Previous Article | Next Article ![]()
Applied and Environmental Microbiology, November 2006, p. 6948-6954, Vol. 72, No. 11
0099-2240/06/$08.00+0 doi:10.1128/AEM.00976-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Entomology, University of California, Riverside, Riverside, California 92521,1 John A. Mulrennan, Sr., Public Health Entomology Research and Education Center, Florida A&M University, Panama City, Florida 32405,2 Interdepartmental Graduate Programs in Genetics and Cell, Molecular and Developmental Biology, University of California, Riverside, Riverside, California 925213
Received 25 April 2006/ Accepted 7 August 2006
|
|
|---|
|
|
|---|
One of the most important subspecies of B. thuringiensis is B. thuringiensis subsp. israelensis. This subspecies is highly insecticidal for the larvae of mosquitoes and blackflies and is presently used in many countries to control pest and vector species of these flies. The insecticidal activity of B. thuringiensis subsp. israelensis results principally from synergistic interactions of major endotoxin proteins, Cry11A (previously CryIVD), Cry4A, Cry4B, and Cyt1A (14, 27, 28). Recently, the nucleotide sequence of the large 128-kb plasmid, pBtoxis, that encodes these proteins was reported (7). Interestingly, comparative analyses of the 158 known and putative proteins encoded by pBtoxis revealed no homology with known plasmid replication proteins. However, comparative nucleotide analysis of pBtoxis and pXO1 of B. anthracis (25) suggested that a putative replication origin is present near the assigned first nucleotide position of pBtoxis (7). The region to the right of this position contains an open reading frame (ORF), pBt001, that encodes a putative peptide that shares about 76% amino acid identity with that encoded by pXO1-49, a gene located close to a putative replication origin of pXO1. Based on the relative positions of these genes and the sequence similarity between pBt001 and pXO1-49, it was suggested that they potentially play a role in plasmid replication (7). In addition, ORF pBt156, which occurs to the left of pBt101, encodes a peptide with weak amino acid similarity to the FtsZ/tubulin-like proteins of Pyrococcus (BAB17294) and pXO1-45. As the FtsZ protein family is known to function in cell division by forming a ring structure at the dividing septum, it was proposed that pBt156 functions in plasmid partitioning (7).
To determine whether the putative origin of replication, along with pBt001 and pBt156, plays a role in pBtoxis replication, we cloned and performed nested deletions and site-directed mutagenesis of nucleotide sequences within this region. Our results suggest that pBt156 and its flanking ORF (pBt157), but not pBt001, are essential for replication of pBtoxis. In addition, we show that these minimal sequences, referred to here as a minimal replicon (minireplicon), can be used as a replicon to construct plasmids for expression of the cry and cyt genes of B. thuringiensis, thereby providing a new plasmid vector for engineering various combinations of endotoxins into this species.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Bacterial strains and plasmids
|
Construction of recombinant plasmids.
To construct pUCE, the 1.2-kb fragment containing the erythromycin resistance gene (erm) from pHT3101 (23) was amplified using primers Em-S-F and Em-S-R (Table 2), digested with SalI, and cloned into the same site in pUC19. Plasmid pUCEG was constructed by inserting a 1.2-kb gyrB gene (AY461780) (21) obtained by PCR with primers gyr-F and gyr-R (Table 2) into the SmaI site of pUCE.
|
View this table: [in a new window] |
TABLE 2. Primers used for PCR amplification
|
![]() View larger version (11K): [in a new window] |
FIG. 1. Deletion mutants of the KpnI fragment used to identify a minireplicon of pBtoxis. The ability of each of the mutants to replicate in B. thuringiensis subsp. israelenses 4Q7 is indicated (+, capable; , incapable).
|
Transformation.
Transformation of B. thuringiensis subsp. israelensis 4Q7, B. megaterium VT1660, B. cereus NRRL B-569, and B. subtilis PY79 with vectors bearing the minireplicon was performed as described previously (29, 31, 34, 35). For controls, B. cereus was transformed with pC194 (18) and B. subtilis and B. megaterium with pBU4 (8).
Site-directed mutagenesis.
The putative translational start codons (ATG) for ORFs pBt156 and pBt157 were replaced with CCC and TTT, respectively, by PCR using plasmid pBtSw as the template and primer pairs ORF156-F and ORF156-R, and ORF157-F and ORF157-R (Table 2), which were designed in inverted tail-to-tail directions to amplify the entire pBtSw. After PCR, the amplified linear DNAs were self-ligated to generate pBt
156 and pBt
157.
DNA probes.
The 2.4-kb fragment containing the erm and gyrB genes in pUCEG (see above) was amplified by PCR with primers gyr-F and Em-S-R (Table 2) and labeled using the Oligonucleotide 3'-end labeling kit (second generation; Roche).
Relative copy number determination.
The plasmid copy number was determined according to the method of Devine et al. (13), using the chromosome-specific gyrB and plasmid-specific erm probes. Fragments that hybridized were analyzed with the AlphaEaseFC software, and the ratio of signals from the erm gene to the gyrB gene was used to estimate of the number of plasmid copies per chromosome equivalent.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
Bacterial strains were grown in 20 ml of NBG broth (26) at 30°C for 5 days, by which time the cells had sporulated and lysed. An equal volume (1.2 ml) from each culture was sedimented, and the proteins were seperated in a 10% polyacrylamide gel by SDS-PAGE (27).
Segregation stability assay.
The stabilities of plasmids pBtKp(+), pBtKp(), and pBtSw in B. thuringiensis subsp. israelensis 4Q7 were determined using methods described previously (22).
|
|
|---|
Minimal sequences (minireplicons) sufficient for replication.
In addition to the presence of a replicon, the 6.3-kb KpnI fragment contains four ORFs, pBt001, pBt156, pBt157, and pBt158, the last three of which putatively code for, respectively, a FtsZ/tubulin-related protein, a DNA-binding protein, and a transcriptional regulator. Using a series of overlapping deletion derivatives of pBtKp, the minimal sequences required for replication included those that contained intact ORF156 and ORF157, i.e., plasmids pBtAg2, pBtXm, and pBtSw (Fig. 1). Plasmids containing intact ORF156 but deficient or partially deficient in ORF157 sequences, or vice versa (pBtXb, pBtAg1, pBtSa1, pBtSa2, and pBtBs), did not replicate (Fig. 1). Thus, the smallest autonomously replicating DNA sequence was located on a 2.2-kb SwaI fragment in pBtSw (Fig. 2). This fragment is defined here as a minireplicon of pBtoxis.
![]() View larger version (85K): [in a new window] |
FIG. 2. Nucleotide sequence of the 2.2-kb SwaI region of pBtoxis from B. thuringiensis subsp. israelensis. The fragment contains a putative operon composed of two ORFs, ORF156 (pBt156) and ORF157 (pBt157). The predicted amino acid sequences are shown beneath the nucleotide sequence. Direct- and inverted-repeat sequences (a, b, c, and d) are indicated by arrows. Signature ribosome binding sequences (RBS) are underlined. Iterons and potential DnaA boxes are boxed and shaded, respectively. The tubulin superfamily signature motif in pBt156 is double underlined, and the helix-turn-helix motifs at positions 43 to 65 with residues that form alpha ( ) helices in pBt157 are underlined in boldface.
|
The pBt157 gene (nt 192 to 506) coded for a putative peptide (ORF157) of 104 amino acid residues (Mr, 11,800), and pBt156 (nt 527 to 1981) coded for a putative peptide of 484 amino acids (ORF156; Mr, 54,400). Previous analyses suggested that ORF157 is a DNA-binding protein with a predicted helix-turn-helix (HTH) motif at amino acids 41 to 62 (7). In our studies, alignment of ORF157 with MarR, a repressor of multiple antibiotic resistance genes in E. coli (1), placed the helix-turn-helix motif at residues 43 to 65 (Fig. 2). Berry et al. (7) reported that ORF156 showed weak similarity to FtsZ/tubulin-like proteins from Pyrococcus (BAB17294; 21% identity) and pXO1-45 (21% identity). Our analyses agree with this conclusion. In addition, we identified the signature motif (GGGTVGTG; residues 138 to 143) (Fig. 2) in ORF156 that is characteristic of the tubulin protein superfamily (11).
pBt156 and pBt157 are required for minireplicon replication.
When the translation start codon of either ORF156 or ORF157 was disrupted, B. thuringiensis subsp. israelensis 4Q7 transformants were not recovered following electroporation with the mutant plasmids (pBt
156 and pBt
157), indicating that both ORF156 and ORF157 are required for minireplicon replication.
Determination of the copy number for pBtSw.
Densitometric analysis of hybridization bands showed that the intensity of the erm gene derived from pBtSw was two to three times that of the single chromosomally derived gyrB gene (data not shown), indicating a copy number of two or three per chromosome equivalent.
Segregation stability.
Studies of the segregation stabilities of pBtKp and pBtSw, which contained the 6-kb KpnI and the 2.2-kb SwaI fragments, respectively, showed that these plasmids were unstable in the absence of positive selection with erythromycin. Virtually 100% of the cells from colonies grown without selection after 20 generations failed to grow in LB medium with erythromycin.
Replication of the pBtoxis minireplicon in other Bacillus species.
Plasmid pBtSw, which contained the pBtoxis minireplicon, was tested for its ability to replicate in B. cereus, B. megaterium, and B. subtilis. Whereas the control plasmids replicated after transformation in these species, pBtSw did not. Specifically, pC194 replicated in B. cereus (
300 CFU/µg DNA) and pBU4 replicated in both B. subtilis and B. megaterium (
500 CFU/µg DNA). No colonies were obtained with these species transformed with pBtSw. Furthermore, when wild-type B. thuringiensis subsp. israelensis IPS82, which contains pBtoxis, was transformed with plasmids containing the minireplicon, very few colonies (
5%) produced crystal toxins, indicating a loss of parental pBtoxis in most transformants. Incompatibility between the B. thuringiensis recombinant and resident plasmids from which the replicons in recombinants were derived has been reported previously (6).
Expression of cry and cyt genes using the minimal replicon.
Recombinant 4Q7 cells transformed with pBtH (Fig. 3A), the plasmid containing the minimal replicon, along with the 10-kb HindIII fragment of pBtoxis that contains the cry11Aa, cyt1Aa, and p20 genes (7), synthesized large inclusions that were clearly visible by light microscopy (Fig. 3B). The presence of Cry11A and Cyt1A was confirmed by SDS-PAGE analysis, which showed that the recombinant strain containing this plasmid synthesized amounts of these toxins similar to those produced by B. thuringiensis subsp. israelensis strain 4Q5 (Fig. 3C).
![]() View larger version (53K): [in a new window] |
FIG. 3. Synthesis of Cry11A and Cyt1A using the minireplicon sequence derived from pBtoxis. (A) Map of recombinant plasmid pBtH constructed using the minireplicon in pBtSw (Fig. 1) and the 10-kb HindIII fragment from pBtoxis that contains the cry11A, cyt1A, and 20-kDa genes. (B) Phase-contrast microscopy showing crystals (arrows) in cells of B. thuringiensis subsp. israelensis strain 4Q7 transformed with pBtH. (C) SDS-PAGE analyses of recombinant strains and wild-type strains of B. thuringiensis subsp. israelensis, strain 4Q7 pBtH (lane 1), strain 4Q5 (lane 2), strain pBtSw (lane 3) (Fig. 1), and strain 4Q7 (lane 4). Lane M, protein molecular mass standards.
|
|
|
|---|
Modes of plasmid replication from a wide variety of prokaryotes fall within two categories of mechanisms, the so-called rolling-circle mechanism and the theta mechanism. Whereas the rolling-circle mechanism is the preferred mode for replicating small plasmids (less than
15 kb), large plasmids typically replicate by the theta mechanism (5, 6, 19, 30). At present, five large plasmids with low sequence homology, three (p43, p44, and p60) from B. thuringiensis subsp. kurstaki HD263 and two (pHT73 and pAW63) from B. thuringiensis subsp. kurstaki HD73, have been shown to replicate by the theta mechanism (4, 6, 15, 33). Although pBtoxis shares no significant homology with other large plasmids in B. thuringiensis (7), the structural features of its minireplicon, which include putative iteron sequences and DnaA boxes, suggest that it also replicates using the theta mechanism.
Although we have shown that ORF156 and ORF157, encoded by, respectively, pBt156 and pBt157, are essential for replicating the pBtoxis minireplicon, the precise roles of these proteins remain unknown. The presence of a conserved HTH motif in ORF157 (7), however, suggests a functional role in DNA binding. The HTH motif is present in many prokaryotic and eukaryotic proteins that function in transcription, DNA replication and repair, RNA metabolism, and protein-protein interactions in diverse signaling pathways (3). As plasmid-encoded Rep proteins, which also contain HTH, together with DnaA, accumulate at AT-rich sites at replication origins to initiate plasmid replication (12), we propose a similar role for ORF157. Furthermore, as Rep-DNA-DnaA complexes are also known to play a role in plasmid partitioning (12), ORF157 could play a similar role. In this regard, ORF156, which shows similarity to FtsZ/tubulin-like proteins, could potentially function with ORF157 in plasmid partitioning. The prokaryotic FtsZ protein is a polymer-forming GTPase that shares structural and functional similarities with eukaryotic tubulins (32). FtsZ assembles into a ring structure on the inner surface of the cytoplasmic membrane at the site of cell division. The so-called Z ring is progressively reduced in diameter, a process which leads to invagination of the dividing septum (32). It is possible that ORF156 and ORF157 function together to mediate partitioning of pBtoxis into daughter cells.
The low copy number (two or three copies per chromosome) for the pBtoxis minireplicon was expected, as it is typical for replicons derived from large plasmids (17). The unstable inheritance of the minireplicon when propagated in the absence of antibiotic selection is also not unexpected, because other loci, such as par (12), associated with efficient partitioning of this artificial plasmid are absent. The inability of the pBtoxis minireplicon to replicate in B. cereus, B. megaterium, and B. subtilis suggests that these bacteria lack replicative proteins that recognize cis elements in the minireplicon or proteins that can interact effectively with ORF156 and ORF157. Alternatively, the recombinant plasmid harboring the minireplicon could have been degraded by host cell restriction modification systems (6, 24).
Finally, for commercial and regulatory purposes, one of our primary interests is to develop expression vectors for the synthesis of insecticidal Cry and Cyt proteins using nucleotide sequences native to B. thuringiensis subsp. israelensis. Current regulations for the registration of genetically engineered bacterial insecticides favor recombinants that have as little foreign DNA as possible. The high levels of Cry11A and Cyt1A synthesis that we obtained (Fig. 3) demonstrated that the pBtoxis minireplicon can be used for this purpose and thus has utility in genetically manipulating B. thuringiensis to produce novel combinations of insecticidal proteins.
This research was supported by a grant to B.A.F. from the U.S. National Institutes of Health (AI 145817).
Published ahead of print on 25 August 2006. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»