OUP user menu

Isolation and characterization of the Serratia entomophila antifeeding prophage

Mark R.H. Hurst, Sam S. Beard, Trevor A. Jackson, Sandra M. Jones
DOI: http://dx.doi.org/10.1111/j.1574-6968.2007.00645.x 42-48 First published online: 1 May 2007


The Serratia entomophila antifeeding prophage (Afp) is thought to form a virus-like structure that has activity towards the New Zealand grass grub, Costelytra zealandica. Through the trans based expression of AnfA1, an RfaH – like transcriptional antiterminator, the Afp, was able to be induced. The expressed Afp was purified and visualized by electron microscopy. The Afp resembled a phage tail-like bacteriocin, exhibiting two distinct morphologies: an extended and a contracted form. The purified Afp conferred rapid activity towards C. zealandica larvae, causing cessation of feeding and a change to an amber colouration within 48 h postinoculation, with increased dose rates causing larval mortality.

  • antifeeding
  • afp
  • RfaH
  • Serratia entomophila
  • Costelytra zealandica
  • PVC


Serratia entomophila and Serratia proteamaculans (Enterobacteriaceae) are the causal agents of amber disease of the New Zealand grass grub, Costelytra zealandica (Coleoptera: Scarabaeidae). The disease was first described by Trought (1982) and is highly host specific, only affecting larvae of a single indigenous species of New Zealand scarab. Infected hosts cease feeding within 1–3 days of ingesting pathogenic bacteria. The larvae gut, which is normally dark in colour, clears (Jackson et al., 1993) and levels of the major gut digestive enzymes (trypsin and chymotrypsin) decrease sharply (Jackson, 1995). The larvae may remain in this state for 1–3 months, before bacteria eventually invade the haemocoel, resulting in rapid death of the insect (Jackson et al., 1993, 2001).

Two regions of the 153-kb plasmid termed pADAP (amber disease-associated plasmid) have been identified as necessary for causing amber disease symptoms. The sep virulence-associated region comprises of three genes sepABC that are responsible for the amber disease symptoms of gut clearance and amber colouration of the larvae (Hurst et al., 2000). The afpantifeeding prophage gene cluster, which encodes a putative virus-like structure theorized to mediate the transport of toxins to a target site, causes a cessation of feeding by the grass grub larvae (Hurst et al., 2004). The afp comprises 18 ORFs, of which the translated products of six scored highly to the tail sheath and phage base plate-related protein domains of the bacteriophages T4 and P2. Another ORF encodes a protein that has a similarity to components of avian adenovirus tail fibres (Fig. 1). Orthologues of the first 16 ORFs of the afp cluster reside as six, now termed Photorhabdusvirulence cassettes (PVC's) in the genome of the insecticidal bacterium Photorhabdus luminescens TTO1 (Duchaud et al., 2003; Hurst et al., 2004; Waterfield et al., 2004). Work by Yang (2006) has identified additional PVCs in the genome of Photorhabdus asymbiotica, and has shown by electron microscopy that a PVC cassette (PaPVCpnf) is able to form a distinct R-type pyocin-like structure. R-type pyocins exhibit two forms: a contracted form, where the outer sheath resembles a bullet-like structure, and an extended form, where the inner core of the contracted form protrudes. The R-type pyocins absorb to a lipopolysaccharide of a target bacterial cell and cause lethality by the rapid contraction of the sheath component and subsequent penetration of the core through the outer bacterial membrane (Michel-Briand & Baysse, 2002).

Figure 1

Schematic of the afp showing ORF-associated protein domains, the location of the amb2 locus (anfA1 and anfA2) and its associated ops element; modified from (Hurst et al., 2004).

DNA analysis of the afp encoding region suggests that the regulation of the afp gene cluster is mediated by AnfA1, a component of the amb2 locus (Nunez-Valdez & Mahanty, 1996). The translated product of anfA1 has a high similarity to members of the RfaH family of transcriptional antiterminator (Hurst et al., 2004). Members of the RfaH family are required for the transcription of genes encoding the synthesis of the F pilus and lipopolysaccharide core attachment of the O-antigen of Escherichia coli and Salmonella (Bailey et al., 1992), and toxins such as α-haemolysin (Leeds & Welch, 1996), among others. They act by binding to a specific DNA sequence (5′-GGCGGTAGNNT-3′) termed the ops (operon polarity suppressor) element, located upstream or internal to the operon they are associated with, and enable the readthrough of terminator structures, thus preventing operon polarity (Bailey et al., 1997, 2000). An ops element with the DNA sequence (5′-GGCGGTAGCAT-3′) is located 108bp upstream of afp1 (gi:48995178 bp 74780–74792; Fig. 1), indicating that AnfA1 may regulate the transcription of the afp cluster. Work by Carter (2004) demonstrated that overexpression of RfaH-type regulators results in the concurrent expression of their associated operons.

In this study, the production of the Afp by the in trans-based arabinose expression of the anfA1 gene is induced.

Materials and methods

Strains and plasmids

The strains and plasmids used in the study are listed in Table 1. Bacteria were grown in Luria–Bertani (LB) broth or on LB agar (Sambrook et al., 1989), at 37°C for E. coli and 30°C for S. entomophila. For E. coli, carbinicillin, chloramphenicol and spectinomycin were used at concentrations of 100, 30 and 100 µg mL−1, respectively. l-arabinose was used to a final concentration of 0.2%.

View this table:
Table 1

Bacterial strains and plasmids used in the study

Strain or plasmidRelevant genotype or descriptionReference
Escherichia coli
DH10BFmcrmrr-hsdRMS-mcrBCΦ80d lacZΔM15ΔlacX74 endA1 recA1 deoara, leu 7697 araD139 galU galK nupG, rpsLorow & Jessee (1990)
Serratia entomophila
A1MO2pADAP pathogenicGrimont et al. (1988)
5.6Heat cured pADAP minus derivative, of A1MO2Glare et al. (1993)
Pseudomonas fluorescensAgResearch Insect Pathogen culture collection (AgResearch, Lincoln, New Zealand)
pACARApACYC184 containing araC gene cloned as a HindIII-ClaI fragment from pAY2-4; CmRThis study
pACYC184Cloning vector; CmR, TcRChang & Cohen (1978)
pADAPAmber disease associated plasmid (153 kb)Glare et al. (1993)
pADK93XbaIΔStuIpADAP derivative from which a 45-kb fragment encompassing the sep-virulence associated region was removed by deletion (121.2 kb); antifeeding pathotype; SpRHurst et al. (2004)
pAF6Antifeeding prophage encoding region in pBR322; ApRHurst et al. (2004)
pAF6-21pAF6 containing mini-Tn10 insertion in sea25, a probable transposase gene; ApR, CmRHurst et al. (2004)
pASANFA1Arabinose inducible anfA1, SpRThis study
pASARApACARA containing MfeI flanked spectinomycin resistance gene in EcoRI site; SpRThis study
pAY2-4Arabinose expression vector, ApRShaw et al. (2003)
pBR322Cloning vector; ApR, TcRBolivar et al. (1977)
pHP45Contains spectinomycin resistant Ω fragment ApR, SpRPrentki & Krisch (1984)

DNA manipulation and sequencing

Standard DNA techniques were carried out as described by Sambrook (1989). Plasmid templates for DNA sequencing were prepared using the High Pure Plasmid Isolation Kit (Roche Diagnostics GmbH). The DNA was sequenced using a capillary ABI3730 Genetic Analyzer, from Applied Biosystems Inc. (http://awcmee.massey.ac.nz/genome-service.htm).

PCR was undertaken using Thermoprime Plus DNA polymerase (ABgene; Advanced Biotechnologies Ltd, UK), 1.5 mM MgCl2, 0.2 mM each dNTP, 2 µM each primer and 1 µL of DNA in a final volume of 25 µL, adjusted with sterile-distilled water. Template DNA was denatured with a preliminary step of 97°C for 2 min, and then five PCR cycles of denaturing at 95°C for 15 s, annealing at 55°C for 15 s and elongating at 72°C for 90 s, followed by 30 cycles at 95°C for 15 s, 50°C for 15 s and 72°C for 90 s. The primer sets used were spF (5′-AAACAATTGCAAACCCTCACTGATCCGC-3′)-spR (5′-AAACAATTGGCCGCGCCGCGAAGCGGC-3′); anfA1F (5′-AAACATATGAAAAAAAAATGGTATTTGATTAG-3′) — anfA1R (5′-AAAGAATTCGACGCAATTAAAACCCCTGAAT-3′); araF (5′-TCCATAAGATTAGCGGATCCTAC-3′); and araR 5′-CATGGGGTCAGGTGGGAC-3′). PCR products were purified using the High Pure PCR Product Purification Kit (Roche Diagnostics GmbH) following the manufacturer's instructions.

Induction and purification of the Afp

From a 3 mL overnight culture of E. coli strain DH10B (Table 1), 1.0 mL of bacteria was inoculated into 50 mL of LB broth and grown to an OD600 nm of 0.6. The cells were harvested by centrifugation at 8000 g for 10 min, resuspended in 1.0 mL of 0.5 × LB broth and transferred to a fresh 50 mL broth of 0.5 × LB, supplemented with the appropriate antibiotics and l-arabinose and left at 20°C (0.02 g) for 5 h.

The culture was harvested by centrifugation at 8000 g for 3 min and resuspended in 1.2 mL phosphate-buffered saline (10 mM sodium phosphate buffer, pH 7.4; 2.7 mM KCl; 137 mM NaCl). Two 0.7 mL samples were transferred to a 1.7 mL microcentrifuge tube and subjected to three 20-sec rounds of sonication on wet ice using a Sanyo Soniprep 150 Sonicater (18 Ω). The sonicated samples were centrifuged at 16 000 g for 3 min and filter-sterilized through a 0.2 µm Sartorius Minisart® filter to a sterile microcentrifuge tube.

For chloroform-based isolation, 600 µL of chloroform and NaCl to a final concentration of 0.5 M was added and the sample was vortexed for 20 s. The culture was left to stand at ambient temperature for 10 min. Bacterial debris was removed by centrifugation at 10 000 g for 30 min at 4°C. A 4-mL volume of the supernatant was centrifuged at 250 000 g for 90 min at 4°C in a Sorvall RZ M120EX Ultra Centrifuge. The pellet was resuspended in 100 µL 0.5 × LB broth and treated for 10 min at ambient temperature with Pancreatic DNaseI and RNaseA at final concentrations of 5 and 1 µg mL−1, respectively. The sample was further centrifuged at 12 000 g for 10 min at 4°C, and the supernatant was applied to the surface of 3.5 mL of 40% glycerol in 0.5 × LB broth and centrifuged at 250 000 g for 90 min at 4°C. The centrifuge tubes were inverted on dry tissue paper for 10 min. The pellet was gently resuspended in 30 µL of 0.5 × LB broth.

Afp temperature and bacterocin sensitivity assay

The effect of temperature on the stability of the Afp was assessed by exposing 500 µL postchloroform-treated samples to 45, 50, 60 and 65°C for 10 min. The heated samples and an untreated control were assessed for their ability to cause antifeeding by standard bioassay. To assess the ability of the Afp to inhibit the growth of bacteria, 10 µL aliquots of the postchloroform-treated-induced Afp was pipetted onto the surface of a freshly prepared soft-agar lawn, containing cells of the bacterium to be tested. The aliquots were allowed to dry on the agar surface and then incubated overnight at the temperature conducive to the growth of the target bacterium.

Transmission electron microscopy

To visualize the Afp, 3 µL of the Afp preparation obtained after ultracentrifugation was applied to a formvar-coated 300-mesh copper grid, the sample was left for 5 min and excess fluid was drawn off with an absorbent filter paper. A drop of 2% phosphotungstic acid was then added to each grid and the excess fluid was drawn off with an absorbent filter paper and the grids were air-dried. Grids were examined with a Hitachi H-600 electron microscope (80 kV) at a magnification of × 150 000.

Activity of isolated Afp against C. zealandica larvae

The efficacy of the isolated Afp was assessed by the application of 5 µL of sample to the surface of a 3 mm3 carrot cube, from which the grass grub larvae would feed. Twelve second- or third-instar larvae collected from the field were used for each treatment. Inoculated larvae were maintained at 15°C in ice-cube trays. Larvae were fed treated carrot on day 1, and on days 3 and 6 were transferred to fresh trays containing untreated carrot. The occurrence of cessation of feeding and amber colouration were monitored at days 1, 2, 3, 6 and 12. Controls tested were the S. entomophila strains A1MO2; A1MO2 (pADK93XbaIΔStuI) and sonicated filtrates of arabinose-induced E. coli pAF6-21 (Table 1). The sterility of the sonicated filtrates and chloroformed-treated samples was tested by plating 100 µL onto LB agar and assessing for growth. Bioassays were performed at least four times per treatment.


Construction of the trans-acting anfA1 arabinose expression vector

To construct an arabinose-inducible vector compatible with the pBR322-based antifeeding clone pAF6-21 (Table 1), the 1686 bp ClaI-HindIII DNA fragment encoding the araC gene from the vector pAY2-4 was excised and ligated into the analogous restriction enzyme site of pACYC184 (Table 1), forming the vector pACARA. To facilitate later antibiotic selection, an MfeI-flanked spectinomycin resistance cassette was amplified using pHP45 DNA and the spF-spR primer set (see ‘Materials and methods’). The 1437-bp amplicon was cloned into the EcoRI site of the chloramphenicol resistance gene of pACARA to form pASARA. The anfA1 gene was PCR-amplified using the primers anfA1F and anfA1R and the 540-bp amplicon cloned into the unique NdeI-EcoRI sites of pASARA, to form the construct pASANFA1. The resultant clone was validated by DNA sequencing using the araF and araR, pASARA-specific primers. Once validated, pASANFA1 was electroporated into the E. coli strain DH10B containing pAF6-21.

Induction and assessment of Afp efficacy

The E. coli strains DH10B (pAF6-21) and DH10B (pASANFA1; pAF6-21) were induced for five hours by the addition of l-arabinose. The samples were sonicated, filter-sterilized and assessed for their effect on grass grub larvae by standard bioassay. Data showed that only the DH10B (pASANFA1; pAF6-21) strain caused the antifeeding effect of grass grub larvae (Table 2), indicating that the induction of anfA1 resulted in the production of the Afp. On the premise that the Afp was able to form a virus-like structure, a chloroform-based bacteriophage isolation was undertaken. Bioassay analysis showed that 5 µL of neat Afp purified in this manner caused cessation of feeding and amber colouration within 48 h postinoculation (Table 2).

View this table:
Table 2

Effect of Escherichia coli DH10B(pASANFA1, pAF6-21) arabinose induced Afp preparations and the effect of temperature on the efficacy of the chloroform Afp preparation towards Costelytra zealandica larvae

Percent antifeeding/percent amber//percent dead
Day 2Day 4Day 12
Sonicated filtrate62.5/62.5//0100/100//2.1100/100//14.6
Glycerol gradient56.3/2.1//0100/2.1//0100/2.1//0
Temperature (°C)
  • Refer to methods ‘Induction and purification of the Afp’.

  • pADAP variant of an antifeeding pathotype (Table 1).

  • Afp negative control, sonicated filtrate of arabinose induced DH10B(pAF6-21).

The effect of temperature on the stability of the Afp was assessed by exposing the Afp to temperatures ranging from 15 to 65°C for a 10-min duration. The results showed that the Afp lost activity towards grass grub larvae at temperatures at or above 60°C (Table 2).

To further purify the Afp, the chloroform-derived supernatant was subjected to two ultracentrifugation steps, allowing the structures to be viewed by electron microscopy.

Structure of the Afp

Through electron microscopy, two forms of the Afp were present: an extended form comprising a sheath and core, and a contracted form with the outer sheath contracted to expose the lower core region (Fig. 2). Comparison of the extended and contracted Afp forms showed that they were of a similar length, ~83–89 nm, but differed in their relative width, with the diameter of the extended form measuring ~15 nm compared with the contracted form, which measured ~12 nm (Fig. 3). The upper apical region of the contracted form was slightly wider than the lower region and has a flattened end from which the predicted core structure can be seen (Fig. 2ac, g). Some of the contracted forms showed a characteristic helix structure located from the apical end to half-way down the sheath (Fig. 2ac), and Afp dimensional data identified that the contracted sheath is approximately half the length of the extended form (Table 3). Both Afp forms showed the presence of a bell-shaped base plate, from which five to six tail fibre elements protruded at the end of stalk-like projections. Located at the end of the tail fibre elements are pads (Fig. 2ac). No variation in the length of the tail fibres (~33.5 nm) was evident. Observations of the extended form showed that 12 distinct striations could be identified, which had a predicted ~8° angle of projection relative to the base plate component (Fig. 2df). Two opposing spiral formations with an ~68° angle could be seen that gave the appearance of triangle-like structures that pointed towards the apical end. Unlike the contracted form, the apical end of the extended form was dome shaped in appearance. The distal end of the predicted core region was visible through the bell-shaped structure (Fig. 2de).

Figure 2

Transmission electron micrographs of the Afp variants. (a–c, g) Contracted form showing the exposed core and the helix-like structure as depicted in c. The apical end appears flattened and wider than the lower region, compared with the conical shape of the extended variant (d–f, h, refer text). The arrow denotes the distal structure located at the end of the core. (d–f, h) Extended form comprising the sheath containing 12 striations, two opposing spirals that are at an ~68° angle (f). The bell-shaped base plate, and associated tail fibres with pads are evident. Internal to the putative base plate, the distal end of the core can be seen (arrow). Electron micrographs were taken at a magnification of × 150 000 (Scale bar=30 nm). Photograph in Fig. 2b and d is derived from digitally photographing the negative on a standard white light box and converting to a negative image using Jasc® Paint Shop Pro™ (version 7.02).

Figure 3

Comparison of the relative dimensions of the contracted and extended forms of the Afp. For dimensional data, refer to Table 3.

View this table:
Table 3

Dimensions of Afp extended and contracted forms

Afp formDimension in nanometres (mean ± SD)
(6 measured)83.53 ± 5.115.01 ± 1.733.5 ± 044.7 ± 2.76.2 ± 0.4
(14 measured)89.45 ± 5.212 ± 5.233.5 ± 0
  • The letters A–H correspond to the diagram shown in Fig. 3.

Owing to the similarity of the Afp to phage tail-like bacteriocins, the Afp was assessed for its ability to form zones of clearing on lawns of various bacteria. Strains tested were S. entomophila strains, A1MO2 and 5.6; the Pseudomonas fluorescens; and the E. coli strain DH10B (Table 1). The purified Afp was unable to inhibit the growth of the tested bacteria (data not shown).


Through the expression of the anfA1, the Afp was produced. Electron microscopy of the Afp showed that it resembled the structure of phage tail-like bacterocins from Rhizobium (Lotz & Mayer, 1972), Yersinia enterocolitica (Strauch et al., 2001), Xenorhabdus nematophilus (Thaler et al., 1995), Budvicia aquatica and Pragia fontium (Smarda & Benada, 2005), and the R-type pyocins of Pseudomonas aeruginosa (Michel-Briand & Baysse, 2002) that have activity against a number of bacteria species. Assessment of the purified Afp for antibacterial activity against E. coli, Pseudomonas and Serratia strains showed no effect on bacterial growth.

Bioassay analysis of the Afp showed that increased levels of the Afp were able to cause amber colouration and cessation of feeding within 48 h postingestion (Table 2). These data concur with that of a previous study showing that the afp located on the pBR322-based pAF6 construct (~50 per chromosome equivalent) caused amber colouration and mortality within 11 days postapplication, relative to pADAP (estimated at one copy per chromosomal equivalent), resulting in the healthy but nonfeeding phenotype (Hurst et al., 2004). This may indicate that in the grass grub system, a finite number of Afp are produced, causing a sublethal effect (antifeeding).

Two forms of the Afp were visualized: an extended form and a contracted form in which the core region was exposed (Figs 2 and 3). The Afp variants measure ~83–89 nm in length and ~12–15 nm in diameter (Table 3; Figs 2 and 3), a size that is comparable to R-type pyocins (Michel-Briand & Baysse, 2002) and the phage tail-like bacteriocins of B. aquatica and P. fontium (Smarda & Benada, 2005). Measurements show that the predicted sheath length of the contracted form is approximately half the length of the extended form (Table 3). Examination of the electron micrographs of the contracted form indicated an intriguing helix shape, and that the apical end was slightly wider than the base of the predicted sheath (Fig. 2). This may indicate that the sheath folds in on itself, causing the expansion and flattening of the apical end. Further evidence for this comes from the comparative diameter of the contracted and the extended forms, with the contracted form ~3 nm wider than the extended form. A similar scenario is seen with the analysis of dimensional data of phage tail-like bacteriocins of B. aquatica and P. fontium (Smarda & Benada, 2005).

It is postulated that the first 16 ORFs comprise the carriage region that forms the R-type pyocin structure, which functions as the delivery system for the putative toxin component/s Afp17, Afp18 (Fig. 1; Hurst et al., 2004). Both the afp17 and afp18 ORFs have been placed under the control of an arabinose promoter, allowing their inductive expression as either independent entities or as an afp17–18 combination and expressed in E. coli. The induced proteins were unable to cause an antifeeding pathotype when fed to grass grub larvae (data not shown), suggesting that these proteins may need to be internalized within a target cell to cause an effect.

Previous analysis of the protein sequences derived from the Afp and its PVC orthologues showed that the putative tail-like fibres are more related to the adenovirus family than bacteriophage-related tail fibres (Fig. 1; Hurst et al., 2004), indicating that the Afp and its orthologues may have an affinity with a site on a eukaryotic cell. Combined, these data suggest that once the Afp/PVC has made contact with the target cell, the extended form becomes contracted, increasing the diameter of the sheath region as it folds upon itself, causing an extrusion of the inner core, allowing the delivery of the toxin into the cell, a process that could define the Afp and its PVC orthologues as micro-injection units.

The results from this study demonstrated that the expression of the anfA1 is able to induce the production of the afp gene cluster, suggesting that the anfA1 is a member of the RfaH family of transcription elongation factors. DNA sequence analysis of pADAP identified the location of a second ops element (pADAP gi:48995178 39549–39559; 5′-GGCGGTAGCGT-3′) positioned internal to sefC, the third ORF of the putative Sef Class I fimbriae cluster (Hurst et al., 2003). Based on this finding and the data of Carter (2004), it is probable that upon the induction of anfA1 both the afp and Class I sef fimbriae are expressed.

The stable virus-like structure suggest that the purified Afp and its orthologues may form the basis of an ecologically sound bio control agent for targeted insect control. Efforts to understand the regulation of anfA1 and how each of the Afp proteins contribute to virulence and structure are in progress.


The authors thank Manfred Ingerfeld at the University of Canterbury for Transmission Electron Microscopy work, and Richard Townsend for the collection of grass grub larvae. This research was funded by grant C10X0313 of the New Economy Research Fund (NERF), administered by the New Zealand Foundation for Research, Science and Technology.


  • Editor: Reggie Lo


View Abstract