.K., Kim, P.S., Sauer, R.T., 1990. Sequence requirements for coiled-coils: Analysis with l repressor-GCN4 leucine zipper fusions. Science 250, 1400–1403. Landschulz, W.H., Johnson, P.F., McKnight, S.L., 1988. The leucine zipper: A hypothetical class of DNA binding proteins. Science 240, 1759–1764.

11

Landschulz, W., Johnson, P.F., McKnight, S.L., 1989. The DNA binding domain of the rat liver nuclear protein C/EBP is bipartite. Science 243, 1681–1688. O’Shea, E., Rutkowski, R., Kim, P., 1989. Evidence that the leucine zipper is a coiled-coil. Science 243, 538–542. O’Sullivan, D.J., Klaenhammer, T.R., 1993a. Rapid mini-prep isolation of high-quality plasmid DNA from Lactococcus and Lactobacillus spp. Appl. Environ. Microbiol. 59, 2730–2733. O’Sullivan, D.J., Klaenhammer, T.R., 1993b. High and low copy number Lactococcus shuttle cloning vectors with features for clone screening. Gene 137, 227–231. Parreira, R., Ehrlich, S.D., Chopin, M.C., 1996. Dramatic decay of phage transcripts in lactococcal cells carrying the abortive infection determinant AbiB. Mol. Microbiol. 19, 221–230. Picard, V., Ersdal-Badju, E., Lu, A., Bock, S.C., 1994. A rapid and efficient one-tube PCR-based mutagenesis technique using Pfu DNA polymerase. Nucleic Acids Res. 22, 2587–2591. Powell, I.B., Arnold, P.M., Hillier, A.J., Davidson, B.E., 1989. Molecular comparison of prolate- and isometric-headed bacteriophages of lactococci. Can. J. Microbiol. 35, 860–866. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Sanders, M.E., Klaenhammer, T.R., 1983. Characterization of phagesensitive mutants from a phage-insensitive strain of Streptococcus lactis: evidence for a plasmid determinant that prevents phage adsorption. Appl. Environ. Microbiol. 46, 1125–1133. Terzaghi, B.E., Sandine, W.E., 1975. Improved medium for lactic streptococci and their bacteriophages. Appl. Environ. Microbiol. 29, 807–813. Turner, R., Tjian, R., 1989. Leucine repeats and an adjacent DNA binding domain mediate the formation of functional cRos–cJun heterodimers. Science 243, 1689–1694. Yanisch-Perron, C., Viera, J., Messing, J., 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.