Blackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382X© 2005 The Authors; Journal compilation © 2005 Blackwell Publishing Ltd? 2005592503512Original ArticleInsecticidal toxin of Yersinia enterocoliticaG. Bresolin et al.
Molecular Microbiology (2006) 59(2), 503–512
doi:10.1111/j.1365-2958.2005.04916.x First published online 1 December 2005
Low temperature-induced insecticidal activity of Yersinia enterocolitica Geraldine Bresolin,1 J. Alun W. Morgan,2 Denise Ilgen,1 Siegfried Scherer1 and Thilo M. Fuchs1* 1 Zentralinstitut für Ernährungs- und Lebensmittelforschung (ZIEL), Abteilung Mikrobiologie, Technische Universität München, Weihenstephaner Berg 3, D-85350 Freising, Germany. 2 Warwick HRI, University of Warwick, Wellesbourne, Warwick CV35 9EF, UK. Summary The insecticidal toxin complexes (Tcs) are produced by several Enterobacteriaceae associated with insects, such as Photorhabdus luminescens, Serratia entomophila and Xenorhabdus nematophilus. Genome sequences revealed tc-like genes in Yersinia spp., but insecticidal activity of this genus associated with the toxins has not been described. Through the search for genes upregulated at low growth temperatures in Yersinia enterocolitica strain W22703, a genomic island of 19 kb termed tc-PAIYe with homologues of the toxin genes tcaA, tcaB, tcaC and tccC was identified. Southern blot and polymerase chain reaction (PCR) analysis of 34 strains demonstrated that the tc-PAIYe is present in biovars 2, 3 and 4, but neither in biovars 1A and 1B, nor in five Yersinia species apathogenic in humans. Using the luxCDABE operon as reporter, the expression of the toxin genes was shown to be completely repressed in cells cultured at 37°C, and to increase by 4.6 orders of magnitude when the growth temperature was decreased gradually to 10°°C. These data provide the first indication that temperature is a critical parameter for induction or repression of tc gene transcription. Whole-cell extracts of Y. enterocolitica strain W22703 cultivated at 10°°C, but not at 30°°C, led to insect mortality when fed to Manduca sexta larvae, in contrast to an insertional tcaA mutant. Overall the results suggest that the tc-PAIYe could play an important role in the transmission and survival of pathogenic Y. enterocolitica strains outside mammalian hosts.
Accepted 15 September, 2005. *For correspondence. E-mail
[email protected]; Tel. (+49) 8161 713859; Fax (+49) 8161 714492.
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd
Introduction The yersiniae are Gram-negative, rod-shaped, non-spore forming, facultative anaerobes that belong to the family Enterobacteriaceae (Robins-Browne, 1997). Three of 11 species, Yersinia pestis, Y. pseudotuberculosis and Y. enterocolitica, are human pathogens, and their virulence correlates with the presence of a highly conserved 70 kDa plasmid (Cornelis et al., 1998). Y. enterocolitica, like Y. pseudotuberculosis, infects a variety of animals and birds and may cause gastroenteritis in humans. Most primary, pathogenic strains of humans and domestic animals belong to biovars 2–5 that are unable to kill mice and are mainly found in Europe and Japan (the so-called ‘Oldworld’ strains) (Wren, 2003). In contrast, Y. enterocolitica strains of biovar 1A are assumed avirulent for the lack of known virulence determinants such as the virulence plasmid pYV (Tennant et al., 2003a), and also because biovar 1A strains are frequently isolated from terrestrial and freshwater ecosystems. However, some strains of this biovar are able to cause disease by mechanisms that are independent of pYV or other known virulence determinants (Tennant et al., 2003a). Biovar 1B bacteria form a geographically distinct group, which were predominately isolated from patients in the USA (‘New-world’ strains). They are particularly virulent for humans due to a highpathogenicity island (Schubert et al., 1998) and are lethal for mice. While Y. enterocolitica is widely found in the environment and often transmitted by unprocessed food due to its ability to grow at low temperature (Robins-Browne, 1997), its association with insects has not been reported. However, Y. pestis, a blood-borne pathogen and the aetiological agent of human plague, has long been known to be transmitted by insects, specifically by fleas (Perry and Fetherston, 1997). Recently, insecticidal Tc homologues have been suggested to be important in the association of Y. pestis and Y. pseudotuberculosis with insects (Waterfield et al., 2001a; 2004). Tc homologues have not only been described in yersiniae, but also in other insectassociated bacteria such as Serratia entomophila (Hurst et al., 2000), Xenorhabdus nematophilus (Morgan et al., 2001) or Photorhabdus luminescens (Bowen et al., 1998; Duchaud et al., 2003). The latter two live in a complex symbiotic relationship with the entomophagous nematodes. Following invasion of an insect by the nematode,
504 G. Bresolin et al. the bacteria are released from the nematode gut into the haemocoel, the open circulatory system of the insect. By expressing a variety of virulence factors including the Tc proteins, the bacteria are thought to help in killing the insect host, thus providing a source of nutrients for both the bacteria and the nematodes (Forst et al., 1997). The Tc proteins were first characterized in, and purified from, P. luminescens strain W14 (Bowen and Ensign, 1998; Bowen et al., 1998; Guo et al., 1999). The respective genes encode four high-molecular-weight toxin complexes named Tca, Tcb, Tcc and Tcd with oral insecticidal activities comparable with those of Bacillus thuringiensis toxins (Bowen et al., 1998). The predicted amino acid sequences of the Tc proteins show three areas with high similarity to each other, and basically the toxins comprise the products of three gene families only (namely tcaAB/tcdA, tcaC/tcdB and tccC). Recent studies based on recombinant expression of tc genes from P. luminescens in Escherichia coli show that the combination of three genes, tcdA, tcdB and tccC, is essential for oral toxicity to Manduca sexta (Waterfield et al., 2001b), and similar results were obtained for xptA1/tcbA, xptB1/tccC1 and xptC1/tcaC from X. nematophilus (Morgan et al., 2001; Joo Lee et al., 2004). Further experiments supported the hypothesis that TccC-like proteins might act as universal activators of, or chaperons for, different toxin proteins, while Tca-like and Tcd-like proteins contribute predominately to the oral toxicity of bacterial supernatants (ffrench-Constant et al., 2003). Despite the well-documented association of Y. pestis and Y. pseudotuberculosis with nematodes and insects (Hinnebusch et al., 1996; Joshua et al., 2003), the expression pattern and the biological role of insecticidal toxins in Yersinia spp. still remain to be disclosed. Here, we describe the discovery of a novel pathogenicity island in Y. enterocolitica W22703 which harbours tc-like genes, and that is present in biovars 2–4, but not in the representatives of biovars 1A and 1B tested. The genetic composition of this island, as well as the temperature-dependent expression of the putative tc-operon, is presented. The most striking result is the demonstration that the Tc-like elements encoded by Y. enterocolitica are orally active against the larvae of M. sexta, the tobacco hornworm, but only if a low-growth temperature was used to culture the bacteria and produce a protein extract.
Results Sequence analysis of a genomic island of Y. enterocolitica W22703 A transposon mutant library was established in Y. enterocolitica W22703 using the plasmid pUT mini-Tn5 luxCDABE Km2. Two independent mutants exhibiting
higher levels of light emission at 10°C were identified. The insertion point of the transposon was determined by inverse polymerase chain reaction (PCR). The translated nucleotide sequences flanking the mini-Tn5 insertion site of both mutants showed significant similarities to parts of the Tc proteins of P. luminescence and Y. pestis. Surprisingly, no homologies to the genome sequence of Y. enterocolitica strain 8081v were found. Starting from the two transposon insertion sites of the mutants W22703tcaA(134)::Tn5lux and W22703-tcaB1(152)::Tn5lux, the surrounding chromosomal region of 20 403 bp was sequenced as described in Experimental procedures. When compared with the genome sequence of strain 8081v, the left hand junction of the nucleotide sequence revealed to be identical to open reading frames (ORFs) YE3795, YE3796 and YE3797, encoding a HlyD family secretion protein, a putative lipoprotein and a putative LysR-like transcriptional regulatory protein respectively. To define the 3′-end of tc-PAIYe, we amplified and sequenced a fragment of approximately 3 kb comprising the last 462 nucleotides of tccC and the last 519 nucleotides of tldD (YE3798) encoding a putative DNA gyrase modulator. The sequence data show that the strain-specific region had been inserted between ORFs YE3797 and YE3798 with respect to the 8081v genome. The annotation of the whole nucleotide sequence revealed 10 strain-specific ORFs that are schematically represented in Fig. 1. The ORFs were named according to the similarity of their amino acid sequences to the proteins encoded by the toxin genes tcaA, tcaB, tcaC and tccC from P. luminescens and Yersinia spp. as outlined in Table 1. According to a sequence deletion of approximately 200 bp as compared with Y. pestis, the tcaB locus is divided into two ORFs that were therefore named tcaB1 and tcaB2. Five genes, tcaA, tcaB1, tcaB2, tcaC and tccC, encode conserved proteins that belong to the three TcaAB-like, TcaC-like and TccClike insect toxin elements. The genes coding for TcaABClike and the TccC-like elements identified here are separated by ORFs encoding two phage-related proteins and a putative exported protein. tccC is followed by approximately 1500 bp containing ORF11 that encodes a hypothetical protein. The gene tcaA is preceded by two genes showing homologies to putative LysR-like regulators. Due to the potential role of these proteins in the regulation of tc-like genes, we provisionally termed the two ORFs tcaR1 and tcaR2. The tcaC sequence shows a frameshift at nucleotide 314 that introduces a stop codon at the position indicated in Fig. 1, but the expression of a functional C-domain of 1433 amino acids cannot be excluded. The frameshift is not present, however, in the respective region of the Y. enterocolitica strains NCTC10460 and H270/78, but was also found in H324/78. No similarities to tRNAs, integrases or IS elements could be found. The strain-specific sequence has an average G+C composi-
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
Insecticidal toxin of Yersinia enterocolitica
505
Fig. 1. Genetic organization of the insecticidal pathogenicity island tc-PAIYe of Y. enterocolitica strain W22703. The genes encoding two putative LysR-like regulators are shown in grey, and the genes encoding the TcaAB-like, TcaC-like and TccC-like elements are depicted in black. The two insertion sites of mini-Tn5 luxCDABE Km2 in the mutants used for initial sequencing and expression profiling are indicated (triangles). The asterisk marks the frameshift in tcaC. Numbers indicate ORFs as described in Table 1. The pathogenicity island is highlighted; genes tcaR2 to ORF11 are specific for strain W22703.
tion of 45.4% which is slightly lower than that of the genome sequence of strain 8081v (47.3%). We propose to name the 19 kb region that spans from tcaR1 to ORF11, the tc-like pathogenicity island of Y. enterocolitica, abbreviated tc-PAIYe. Distribution of homologous tc genes in Y. enterocolitica biovars The complete tc-PAIYe sequence was obtained from Y. enterocolitica strain W22703 and no nucleotide similarities to the genome sequence of strain 8081v were found, with the exception of the putative regulator gene tcaR1. To gain further information on the distribution of tc-PAIYe and tc-like genes in five biogroups of Y. enterocolitica, we subjected a total of 29 strains to PCR analysis using primers specific for tcaA, tcaB1, tcaB2 and tcaC. The
strains had been isolated from humans, the environment, unprocessed food or domestic animals. The results shown in Table 2 indicate that all four toxin genes are present in biovars 2–4. Interestingly, no PCR fragments were obtained from chromosomal DNA of biovars 1A and 1B. The observation of a biovar-specific distribution of genes encoding Tcs was further confirmed by Southern blot analysis under standard and low-stringency conditions using a DNA probe amplified from tcaB1. The respective 428 bp fragment is highly conserved between Y. enterocolitica, Y. pseudotuberculosis and P. luminescens. This fragment hybridized to chromosomal DNA of all tested Y. enterocolitica biovars 2–4, as well as to Y. pseudotuberculosis and P. luminescens (data not shown), but not to DNA of biovars 1A and 1B, demonstrating the presence of a homologue of tcaB1 in the respective strains (Fig. 2). No hybridization of the tcaB1 probe to
Table 1. Nomenclature of tc-PAIYe ORFs as derived from identity and similarity of predicted proteins with and to sequences in the SWISSPROT and EMBL databases. Gene
aa
MW (kDa)
Homologous protein/putative function
Identity/similarity (%)
tcaR1 tcaR2 tcaA tcaB1 tcaB2 tcaC ORF7 ORF8 ORF9 tccC ORF11
303 288 719 352 769 1500 103 133 119 969 264
34.0 33.0 81.2 40.9 85.5 168.5 11.3 15.5 13.4 106.5 29.1
Putative LysR-family transcriptional regulatory protein Putative LysR-family transcriptional regulatory protein Tc subunit TcaA Tc subunit TcaB Tc subunit TcaB Tc subunit TcaC Putative phage-related protein; putative holin protein Putative phage-related protein; putative endolysin Putative exported protein Tc subunit TccC Hypothetical protein
99/99Ye; 93/97Yptb; 92/96Yp; 29/52Pl 66/77Ye; 62/78Yptb; 59/74Yp; 36/55Pl 40/54Yptb; 39/53Yp; 38/59Pl; 21/40Ye 56/73Yptb; 56/73Yp; 36/55Pl; 33/51Ye 63/76Yp; 62/75Yptb; 45/59Pl; 42/55Ye 71/80Ye; 59/72Yptb; 58/71Yp; 49/62Pl 74/82Ye; 69/82Yptb; 69/82Yp; 54/72Pl 80/85Ye; 71/84Yp; 71/84Yptb; 54/72Pl 78/91Ye; 61/75Yp; 60/73Yptb 65/74Ye; 58/69Yptb; 58/69Yp; 51/64Pl 34/49Pl
For BLAST analysis, the amino acid sequences of the two tcaC frames were coupled. aa, amino acids; MW, predicted molecular weight; Yp, Y. pestis CO92 or KIM or biovar Medievalis strain 91001; Yptb, Y. pseudotuberculosis IP32953; Ye, Y. enterocolitica 8081v or T83; Pl, P. luminescens W14 or ssp. laumondii TT01. © 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
506 G. Bresolin et al. Table 2. Distribution of tc-PAIYe genes tcaA, tcaB1, tcaB2 and tcaC in representatives of five Y. enterocolitica biovars. Biovar
Serovar
Origin
Code
PCR
Southern blot
1A
O:10 O:4,33 O:41,43 O:41,43 O:41,43 O:5 O:5 nd
Patient (Germany) Patient (Germany) Food Food Patient (Germany) Patient Patient Milk
SZ671/04 SZ1167/04 SZ593/04 SZ554/04 SZ634/04 H79/83 H1527/93 WS1968
– – – – – – – –
– – – – – – – –
1B
O:8 O:8 O:8
Patient (Germany) Patient (Germany) Patient (USA)
SZ506/04 SZ375/04 8081v
– – –
– – –
2
O:5,27 O:5,27 O:9 O:9 O:9
nd Patient (Germany) nd Patient nd
H280/83 SZ1249/04 H692/94 H621/87 W22703
+ + + + +
+ nd + + +
3
O:1 O:5,27 O:5,27 O:9 O:9 O:9
Chinchilla (Denmark) Patient Patient Swine nd nd
NCTC10460 H230/89 H582/87 H324/78 H7580/93 H7692/93
+ + + + + +
+ nd nd + + +
4
O:3 O:3 O:3 O:3 O:3 O:3 O:3
Dog faeces Swine Patient Patient Swine Swine Dog faeces
H270/78 H31/80 H608/87 H450/87 H469/87 SZ425/04 SZ687/04
+ + + + + + +
+ nd nd nd nd nd nd
H 27 0/ 78 H (4 69 ) 2/ 9 4 H 32 (2 4/ ) 78 H (3 28 ) 0/ 83 H 79 (2) /8 3 (1 N A) C TC 10 W 46 22 0 (3 70 ) 80 3 (2 81 ) v (1 B)
[bp]
M
The strain code refers to the collection of the Institut für Hygiene und Umwelt, Hamburg. +, positive signal; –, no signal; nd, not determined.
chromosomal DNA of the non-pathogenic species Yersinia ruckeri, Y. kristensenii, Y. frederiksenii, Y. intermedia and Y. aldovae was observed. Low temperature-dependent transcription of tcaA
8576 7427 6106 4889 3639 2799
1953
Fig. 2. Southern blot analysis supports the hypothesis that Y. enterocolitica biovars 2–4 carry a pathogenicity island encoding insecticidal toxin genes. The highly conserved region of the tcaB gene from strain W22703 was used as probe. Chromosomal DNA was digested with EcoRI and hybridization was performed at standard conditions and visualized using the digoxigenin system. The biovar is indicated in parenthesis.
Mutant W22703-tcaA(134)::Tn5lux was initially identified by screening a mutant library for genes with upregulated expression during growth at 10°C compared with 30°C. To gain a further insight into the temperature-dependent expression of the tc genes of Y. enterocolitica, the transcriptional activity of tcaA at different temperatures was investigated. Cultures of W22703-tcaA(134)::Tn5lux were grown overnight at 30°C, diluted as described and incubated at 10°C, 15°C, 20°C, 25°C, 30°C and 37°C. Bioluminescence was measured over all growth phases until cells reached stationary phase (Fig. 3A). No light emission above background was observed during growth at 37°C, suggesting a complete block of tcaA expression at this temperature. Induction of luciferase activity as a reporter for tcaA transcription, however, was observed when the growth temperature was gradually decreased to 10°C. The absolute, relative light units (RLU) at optical density at 405 nm (OD405) of ∼1.0 ranged from 2742 (30°C) to 54 158 (25°C) and 341 912 (15°C), indicating a
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
Insecticidal toxin of Yersinia enterocolitica
10 6
A
RLU
10 5
10 4 10 3
10 2 10 1 0.0
0.2
0.4
0.6
0.8
1.0
1.2
OD405
107
B
10°C
15°C
20°C
25°C
30°C
37°C
cRLU
5
104 103 102 101 0
temperature optimum of the luciferase at 37°C, 76 independent mutants that had not shown temperature-dependent luxCDABE transcription were taken from the transposon library, and their light emission was measured at different temperatures. For each factor, the mean of five RLU values at four different time points was divided by the respective OD405 values, resulting in the approximated correction factors 1.6, 2.5, 3.1, 5.1 and 19.7 at 30°C, 25°C, 20°C, 15°C and 10°C. A temperature optimum of 10–15°C for tcaA transcription was determined by multiplication of the respective correction factors with the RLU of cells reaching the stationary phase at OD405 of ∼1.1 (Fig. 3B). The logarithm of the quotient of the corrected RLU values at 10°C (2142 493) and 37°C (52) is 4.62, indicating that the expression of tcaA increased by more than four orders of magnitude when Y. enterocolitica was grown at low temperature compared with mammalian body temperature. Insecticidal activity
106 10
507
10°C
15°C
20°C
25°C
30°C
37°C control
temperature Fig. 3. Temperature-dependent transcription of tcaA. A. Using mutant W22703-tcaA(134)::Tn5lux and the luciferase as transcriptional reporter, the expression profile of tcaA at different temperatures was derived. B. The RLU values detected at OD405 of ∼1.0 were multiplied with a factor that considers the temperature-dependent activity of the enzymes involved in light emission (corrected RLU, cRLU). A maximal transcription of tcaA at 10°C, and a complete lack of luciferase activity at 37°C was observed. The data represent the means and standard errors of means for five independent measurements. The control value was derived from 10 wells filled with LB medium.
striking increase of tcaA expression at temperatures below 30°C. By relating RLU to the growth phases (RLU/ OD405), initial induction of bioluminescence was observed at OD405 of 0.2 at 15°C, and transcriptional activity steadily increased until cells entered the stationary phase. A similar expression profile was derived from mutant W22703tcaB1(152)::Tn5lux (data not shown). As the activity of an enzyme decreases with lower temperatures, the ‘real’ light emission can be determined as the product of the RLU value and a temperature-dependent factor. To determine this factor with respect to the assumed
As model organisms to test for insecticidal activity M. sexta and Pieris brassicae were selected. M. sexta was chosen as the tc-PAIYe genes showed greatest similarity to the Tc elements of P. luminescens that are active against this insect. X. nematophilus and P. brassicae combination served as a highly sensitive control. The culture supernatants or cell extracts recovered from Y. enterocolitica cultures grown at 30°C did not show insecticidal activity to M. sexta above that of the negative control. However, cell extracts from the two Y. enterocolitica strains grown at 10°C showed a significant level of insecticidal activity against M. sexta, with a slightly lower activity of NCTC10460 compared with strain W22703. This effect might rather be the result of sequence variations within the genes encoding the enzymatically active toxins, than of the frameshift in tcaC. The culture supernatants from the strains grown at 10°C and 30°C did not show insecticidal activity. As expected, culture supernatants and cell extracts of X. nematophilus displayed good insecticidal activity towards P. brassicae, and no detectable insecticidal activity towards M. sexta was observed. Wild-type strain 8081v that lacks tc-PAIYe and any tc-like gene in its genome showed no insecticidal activity towards either of the insect tested. When the mutants W22703-tcaA(134)::Tn5lux and W22703tcaB1(152)::Tn5lux were tested for activity against M. sexta, the results indicated that insertional knockout of tcaA, but not tcaB1, resulted in the loss of insecticidal activity when extracts were gained from 10°C cultures (Table 3). As the transposon unit contains two putative transcriptional terminators downstream of luxCDABE and the kanamycin resistance cassette, polar effects of an insertion in tcaA on genes within a putative operon could not be excluded. Therefore, a non-polar tcaA deletion
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
508 G. Bresolin et al. Table 3. Insecticidal activity of Y. enterocolitica and X. nematophilus towards M. sexta and P. brassicae. Species
Strain code
T
Probe
Insect
Toxicity
Y. enterocolitica
W22703 W22703 W22703 8081v NCTC10460 W22703-tcaA(134)::Tn5lux W22703-tcaB1(152)::Tn5lux W22703∆tcaA/pACYC184 W22703∆tcaA/pACYC184-F37.R43(s) W22703/pACYC184
10°C 10°C 30°C 10°C 10°C 10°C 10°C 10°C 10°C 10°C
Cell extract Supernatant Cell extract Cell extract Cell extract Cell extract Cell extract Cell extract Cell extract Cell extract
M M M M M M M M M M
+ – – – + – + – + +
30°C 30°C 30°C 30°C
Cell extract Supernatant Cell extract Supernatant
P P M M
++ ++ – –
X. nematophilus
Data of three independent bioassays were exploited. Positive results were scored when >50% of larvae were killed within 1 day (++) or within 5 days (+). Negative results (–) were scored if less than 5% of larvae were dead by the end of experiments, equivalent to negative control samples. T, growth temperature for the bacterial culture; M, M. sexta, P, P. brassicae.
mutant was constructed that showed no insecticidal activity in the bioassay. The in trans complementation of tcaA restored the wild-type insecticidal activity. In each case, we used strains transformed with plasmid pACYC184 to confirm that the observed effects are unequivocally a result of the cloned fragment. Table 3 summarizes these data. Taken together, tcaA, but not tcaB, is essential for the observed toxic activity of Y. enterocolitica strains against M. sexta larvae. Discussion Bacterial symbionts of nematodes, such as P. luminescens and X. nematophilus, and bacteria that use insects either as hosts (S. entomophila) or as vectors (Y. pestis) carry homologues of Tc elements (Hurst et al., 2000; Parkhill et al., 2001). However, no insecticidal activity has so far been reported for either Y. pestis or Y. pseudotuberculosis. The presence of tc-like genes and other determinants involved in insect association in pathogenic Yersinia spp. has raised questions about their role in the life cycles of these bacteria and the involvement of insects (Waterfield et al., 2004). A low level of expression of the tc loci under the laboratory conditions used, and the lack of an appropriate host insect, are the possible reasons for the failure to demonstrate insecticidal activity of Yersinia spp. in the past. We have reported here that the expression of at least two tc genes, namely tcaA and tcaB1, is strictly temperature-dependent, showing a transcriptional minimum at 37°C and a maximal induction when cells are cultured at 10–20°C (Fig. 3). Indeed, whole-cell extracts of Y. enterocolitica W22703 are toxic for M. sexta larvae only when cells had been cultivated at low temperature. Earlier observations have already shown that tc genes are not constitutively
expressed. Oral toxicity of P. luminescens W14 supernatant is seen only in stationary phase when the bacteria are cultured in vitro (Bowen and Ensign, 1998), and the oral toxicity of Tcd-producing E. coli supernatants has been shown to increase either with the age of culture or with the addition of mitomycin C that might increase cell lysis (Waterfield et al., 2001b). The data presented suggests that elevated Tca production can be obtained through growth at low temperature not only in Y. enterocolitica, but also in other bacteria harbouring tclike genes that aid in the killing and bioconversion of an insect host. Besides a predicted coding sequence similar to an insect virus enhancing gene and the haemin storage locus hms that is required for biofilm formation in nematodes (Joshua et al., 2003) and for blocking the foregut in fleas (Hinnebusch et al., 1996), the pathogenicity island tc-PAIYp can be regarded as another chromosomal determinant important in the association of Y. pestis with insects (Dobrindt et al., 2004). With respect to a frameshift in tcaB in Y. pestis strain CO92, it was argued that the remaining tc gene homologues contribute to a functional insecticidal toxin (Waterfield et al., 2004), and it is therefore interesting to note that the frameshift of tcaC in strain W22703 does not reduce its overall insecticidal activity. The pathogenicity island tc-PAIYe has a slightly lower G+C content in comparison with the genome sequence (45.4% versus 47.3%). Despite the lack of transposase fragments and inverted repeats within the sequenced region, the genome island might have been acquired by horizontal gene transfer by an ancient ancestor before the divergence of the pathogenic Yersinia species. This hypothesis is supported by the finding that TcaA-like and TcaB-like protein sequences from different bacteria, namely P. luminescens and Y. pestis, are closely related
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
Insecticidal toxin of Yersinia enterocolitica (Waterfield et al., 2001a). The tc-PAIYe and their counterparts in Y. pestis KIM and CO92 not only harbour putative regulatory genes and tc-like genes, but also a phagerelated gene encoding an endolysin, raising the possibility that the insecticidal components are released from the bacteria via cell lysis (ffrench-Constant et al., 2003). The presence of a putative haemolysin in the neighbourhood of the insecticidal genome island in Y. enterocolitica is also notable. However, the functional role of these putative virulence determinants is unknown. Most gene products of tc-PAIsYe from W22703 are more closely related to those of Y. enterocolitica strain T83, from which the 20 kb sequence of a similar tc-PAIYe is available (Accession No. AY647257). Interestingly, three Tca-like proteins of strain W22703, namely TcaA, TcaB1 and TcaB2, show a higher similarity to their homologues in Y. pseudotuberculosis and Y. pestis, respectively, than to those in T83, indicating greater variability of the Tca-like elements (Table 1). Indeed, amplification of an internal tcaA fragment, and a fragment comprising the 3′-end and part of the intergenic region between tcaA and tcaB1 from the genome of some biovars 2–4 strains failed, possibly due to insufficient DNA homologies between template DNA and oligonucleotides. However, using a second set of oligonucleotides identical to the 3′-end of tcaR2 and the 5′-end of tcaA PCR fragments were produced in all cases. The non-toxic phenotype of mutant W22703-tcaA(134)::Tn5lux confirms the suggestion that the Tca/Tcd-like elements rather than Tcclike proteins are the carrier of the enzymatic activity of the Tcs (ffrench-Constant et al., 2003) and might thus contribute to the host-specificity of tc-PAIYe. Initially, the tc-PAIYe was detected in W22703 (biovar 2), but not in 8081v (biovar 1B), the only Y. enterocolitica strain for which the whole genome sequence is available. By additional Southern blot and PCR analysis, we detected homologues of the tc-PAIYe in all biovars 2–4 tested, but not in any of the 11 strains of biovar 1A or 1B. The only known exception to this is biovar 1A strain T83, in which fragments with homology to three insecticidal toxin genes have recently been identified (Tennant et al., 2003b). The observed distribution pattern of tc-PAIYe among the five biovars gives rise to speculations about the evolutionary separation of the tc negative biovars 1A and 1B from the ‘Old-world’ strains. In the light of our finding that tc genes of Y. enterocolitica are biologically active towards insects, it is tempting to speculate that the life cycle of some biotypes of Y. enterocolitica includes an insect stage, like that of Y. pestis that alternates between fleas and mammals. Furthermore, we show that temperature is the environmental signal for Tc protein expression. This would switch off expression in a mammalian host, and switch it on when Y. enterocolitica replicates in the soil environment and possibly poikilothermic insects. As not only Y. pestis, but
509
also strains of Y. pseudotuberculosis can infect the nematode worm Caenorhabditis elegans (Joshua et al., 2003), the association of Y. enterocolitica with insects or nematodes requires further investigation. A striking result of our study is that homologues of insecticidal toxin genes could be found in all but one human pathogenic Y. enterocolitica biovars, but in none of five non-pathogenic Yersinia species. Thus, the presence or absence of tc-PAIYe might have a yet underestimated impact on the transmission of pathogenic yersiniae. Experimental procedures Bacterial strains, plasmids and growth conditions Bacterial strains used in this study were taken from the strain collection of the Abteilung Mikrobiologie, ZIEL, Weihenstephan, or have been obtained from the Institut für Hygiene und Umwelt, Hamburg, Germany. The sources of other strains and plasmids are indicated in Table 4. All cultures were grown in Luria–Bertani (LB) broth (10 g l−1 tryptone, 5 g l−1 yeast extract, 5 g l−1 NaCl), or on LB agar (LB broth supplemented with 1.5% agar). E. coli was grown at 37°C, and X. nematophilus and Y. enterocolitica at 30°C or as indicated. If necessary, kanamycin (50 µg ml−1), streptomycin (50 µg ml−1), tetracycline (12 µg ml−1) or nalidixic acid (20 µg ml−1) was added to the media. Strains were frozen in LB containing 13% glycerol at −70°C.
General molecular techniques DNA manipulation and isolation of chromosomal DNA was performed according to standard procedures (Sambrook and Russell, 2001), or to the manufacturer’s protocol. Transposon mutagenesis using pUT mini-Tn5 luxCDABE Km2 was carried out essentially by the method of Winson et al. (1998) using E. coli strain S17.1 λpir. PCR was carried out with Thermoprime Taq polymerase (ABgene, Hamburg, Germany) and the following programme: 95°C for 2 min; 30 cycles at 95°C for 10 s, specific annealing temperature for 30 s, and 72°C for 20–180 s depending on the expected fragment length; and a final extension of 72°C for 10 min. Chromosomal DNA (100 ng) was used as template for PCR amplification. Conjugational transfer was performed using the mobilizing E. coli strain SM10 as the donor for the matings (Simon et al., 1983; Herrero et al., 1990).
Sequencing of insecticidal toxin genes Chromosomal DNA (400 ng) of transposon mutants W22703tcaA(134)::Tn5lux and W22703-tcaB1(152)::Tn5lux was completely digested with ClaI and HindIII. Fragments were treated with T4 DNA ligase (Gibco, CA, USA) to generate circular molecules by self-ligation, and inverse PCR (Ochman et al., 1988) was performed using transposon-specific primer pairs IF1 (5′-tgttccgttgcgctgcccgg-3′)/ClaIR1 (5′-gcgca tcgggcttcccatac-3′) and INV1F (5′-gttgcgctgcccggattacag-3′)/ INV2R (5′-gaaatcaccatgagtgacgactg-3′) derived from the Iend of mini-Tn5. The PCR fragments obtained were sequenced initially using the primers IF1 and INV1F. Chro-
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
510 G. Bresolin et al. Table 4. Strains and plasmids used in this study. Genotype/relevant features
Reference or source
F– φ80∆lacZ∆M15 ∆(lacZYA-argF) U169 recA1 endA1 hsdR17(rk–, mk+) phoA supE44 λ-thi-1 gyrA96 relA1 λ-pir lysogen of S17.1 (Tpr Strr thi pro hsdR–M+ recA RP4::2-Tc::MuKm::Tn7) lacY, tonA, recA, Muc+, thi, thr, leu, supE, RP4-2-Tc::Mu, Kmr, λpir
Invitrogen, Karlsruhe, Germany
Y. enterocolitica W22703 8081v NCTC10460 WS1968 W22703-tcaA(134)::Tn5lux W22703-tcaB1(152)::Tn5lux W22703∆tcaA
Nalr, Res– Mod+, pYV– Wild type, Nalr Camr Isolate from milk Kmr; insertion of Tn5 luxCDABE Km2 into tcaA at nucleotide 134 Kmr; insertion of Tn5 luxCDABE Km2 into tcaB1 at nucleotide 152 Mutant with a non-polar deletion of tcaA
Cornelis and Colson (1975) Miller, St Louis, USA NCTC, London, UK Strain collection Weihenstephan This study This study This study
X. nematophilus ATCC19061
Wild type
ATCC, London, UK
Conditionally replicating vector; R6K origin, mini-Tn5 luxCDABE Km2 transposon, oriT (RP4), Ampr, Kmr Conditionally replicating vector; R6K origin, mobRK2 transfer origin, sucrose-inducible sacB, Strr p15A origin, Camr, Tetr pACYC184 with an EcoRI fragment containing tcaA and its promoter region, Cams
Winson et al. (1998)
Strains E. coli DH5α S17.1 λpir SM10
Plasmids pUT mini-Tn5 luxCDABE Km2 pKNG101 pACYC184 pACYC184/F37.R43(s)
mosomal walking was performed by inverse PCR using the restriction enzymes SphI, HpaI, EcoRI, MunI and DraI (MBI Fermentas) and sequencing with island-specific primers. In addition, a 1053 bp fragment of tcaC was amplified from the chromosomal DNA of strains NCTC10460, H270/78 and H324/78 with the primers toxIF21 (5′-ctgttgcaaagcctgagc-3′) and toxIR17 (5′-gtggaacatcaagacttgc-3′), and sequenced. SequiServe (Vaterstetten, Germany) and Biolux (Stuttgart, Germany) performed sequencing.
Southern blotting and DNA hybridization Chromosomal DNA was completely digested with EcoRI, size fractionated by agarose gel electrophoresis (0.8% agarose), denaturated and transferred to Biodyne® Plus membrane (PALL, Dreieich, Germany) using the VacuGene™XL vacuum blotting system (AmershamPharmacia, Freiburg, Germany) as recommended by the manufacturer. Hybridization, washing and detection of the digoxigenin (DIG)-labelled probe (Roche, Mannheim, Germany) were performed according to the manufacturer’s instructions. A 428 bp fragment was generated by PCR using the primer combination toxIFI (5′-ggtgctgaagtcaacacc-3′) and toxIR2 (5′-aggaacttcctgactgcg-3′) that binds to nucleotides 144–572 of tcaB1. The standard conditions for hybridization of this probe were 15 h at 38°C in 5× SSC containing 50% formamide, followed by two high-stringency washing steps for 15 min each at 65°C in 0.1× SSC and 0.1% SDS. To apply low-stringency conditions, hybridization was performed at 30°C, and washing was carried out for 15 min at room temperature in 2× SSC and 0.1% SDS.
Construction of W22703∆tcaA and pACYC184/F37.R43(s) To construct an in-frame deletion of tcaA, two fragments of
Simon et al. (1983) Simon et al. (1983)
Kaniga et al. (1991) Chang and Cohen (1978) This study
825 bp and 808 bp were amplified using the oligonucleotide pairs tcaA.delF1 (5′-cagccgtacgaccgc-3′)/tcaA.delR1 (5′gagaattcacgttctttatttggcatagctac-3′) and tcaA.delF2 (5′gagaattcttacaaataacaataaaaaac-3′)/tcaA.delR2 (5′-actgcattctcagtgag-3′), and ligated via the introduced EcoRI sites. Following nested PCR with the oligonucleotides tcaA.nestedAB (5′-cgggatccttaaattaactgatagag-3′) and tcaA.nestedCD (5′-cgggatccaatatagtcggccgc-3′) and the ligation mixture as a template, the resulting fragment was cloned into pKNG101 via BamHI, giving rise to pKNG101∆tcaA. This construct was transformed from SM10 into W22703 by conjugation, and streptomycin-resistant bacteria growing at 30°C harbouring the chromosomally integrated plasmid were isolated. Making use of levansucrase encoded by sacB as a positive marker for plasmid excision, bacteria were selected on agar plates containing 5% sucrose, and streptomycinsensitive clones were screened by PCR with appropriate primers to identify mutant W22703∆tcaA in which the second recombination step resulted in the deletion of tcaA. The gene deletion was confirmed by sequencing. The first 18 nucleotides at the 5′-end and the last 36 nucleotides of the 3′-end of tcaA were retained to maintain translation start signals and possible translational coupling with tcaB1. To complement W22703∆tcaA, the complete coding sequence of tcaA and 291 nucleotides of its upstream sequence were amplified with the oligonucleotides toxIF37 (5′-ccggaattcgaaaaaggtgctg gac-3′) and toxIR43 (5′-ccggaattctgagtgggtaagttcatc3′) and cloned into pACYC184 via EcoRI. In the resulting recombinant plasmid, pACYC184/F37.R43(s), the direction of tcaA transcription corresponds to that of the disrupted plasmid gene encoding the chloramphenicol acetyltransferase. Cloning of this construct was performed in E. coli strain DH5α and then transferred into W22703∆tcaA by electroporation.
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
Insecticidal toxin of Yersinia enterocolitica Measuring transcriptional profiles using a luxCDABE reporter Bioluminescence measurements were performed in 96-well plates (Greiner Bio-one, Frickenhausen, Germany). Fresh cultures were prepared by diluting 10 µl of each overnight culture into 190 µl of LB medium and overnight shaking at 30°C and 350 r.p.m. Cultures of the transposon mutants were then diluted 1:4000 in microtitre plates filled with 200 µl of temperated LB medium. The strains were then grown at the appropriate temperature with shaking (500 r.p.m.) until reaching stationary phase. Bioluminescence (490 nm) and OD (405 nm) from all plates were measured in parallel every hour using a Wallac VICTOR2 1420 multilabel counter (Perkin Elmer Life Sciences, Turku, Finland). Bioluminescence was recorded as RLU.
511
sanger.ac.uk/Projects/Y_enterocolitica) and databases at the National Center for Biotechnology Information (http:// www.ncbi.nlm.nih.gov) were used. Sequence similarity was determined using the BLAST algorithm. The nucleotide sequence of the tc-PAIYe recently found in strain T83 is available under Accession No. AY647257. The nucleotide sequences reported here will appear in the EMBL, GenBank and DDBJ databases under the Accession No. AJ920332.
Acknowledgements We thank Peter Roggentin for the kindly gift of Y. enterocolitica strains and Patrick Schiwek for technical assistance.
References Insect bioassay Culture supernatants and cell lysates of Y. enterocolitica and X. nematophilus were prepared according to Sergeant et al. (2003) with minor modifications. Bacterial cells from overnight cultures were inoculated in 300 ml of LB and grown at 30°C and 10°C for 28 h and 48 h respectively. Cells were collected by centrifugation (7000 g for 15 min at 4°C) and the culture supernatants filter-sterilized (0.2 µm pore size) and stored at 4°C. The cell pellet was washed three times by resuspension in 10 ml of PBS (10 mM phosphate buffer, pH 7.4; 2.7 mM KCl; 137 mM NaCl). The final pellet was resuspended in 3 ml of PBS and lysed by sonication at 25% power for 30 s (Sonopuls HD2200, Bandelin electronic, Berlin). Sonication was repeated four times with a 1 min interval on ice between each step. The cell extract was filtered (0.2 µm pore size) and stored at 4°C. The protein concentration was measured with the bicinchoninic acid protein assay kit (Pierce, Rockford, III) following the manufacturer’s instructions. The protein concentration of the cell extract samples was normalized to 0.2 µg ml−1, while culture supernatant samples were used neat. For each a 50 µl sample was spread onto an agar-based artificial diet (David and Gardiner, 1965) which contained streptomycin (20 µg ml−1), cefataxime and tetracycline (each at 100 µg ml−1). To each container (4.5 cm in diameter), 10 larvae were added, and the assay pots were incubated at 25°C (16 h day length period), with a relative humidity of 80% for 120 h. Larvae of P. brassicae and M. sexta were used in these assays, and daily recordings of larvae mortality were made. Three independent replicate assays were carried out. A positive result was scored if over 50% of larvae were dead by the end of the assay, and a negative result if less than 5% of larvae were dead as occasional mortality was detected in control samples. E. coli cells, and separately water, were used as controls for the assay. General visual observations on larval size were also noted.
Bioinformatics For the analysis of the nucleotide and amino acid sequences obtained, the Husar Analysis Package (version 4.0; http:// genome.dkfz-heidelberg.de), the genome sequence annotation for Y. enterocolitica (Sanger Institute, http://www.
Bowen, D.J., and Ensign, J.C. (1998) Purification and characterization of a high-molecular-weight insecticidal protein complex produced by the entomopathogenic bacterium Photorhabdus luminescens. Appl Environ Microbiol 64: 3029–3035. Bowen, D., Rocheleau, T.A., Blackburn, M., Andreev, O., Golubeva, E., Bhartia, R., and ffrench-Constant, R.H. (1998) Insecticidal toxins from the bacterium Photorhabdus luminescens. Science 280: 2129–2132. Chang, A.C., and Cohen, S.N. (1978) Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134: 1141–1156. Cornelis, G., and Colson, C. (1975) Restriction of DNA in Yersinia enterocolitica detected by recipient ability for a derepressed R factor from Escherichia coli. J Gen Microbiol 87: 285–291. Cornelis, G.R., Boland, A., Boyd, A.P., Geuijen, C., Iriarte, M., Neyt, C., et al. (1998) The virulence plasmid of Yersinia, an antihost genome. Microbiol Mol Biol Rev 62: 1315– 1352. David, W.A.L., and Gardiner, B.O.C. (1965) Rearing Pieris brassicae larvae on a semi-synthetic diet. Nature 207: 882–883. Dobrindt, U., Hochhut, B., Hentschel, U., and Hacker, J. (2004) Genomic islands in pathogenic and environmental microorganisms. Nat Rev Microbiol 2: 414–424. Duchaud, E., Rusniok, C., Frangeul, L., Buchrieser, C., Givaudan, A., Taourit, S., et al. (2003) The genome sequence of the entomopathogenic bacterium Photorhabdus luminescens. Nat Biotechnol 21: 1307–1313. ffrench-Constant, R., Waterfield, N., Daborn, P., Joyce, S., Bennett, H., Au, C., et al. (2003) Photorhabdus: towards a functional genomic analysis of a symbiont and pathogen. FEMS Microbiol Rev 26: 433–456. Forst, S., Dowds, B., Boemare, N., and Stackebrandt, E. (1997) Xenorhabdus and Photorhabdus spp. bugs that kill bugs. Annu Rev Microbiol 51: 47–72. Guo, L., Fatig, R.O., 3rd, Orr, G.L., Schafer, B.W., Strickland, J.A., et al. (1999) Photorhabdus luminescens W-14 insecticidal activity consists of at least two similar but distinct proteins. Purification and characterization of toxin A and toxin B. J Biol Chem 274: 9836–9842. Herrero, M., de Lorenzo, V., and Timmis, K.N. (1990) Trans-
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
512 G. Bresolin et al. poson vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in gram-negative bacteria. J Bacteriol 172: 6557–6567. Hinnebusch, B.J., Perry, R.D., and Schwan, T.G. (1996) Role of the Yersinia pestis hemin storage (hms) locus in the transmission of plague by fleas. Science 273: 367–370. Hurst, M.R., Glare, T.R., Jackson, T.A., and Ronson, C.W. (2000) Plasmid-located pathogenicity determinants of Serratia entomophila, the causal agent of amber disease of grass grub, show similarity to the insecticidal toxins of Photorhabdus luminescens. J Bacteriol 182: 5127–5138. Joo Lee, P., Ahn, J.Y., Kim, Y.H., Wook Kim, S., Kim, J.Y., Park, J.S., and Lee, J. (2004) Cloning and heterologous expression of a novel insecticidal gene (tccC1) from Xenorhabdus nematophilus strain. Biochem Biophys Res Commun 319: 1110–1116. Joshua, G.W., Karlyshev, A.V., Smith, M.P., Isherwood, K.E., Titball, R.W., and Wren, B.W. (2003) A Caenorhabditis elegans model of Yersinia infection: biofilm formation on a biotic surface. Microbiology 149: 3221–3229. Kaniga, K., Delor, I., and Cornelis, G.R. (1991) A wide-hostrange suicide vector for improving reverse genetics in gram-negative bacteria: inactivation of the blaA gene of Yersinia enterocolitica. Gene 109: 137–141. Morgan, J.A., Sergeant, M., Ellis, D., Ousley, M., and Jarrett, P. (2001) Sequence analysis of insecticidal genes from Xenorhabdus nematophilus PMFI296. Appl Environ Microbiol 67: 2062–2069. Ochman, H., Gerber, A.S., and Hartl, D.L. (1988) Genetic applications of an inverse polymerase chain reaction. Genetics 120: 621–623. Parkhill, J., Wren, B.W., Thomson, N.R., Titball, R.W., Holden, M.T., Prentice, M.B., et al. (2001) Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413: 523–527. Perry, R.D., and Fetherston, J.D. (1997) Yersinia pestis – etiologic agent of plague. Clin Microbiol Rev 10: 35–66. Robins-Browne, R.M. (1997) Yersinia enterocolitica. In Food Microbiology – Fundamentals and Frontiers. Doyle, M.P., Beuchat, L.R., and Montville, T.J. (eds). Washington, DC: American Society for Microbiology Press, pp. 192–215.
Sambrook, J., and Russell, D.W. (2001) Molecular Cloning: A Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Schubert, S., Rakin, A., Karch, H., Carniel, E., and Heesemann, J. (1998) Prevalence of the ‘high-pathogenicity island’ of Yersinia species among Escherichia coli strains that are pathogenic to humans. Infect Immun 66: 480–485. Sergeant, M., Jarrett, P., Ousley, M., and Morgan, J.A. (2003) Interactions of insecticidal toxin gene products from Xenorhabdus nematophilus PMFI296. Appl Environ Microbiol 69: 3344–3349. Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for in vitro genetic engineering: transposon mutagenesis in gram negative bacteria. Biotechnology 1: 784–791. Tennant, S.M., Grant, T.H., and Robins-Browne, R.M. (2003a) Pathogenicity of Yersinia enterocolitica biotype 1A. FEMS Immunol Med Microbiol 38: 127–137. Tennant, S.M., Skinner, N.A., Joe, A., and Robins-Browne, R.M. (2003b) Yersinia enterocolitica biotype 1A: not as harmless as you think. Adv Exp Med Biol 529: 125– 128. Waterfield, N.R., Bowen, D.J., Fetherston, J.D., Perry, R.D., and ffrench-Constant, R.H. (2001a) The tc genes of Photorhabdus: a growing family. Trends Microbiol 9: 185– 191. Waterfield, N., Dowling, A., Sharma, S., Daborn, P.J., Potter, U., and ffrench-Constant, R.H. (2001b) Oral toxicity of Photorhabdus luminescens W14 toxin complexes in Escherichia coli. Appl Environ Microbiol 67: 5017– 5024. Waterfield, N.R., Wren, B.W., and ffrench-Constant, R.H. (2004) Invertebrates as a source of emerging human pathogens. Nat Rev Microbiol 2: 833–841. Winson, M.K., Swift, S., Hill, P.J., Sims, C.M., Griesmayr, G., Bycroft, B.W., et al. (1998) Engineering the luxCDABE genes from Photorhabdus luminescens to provide a bioluminescent reporter for constitutive and promoter probe plasmids and mini-Tn5 constructs. FEMS Microbiol Lett 163: 193–202. Wren, B.W. (2003) The yersiniae – a model genus to study the rapid evolution of bacterial pathogens. Nat Rev Microbiol 1: 55–64.
© 2005 The Authors Journal compilation © 2005 Blackwell Publishing Ltd, Molecular Microbiology, 59, 503–512
