OUP user menu

Complete nucleotide sequence and molecular characterization of two lytic Staphylococcus aureus phages: 44AHJD and P68

Dietmar Vybiral, Marian Takáč, Martin Loessner, Angela Witte, Uwe von Ahsen, Udo Bläsi
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00028-4 275-283 First published online: 1 February 2003


The first complete nucleotide sequences of two lytic Staphylococcus aureus double stranded DNA phages, 44AHJD (16 784 bp) and P68 (18 227 bp), are reported. Both are small isometric phages, with short, non-contractile tails and a pre-neck appendage. Based on their morphology, their genome size, the similarity of the encoded gene products, the type of infection and on the possession of a type B DNA polymerase, 44AHJD and P68 are allocated to the order Caudovirales, family Podoviridae, genus ‘φ29-like phages’. The genome of 44AHJD differs from that of P68 by a deletion spanning nucleotides 10 091 to 11 531 of the P68 genome. The electrophoretic analysis of the terminal DNA fragments of P68 DNA and P68 DNA protein complex suggested the presence of a terminal protein at either DNA end. In contrast to the lysis cassette of the φ29-like phages, which is located at the end of the late operon, the lysis cassette of 44AHJD and P68 is located within the structural genes.

  • Lytic phage
  • Nucleotide sequence
  • Staphylococcus aureus

1 Introduction

Staphylococcus aureus causes a variety of clinical symptoms including skin lesions, serious infections like pneumonia and meningitis, urinary tract infections, and deep-seated infections such as endocarditis. Moreover, S. aureus is a major cause of hospital-acquired wound infections, infections associated with indwelling medical devices, and is the leading cause of food-borne intoxications due to the production of a heat-stable enterotoxin. The occurrence of multidrug-resistant S. aureus strains has prompted research efforts towards alternative therapeutic strategies such as phage therapy. In the pre-antibiotic age, phages were extensively used as antibacterial agents but their use was mainly hampered by inadequate scientific measures [1]. Given that several phages of various pathogens are known to carry virulence determinants [2], the knowledge of the genome sequence and the encoded information are important for the safe use of therapeutic phage. Up to now, the genomes of three temperate S. aureus phages have been sequenced: φPVL, carrying the Panton–Valentine leukocidin genes [3], φETA [4] and φSLT [5]. Here, we report the first complete DNA sequences of two lytic S. aureus phages, 44AHJD and P68, analyzed their genome at the molecular level and determined the evolutionary relatedness to other lytic dsDNA phages of Gram-positive bacteria.

2 Materials and methods

2.1 Biological materials, isolation of phage particles and phage DNA preparations

Both phages, 44AHJD and P68, were kindly provided by Dr. H. Ackermann, Felix d'Herelle Reference Center for Bacterial Viruses, Quebec, Canada, and propagated on the corresponding S. aureus host strains. S. aureus strains were grown in brain heart broth (Merck) at 37°C with vigorous shaking to an OD600 of 0.3–0.4, when the corresponding phage was added at a multiplicity of infection of 3. After 2 h, the lysate was centrifuged for 15 min at 15 000×g at 4°C to remove cellular debris. Phage particles were isolated and phage DNAs as well as P68 DNA protein complex were prepared exactly as described for Bacillus subtilis phage φ29 [6].

2.2 Sequencing of the genomes of 44AHJD and P68

DNA sequencing was performed by MWG Biotech AG, Ebersberg, Germany. In brief, the DNA was fractionated by hydrodynamic shearing, shotgun cloned into the pGEM-T vector (Promega). The inserts were sequenced according to the ABI standard protocol (BigDye-terminator version 3), with an ABI3700 capillary electrophoresis sequencer. The assembly was done using PHRED/PHRAP [7] as well as the STADEN [8] packages.

2.3 DNA sequence analysis and bioinformatics

Open reading frame (ORF) identification was performed using the GeneRunner 3.0 program package (Hastings software, Inc., 1994) and the GeneMark.hmm program [9]. Based on the codon usage for S. aureus, ORFs preceded by a Shine and Dalgarno (SD) sequence with an appropriate distance (3–18 bp) to the initiation codon, and coding for a polypeptide of more than 50 amino acids were considered as putative genes. Similarity searches for nucleotide sequences and for the deduced amino acid sequences were performed using the FASTA [10], BLAST [11], PARALIGN [12], PSI-BLAST [13], and SSEARCH [10] programs available on the Online Analysis Tools home page (http://molbiol-tools.ca/), with sequences present in the EMBL/GenBank/DDBJ nucleotide sequence database and the SWISSPROT and PIR protein sequence databases. Protein motifs from the PROSITE pattern library [14] and the PROSITE profile library [14] as well as protein domains from the ProDom database [15] and the Pfam database [16,17] were searched with the MOTIF program (GenomeNet, Institute for Chemical Research, Kyoto University, Japan). Multiple sequence alignments were performed using the program CLUSTAL W 1.8 [18], available on the BCM Search Launcher (http://searchlauncher.bcm.tmc.edu), and the results were displayed using BOXSHADE (http://www.ch.embnet.org/software/BOX_form.html).

2.4 Electron microscopy

P68 and 44AHJD were negatively stained with 2% uranyl acetate and examined using a Zeiss transmission electron microscope EM902 at an accelerating voltage of 80 kV.

3 Results and discussion

3.1 Morphology of 44AHJD and P68

Electron microscopy of 44AHJD and P68 revealed small isometric headed phages, 75 nm in diameter with a short, non-flexible, non-contractile tail of ~40 nm in length for P68 and ~27 nm in length for 44AHJD (Fig. 1a and b). The phages were allocated to the order Caudovirales wherein they fall into the family of Podoviridae according to the classification of Ackermann [19]. A pre-neck appendage was found for both phages, which is characteristic for ‘φ29-like phages’. Previously, the International Committee on Taxonomy of Viruses classified S. aureus phage 44AHJD into the genus of ‘φ29-like phages’ (species number: These morphological data and the genotypical data presented below corroborate this classification.

Figure 1

Electron micrographs of the phages 44AHJD (a) and P68 (b) negatively stained with 2% uranyl acetate. The bar represents 75 nm.

3.2 Genomes of 44AHJD and P68

The complete DNA sequence of both strands of the phages was determined. The genome of 44AHJD consists of 16 784 bp (GenBank accession no. AF513032) and the genome of P68 comprises 18 227 bp (GenBank accession no. AF513033). The total G+C content is 29.6 and 29.3% for 44AHJD and P68, respectively. Identical inverted terminal repeats with a length of 210 bp were found in both phage genomes at either DNA end. The overall DNA sequence identity between 44AHJD and P68 is 92.2%. Phage 44AHJD differs from phage P68 by a deletion of 1440 bp, comprising nt 10 091–11 531 with regard to the P68 sequence coordinates. 22 open reading frames (ORFs) were identified within the genome of P68 (Table 1; Fig. 2), with 12 orientated in opposite direction from the others. 44AHJD contains 21 putative ORFs of which 11 are orientated in opposite direction (Table 1; Fig. 2). Except for ORF14, which is absent in 44AHJD, the predicted gene products of P68 and 44AHJD are identical. All ORFs begin with an ATG except ORF8, which starts with TTG.

View this table:
Table 1

Location of ORFs in P68 and 44AHJD, translational starts and putative functions of the predicted polypeptides

ORFFrameStartEndPutative translation initiation sitesNo. of amino acidsSize (kDa)PiPutative function
4+11 1111 479TAAATAAGAGGAGAAATAAA ATG12214.35.1single stranded DNA binding protein
6+11 7082 118TTAAGGAGATATAAAAATG ATG13616.54.8
8+12 2842 766GCTTTATATGGAGGTTGATA TTG16019.48.6
9+32 9254 061CAAATAGAATTAGTTGATGA ATG37845.96.0encapsidation protein
10+34 0956 362AAGATTATGGGATTACTAGA ATG75589.75.5DNA-polymerase
11−17 9166 477ATGATACTGAAAAGAGTGAT ATG47952.59.2
12−38 3137 891TAGAGAGGGGGTATAATATG ATG14016.37.9holin
13−210 0788 315CTATTTTTATTGGAGGAAAAA ATG58768.57.0tail protein
14−211 58110 136CAAATAAGAGGTGTAAACAA ATG48154.75.0minor tail protein
15−111 89711 619CCAATACGGCACATATTATA ATG9210.58.9holin
16−312 39611 644ACATCAAAAATAGGAGTGAT ATG25028.58.6amidase
17−114 35112 408TGGTAGAGGTGGTTAAATAA ATG64774.85.9minor structural protein
18−315 12014 365AGATGAAAGCAGTGATATAA ATG27529.15.2lower collar protein
19−215 86215 113TTAATGTAGTTGTTGGTGAA ATG24928.54.4upper collar protein
20−317 33716 111ACGTTGAGGAGGAATAATAA ATG40846.95.8major head protein
21−317 52617 344ATTTAGATTAGGAGGAATTT ATG726.94.4
22−217 94117 540AATATTTGGAGGTGTCTAAA ATG13315.23.5
  • All nt positions are given according to the P68 coordinates.

  • The putative SD sequences and the start codons are indicated in bold.

  • Calculated isoelectric point.

  • This ORF is absent in 44AHJD.

Figure 2

Depiction of the genome organization of 44AHJD and P68. Filled black arrows depict proteins encoded by the respective ORFs for which a putative function was predicted. Open arrows depict ORFs for which no function was predicted. The protein encoded by ORF15 is indicated by a light gray arrow indicating that this protein was identified based on the secondary structure of its amino acid sequence and on the location of the ORF within the amidase gene (ORF16). The protein encoded by ORF14 is depicted in dark gray indicating that this ORF is only present in P68. LITR and RITR denote the inverted terminal repeats at the 5′- and 3′-ends of the genome. The putative transcriptional terminators (T) are indicated by open circles and the positions of the putative promoters (PR, promoter right; PL, promoter left) are indicated by arrows pointing in the direction of transcription. Cleavage sites for the restriction enzymes Spe I and Xcm I (see Fig. 3) are shown.

3.3 Presence of a terminal protein at either DNA end of P68

In the φ29-like Bacillus phages, the Escherichia coli and Streptococcus pneumoniae phages PRD1 and CP-1, respectively, linear plasmids and the eukaryotic adenovirus the inverted terminal repeats are involved in initiation of DNA replication by providing a binding site for the terminal protein (TP) [20]. Although the product encoded by ORF11 of 44AHJD and P68 shows some traits of a TP in that it contains a predicted coiled coil domain typical for the TP of the φ29-like phages and conserved amino acid residues [20] which could interact with the DNA polymerase, database searches did not reveal a significant homology of gp11 with known TPs (data not shown).

Since the terminal ends of φ29 DNA containing the TP do not migrate during gel electrophoresis [6], the presence of a TP can be assessed by comparing the electrophoretic patterns of 3′- and 5′-end fragments of phage DNA prepared in the presence and absence of protease. Digestion of protease-treated P68 DNA with Spe I (cleaves at position 17 562) generated the 3′-end 665 bp fragment (Fig. 3, lane 3) whereas cleavage with Xcm I (cleaves at positions 779, 11 708 and 13 088) generated the 779 bp 5′-end fragment and a 1380 bp internal fragment (Fig. 3, lane 6). In contrast, when the P68 DNA protein complex was digested with either enzyme neither the 3′-end (Fig. 3, lane 2) nor the 5′-end fragment (Fig. 3, lane 5) was detected. However, both terminal fragments re-appeared when the P68 DNA protein complex digested with either enzyme was subsequently treated with protease K. Therefore, these experiments suggested the presence of a TP in P68 DNA, and, as the DNA ends of both phages are identical, by inference also in 44AHJD DNA.

Figure 3

Evidence for the presence of a terminal protein in P68 DNA. P68 DNA and P68 DNA protein complex were cleaved with the restriction enzyme Spe I or Xcm I, and the corresponding fragments were resolved on a 4% polyacrylamide gel. Lanes 3 and 6, P68 DNA was cleaved with Spe I and Xcm I, respectively. Lanes 2 and 5, P68 DNA protein complex was cleaved with Spe I and Xcm I, respectively. Lanes 1 and 4, P68 DNA protein complex was first digested for 2 h with Spe I or Xcm I. Then, protease K (200 µg ml−1) was added and the incubation was continued for 30 min at 37°C. The 5′- and 3′-end fragments as well as an internal fragment generated by both enzymes (see text) are marked by arrows. The position of DNA markers is shown on the right.

3.4 Promoters and terminators

Five putative promoter sequences were identified by searching the genomes of both phages for a S. aureusσ70-dependent promoter consensus motif containing the Pribnow box consensus sequence (TATDHT) followed by a spacer of 16–18 nucleotides and a consensus −35 region (WTNAND). In addition, three putative promoter sequences were found, when the DNA sequences of both phages were inspected for the Pribnow box signature GGGTAT, which is found in S. aureusσB-dependent promoters [21]. Promoters PR1, PR2, PR3, PL1, and PL5 (Table 2) have a consensus sequence for σ70-dependent S. aureus promoters. The PL5 promoter matches exactly the promoter sequence of the σ70-dependent promoter P1 which is involved in regulation of the sar operon of S. aureus [22]. The PL2, PL3, and PL4 (Table 2) promoters were identified by the presence of the σB Pribnow box signature. The locations of the different promoters are depicted in Fig. 2 and the corresponding nucleotide positions are given in Table 2. The promoters PR1, PR2, PL1, PL4 and PL5 contain the TG motif positioned 1 bp upstream of the −10 box (Table 2). The motif is frequently found in σA-dependent B. subtilis promoters, where they provide contact sites for the B. subtilisσA RNA polymerase [23]. Regions forming stem-loop structures, probably representing transcriptional terminators (see Fig. 2) are located: (1) downstream of ORF4 between nt 1471 and 1530 (T1), and (2) in the intergenic region between ORF10 and ORF11 spanning nt positions 6331–6390 (T2 and T2L (lower strand)). T2 has a uridine-rich tail at either strand following the stem-loop suggesting that T2 constitutes a bidirectional Rho-independent transcriptional terminator.

View this table:
Table 2

Putative promoter sequences of P68 and 44AHJD


3.5 Other putative regulatory signals

It has been suggested that the TAAATTAA motif is involved in down-regulation of the P1 promoter activity of the sar locus of S. aureus by providing a binding site for a regulatory protein [22]. A manual search for this motif within the genomes of both phages revealed that it occurs five times. The distribution of the motif within the genomes of 44ADHJ and P68 is depicted in Fig. 2.

3.6 Analysis of gene products involved in DNA replication and packaging

3.6.1 DNA replication: ORF4 and ORF10

The protein encoded by ORF4 showed 47% similarity in a 107 amino acid overlap to the Lactobacillus delbrueckii bacteriophage LL-H ssDNA binding protein, which is involved in phage DNA replication [24]. ORF10 protein showed similarities to DNA polymerases of the members of the φ29-like family of phages with 36% similarity to the DNA polymerases of both the B. subtilis phages M2 [25] and φPZA [26], 35% sequence similarity to the B. subtilis phages φ29 [27] and φB103 [28]. A multiple alignment of the protein encoded by ORF10 with the protein sequences of the DNA polymerases of φ29 (group 1 of the φ29-like genus of Bacillus phages), φB103 (group 2) and M2 (group 2) is shown in Fig. 4. All conserved motifs, characteristic for B-type proofreading-proficient DNA polymerases, such as the Exo I, Exo II, and Exo III motif as well as the motifs of the C-terminal domain, motif 1 (also called A), motif 2a (also called motif B) and 2b, motif 3 (also called C) and motif 4, as well as the CT (cross-talk) motif were also found in the putative DNA polymerase of 44AHJD and P68.

Figure 4

Alignment of the DNA polymerase encoded by Bacillus phages M2 (accession no. P19894), B103 (accession no. Q37882) and φ29 (accession no. X53370) with the putative DNA polymerase of P68 and 44AHJD encoded by ORF10. Conserved residues are boxed in black, conservative amino acid changes are shown in gray. Conserved motifs (see text) of the type B proofreading-proficient DNA polymerases are indicated above the sequence.

3.6.2 DNA packaging: ORF9

The deduced protein sequence of ORF9 displayed 40% similarity to the ATPase protein p16 of the Bacillus phages φPZA [29] and φ29 [30] as well as a high similarity to the Bacillus phage φGA-1 encapsidation protein [20]. These proteins are considered to be required for ATP hydrolysis, which generates the energy for DNA encapsidation. The φ29 p16 sequence contains two typical ATP binding sites [31]: a Walker A and a Walker B motif are located in the N-terminal (amino acids 24–39) and in the C-terminal (amino acids 248–256) part of the φ29 p16 protein, respectively. Both motifs are conserved in the p16 proteins encoded by B103 and GA-1 [20]. The Walker A motif is also present in protein 9 of 44AHJD and P68 but no Walker B motif was found (not shown).

3.7 Capsid proteins

3.7.1 Collar proteins: ORF18 and ORF19

The protein encoded by ORF18 showed 37% similarity in an 163 amino acid overlap to the lower collar protein of the Streptococcus phage CP-1 [32] and 36% similarity in a 204 amino acid overlap to the Bacillus phage B103 lower collar protein [28]. The deduced protein encoded by ORF19 showed high sequence similarity to the upper collar (connector) proteins of the φ29 family of phages. In detail, 46% similarity was found to the upper collar protein from the Bacillus phage GA-1 (165 amino acid overlap; [20]), φPZA (160 amino acid overlap; [29]) and φ29 (160 amino acid overlap; [33]). Slightly lower similarity (45%) was found for the upper collar protein of the Bacillus phage B103 within a 160 amino acid overlap [28].

3.7.2 Head protein: ORF20

The product of ORF20 showed 32% similarity to the major head protein of the Bacillus phage φ29 [34] within an overlap of 252 amino acids.

3.7.3 ORF17

The protein encoded by ORF17 showed 45% similarity within a 487 amino acid overlap to the product of ORF56 of the temperate S. aureus phage φETA [4] as well as to ORF636 of the S. aureus temperate phage φSLT [5] within a 558 amino acid overlap. These two proteins are similar to the minor structural protein gp89 of L. delbrueckii bacteriophage LL-H [24].

3.7.4 Tail proteins: ORF13 and ORF14

The predicted product of ORF13 was identified as the tail protein of 44AHJD and P68 based on the overall similarity to tail proteins of the φ29 family of phages. In particular, 39% similarity was found to the entire tail protein of φPZA and 37% similarity was found to the major tail proteins from the Bacillus phages B103 and φ29.

Comparison of the deduced amino acid sequence of ORF14 with known protein sequences in databases revealed 46% similarity in an 251 amino acid overlap to the N-terminal part of a tail protein of the temperate S. aureus phage φSLT (ORF1374; [5]). Furthermore, 40% similarity was found to gp14 of the Enterobacteria phage HK022, which is similar to gpJ of phage λ, and thought to be involved in host range specificity [35]. The host range of both phages 44AHDJ and P68 was determined on 35 clinical S. aureus isolates and shown to differ (not shown). Since both phages are identical at the molecular level except for ORF14, the encoded protein seems to play a role in host range specificity.

3.8 Lysis proteins

3.8.1 ORF16

The predicted product of ORF16 of 44AHJD and P68 was identified as an amidase. The highest similarity was found to the N-acetylmuramoyl-l-alanine amidase of the S. aureus phage Twort [36]. The N-terminal 149 amino acids showed 68% similarity while the C-terminal part of the φ44AHJD and φP68 amidase showed 44% similarity to the amidase of phage Twort.

3.8.2 ORF12 and ORF15

ORF15 is preceded by a weak SD sequence and codes for a putative protein of 106 amino acids. Although no similarities were found to known protein sequences in databases, the overlap of ORF15 with ORF16 (Fig. 2) resembles the holin-endolysin entity in S. aureus phage φ187, where the holin gene is fully embedded in the reading frame of the endolysin gene [37]. Gp15 is predicted to contain two transmembrane helices, a typical trait of holin proteins [38]. In addition, a dual start motif (5′-ATG-AAA-ATG-…-3′) is present in ORF15, which in other holin genes directs the synthesis of two products, involved in scheduling of host cell lysis [38].

Another candidate holin is encoded by ORF12. The 104 amino acid protein showed a high sequence similarity to the holin proteins of the φ29 family of phages. The predicted protein 12 has 64 and 44% similarity to the holin proteins of the Bacillus phage φ105 (accession no. AB016282) and the Streptococcus phage Cp-1 [32], respectively. Two transmembrane helices are predicted for protein 12. In contrast to the lysis cassette of φCP-1 and the φ29-like phages, which is located downstream of the gene cluster for structural proteins, in 44AHJD and P68 the lysis cassette is located within the structural genes. The location of the putative regulatory element (TAAATTAA) within the 5′ region of ORF16 (amidase) as well as the occurrence of the σB-dependent promoter PL4 upstream of ORF16 could further indicate that the lysis genes are transcribed separately from the structural protein encoding genes.


We are grateful to Dr. H. Ackermann for providing the phages and to T. Narzt for the electron micrographs. This work was supported by InterCell Biomedical Research and Development AG, Vienna, Austria.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
View Abstract