![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Department of Microbiology, University of Illinois, Urbana, Illinois 61801
Received 13 November 2001/ Accepted 17 April 2002
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Bacteroides spp. are hosts to a variety of transmissible elements, including plasmids, transposons, mobilizable transposons, and conjugative transposons (CTns) (18, 20, 21). The results of a recent survey suggest that it is this last group of elements, the CTns, that is responsible for the significant increase in tetracycline and macrolides-lincosamides-streptogramin-B-type resistance in the Bacteroides group (25).
Bacteroides CTns range in size from 50 to 150 kb (1). The two best-characterized Bacteroides CTns are CTnERL and CTnDOT, which are almost identical except that CTnDOT contains a 13-kb insertion which contains an ermF resistance determinant (Fig. 1A) (34). A novel feature of the Bacteroides CTnDOT/CTnERL family of CTns is that self-transfer and the mobilization of coresident plasmids and mobilizable transposons are all stimulated 100- to 1,000-fold by pregrowth in a medium containing tetracycline. A regulatory region that includes four genes designated tetQ, rteA, rteB, and rteC mediates this tetracycline enhancement effect. This region is located near the middle of CTnDOT (31). In the presence of tetracycline, tetQ, rteA, and rteB are transcribed, and RteB directly or indirectly activates transcription of rteC (31), which subsequently controls the excision and transfer of the CTn.
|
RteC is known to play an important role in the excision of the CTnDOT/CTnERL family of CTns. When a plasmid containing constitutively expressed rteC is provided in trans with a coresident CTnERL element, the CTn excises and transfers constitutively, even in a strain in which rteB has been disrupted (13). Although RteC is necessary for excision and was known to have some undetermined role in transfer of the CTn, RteC alone is not sufficient for either process (6, 13). This suggested that sequences encoded outside the transfer region and the central regulatory region of CTnDOT are required for repression and enhancement of conjugal transfer. The link between RteC and CTn excision has recently been determined (6), but the link between RteC and the modulation of transfer was, until now, unknown.
In this work, a minimal transfer system for CTnDOT was constructed in an effort to localize the regions of CTnDOT responsible for the repression of pLYL72 transfer in the absence of tetracycline and RteC-mediated enhancement of pLYL72 transfer in the presence of tetracycline.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MCR (Bethesda Research Laboratories Inc., Gaithersburg, Md.) or HB101 (4) was grown in Luria broth or Luria broth agar. Antibiotic concentrations (in micrograms per milliliter) were as follows: ampicillin, 100; cefoxitin, 20; chloramphenicol, 10; erythromycin, 10; gentamicin, 200; kanamycin, 50; tetracycline, 1; thymidine, 100; and trimethoprim, 100. The methods utilized for DNA extraction, cloning, and Southern analysis have been described previously (17, 22).
|
QABC. A similar method was employed to introduce single copies of tetQ, rteA, and rteB (QAB) into B. thetaiotaomicron strain BT4100N1S1 (G.-R. Wang, unpublished data).
Recently, it was shown that a strain of B. thetaiotaomicron 4001 containing an integrated minielement (ME) (BT40012.6pGERM) that carried only the CTnDOT int and a small open reading frame, orf2, was able to excise from the Bacteroides chromosome if two other plasmids were provided in trans (6). One of these plasmids, pAMS9, contained the central regulatory region of the CTn (tetQ, rteA, rteB, and rteC), while the other plasmid (pKSO1) contained a gene, exc, shown to be essential for CTn excision (6). Consequently, to check that the regulatory region cloned into pGWA35(1.3) was functional, the construct was put into B. thetaiotaomicron strain BT40012.6pGERM(pKSO1), which contains a site-specifically integrated copy of the CTnDOT integrase and exc. If the integrated copy of QABC were functional, then the ME would be able to excise from the chromosome. Excision of the ME in BT40012.6pGERMQABC(pKSO1) was observed by Southern blot analysis (Fig. 2).
|
QABC, other constructs were introduced into BT4001QABC or BT4100N1S1QAB for transfer assays.
Mobilization and conjugal-transfer experiments.
E. coli HB101 containing IncP plasmid RP1 was utilized to mobilize plasmids transformed into E. coli DH5MCR into Bacteroides strains using a triparental mating procedure described elsewhere (24). Transfers of pLYL72 from B. thetaiotaomicron to E. coli were performed as described previously (13). The transfer frequency is expressed as the number of transconjugants per recipient.
Western blot analyses. To detect the expression of transfer proteins, TraG, TraN, and TraP, the membrane fraction (11, 12) was isolated from the appropriate B. thetaiotaomicron strains (Table 1) and the protein concentration was determined using the DC protein assay (Bio-Rad). Between 100 and 150 µg of each membrane sample was loaded into each well of a 10% sodium dodecyl sulfate-polyacrylamide gel for separation and subsequently electrotransferred to Trans-blot nitrocellulose membranes (Bio-Rad). Western blots were incubated with polyclonal antibodies to TraG, TraN, or TraP (3) diluted 500- to 1,000-fold in TTBS (20 mM Tris [pH 7.5], 0.2% Tween 20, 0.5 M NaCl) with 1% bovine serum albumin at room temperature for 1 to 16 h, depending upon the titer of the antibody. The TraG, TraN, and TraP proteins were detected using the colorimetric Opti-4CN substrate and detection kit (Bio-Rad) and subsequently scanned using a Bio-Rad gel documentation system and Quantity One software.
PCR cloning. In some experiments, derivatives of pKSO1 were generated by using PCR to construct deletions of selected open reading frames (Table 1 and Fig. 3). These constructs were assembled by amplification of fragments of various sizes from a region upstream of orf3 or including the 5' end of orf3 and amplification of a second fragment from inside the 5' end of orf3 to a region 200 bp downstream of exc. These amplicons were subsequently ligated together using restriction sites engineered into the oligonucleotide sequences. The primers utilized for these experiments are summarized in Table 2.
|
|
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
QABC(pLYL72)]. This Bacteroides strain contains the transfer region (pLYL72) of CTnDOT and a single copy of the tetracycline-induced central regulatory region (tetQ, rteA, rteB, rteC, or QABC) that was introduced site specifically into the chromosome (BT4001QABC) using the NBU1 integrase and attP (Fig. 1B) (26). Other regions of CTnDOT that were potentially involved in the modulation of pLYL72 transfer could then be introduced into strain BT4001QABC(pLYL72) and tested for the ability to stimulate or repress transfer (Fig. 1B).
Localization of sequences responsible for repression and enhancement of the conjugal transfer of pLYL72.
Plasmid constructs containing DNA fragments from a 13.0-kb region located between the ermF region and tetQ of the central regulatory region of CTnDOT were tested for the ability to repress or enhance conjugal transfer (Fig. 1A). This region was chosen because it is outside the transfer and central regulatory regions and is present in both CTnDOT and CTnERL. Fragments from this 13.0-kb region (Fig. 3) were transferred to Bacteroides host strain BT4001QABC(pLYL72), and the subsequent frequency of pLYL72 transfer was determined in the presence and absence of tetracycline induction. The results of these transfer experiments are summarized in Table 3.
|
QABC) alone were not sufficient for either repression or enhancement of pLYL72 transfer, since transfer of pLYL72 was constitutive in these backgrounds (Table 3). However, a plasmid containing a 7.6-kb region (pKSO1) caused a 100- to 1,000-fold decrease in the frequency of pLYL72 transfer in the absence of tetracycline and a 100- to 1,000-fold increase in pLYL72 transfer in the presence of tetracycline (Table 3 and Fig. 3). Hence, sequences cloned on pKSO1 were sufficient for both repression and enhancement of pLYL72 transfer. pKSO1 contains nine possible open reading frames (Fig. 3) (GenBank accession number AJ431573). It was previously reported that the putative protein encoded by exc had significant amino acid similarity to proteins encoded by some conjugal plasmids and a CTn, Tn1549 (10), from gram-positive organisms and to chromosomally encoded topoisomerases (6). None of the other putative open reading frames contained by pKSO1 had significant similarity to sequences available in the GenBank database, nor did they contain any motifs that could be indicative of function.
In an attempt to further localize the determinants responsible for the repression and enhancement of conjugal transfer, pKSO1 was subcloned further. In one of these subclones, pKSO4, the 3' end of the exc gene had been truncated, and this neither positively nor negatively affected pLYL72 transfer. This indicated that both repressor and enhancer functions were located in the region of pKSO1 that has been truncated in pKSO4 (Table 3 and Fig. 3). The results showed that although the central regulatory region of CTnDOT was not sufficient for the enhancement of pLYL72 transfer, the central regulatory region is important for the activation of the positive activator(s) of conjugal transfer and that exc is required for enhancement of conjugal transfer. Interestingly, this region of CTnDOT, and exc in particular, has recently been shown to be required for the tetracycline-dependent excision of the CTnDOT ME (6).
Localization of sequences responsible for the repression of pLYL72 conjugal transfer. A 2.5-kb region containing exc, orf2E, and rteR (pKSO5) was sufficient for repression of conjugal transfer. In an effort to further localize the repressor of pLYL72 transfer, subclones of the 2.5-kb fragment containing repressor activity were constructed (Fig. 3). In pGW59, the 7 bp from the C-terminal region of rteR has been deleted, and no repression of pLYL72 transfer was detected. Instead, transfer of pLYL72 appeared to be constitutive. In contrast, the repressor appeared to be functional in subclones pGW60 and pGW88 (data not shown) and in the smallest subclone, pGW87, which contained a fragment of only 778 bp (Table 3 and Fig. 3). This fragment contains a putative open reading frame, designated rteR, which may encode the repressor of conjugal transfer.
Localization of sequences responsible for the enhancement of pLYL72 conjugal transfer. Although pKSO5 contained sequences sufficient for repression of pLYL72 transfer, sequences present in pKSO5 were not sufficient for the enhancement of conjugal transfer. This suggested that sequences upstream of exc were required in conjunction with exc for enhancement of pLYL72 transfer. This was not surprising, since exc was previously shown to be in an operon with orf3 (6). A subclone containing orf2ABCD, an in-frame deletion of orf3, and an intact copy of the exc gene (pKSO7) resulted in enhancement of transfer, indicating that orf3 and orf4A are not required for enhancement of pLYL72 transfer (Fig. 3). Further truncation of the upstream region showed that orf2A (pGW45) and orf2B (pGW46) were also not required for enhancement of pLYL72 transfer.
Deletion of sequences containing putative genes orf2C and orf2D (pYS41 and pKSO6) from the upstream region abolished enhancement of transfer (Table 3). However, since pYS41 was able to complement an exc disruption in CTnERL (CTnERLexc), exc was still produced from this construct, and hence the promoter was still intact (Y. S. Sutanto, N. B. Shoemaker, J. F. Gardner, and A. A. Salyers, submitted for publication). These results suggest that other trans factors in this region are required to work in conjunction with exc to result in enhancement of conjugal transfer.
The expression of the pLYL72 transfer proteins TraG, TraN, and TraP is modulated in the presence and absence of tetracycline. Transfer assays led to the identification of sequences responsible for the 100- to 1,000-fold enhancement of pLYL72 transfer in the presence of tetracycline and the 100- to 1,000-fold repression of pLYL72 transfer in the absence of tetracycline. However, it was unknown whether the modulation of pLYL72 transfer was associated with a change in transfer protein expression or otherwise. Western blot analyses of TraG, TraN, and TraP were undertaken to determine whether the expression of transfer proteins differed when the conjugal transfer of pLYL72 was repressed, constitutive, or enhanced. Antibodies against these three transfer proteins were used because traG, traN, and traP represent three of four operons thought to make up the transfer region of CTnDOT (2), and we wanted to know whether all three operons were regulated similarly.
Results of Western blots with antibodies that detect TraG, TraN, or TraP yielded identical profiles. Thus, the three tra operons are regulated similarly. The results for TraG production show that in the absence of a coresident CTn, the expression of TraG from pLYL72 is constitutive (Fig. 4A). Similarly, when pLYL72 was present in a strain containing the regulatory region from CTnDOT (QABC), the expression of transfer proteins was also constitutive, indicating that the central regulatory region alone does not encode the functions necessary for repression or enhancement of transfer (Fig. 4A). In contrast, when pKSO1, which contains both the repressor and enhancer of transfer, was present in host strain BT4001QABC(pLYL72), TraG expression was repressed in the absence of tetracycline and enhanced above the constitutive level in the presence of tetracycline (Fig. 4A). When exc was truncated, as in pKSO4, or when the sequences upstream of exc alone (pGW41) were present in B. thetaiotaomicron strain BT4001QABC(pLYL72), expression of TraG was no longer repressed or enhanced (Fig. 4A). These results are consistent with transfer experiments that suggest that exc, and sequences downstream of exc, are important for both positive and negative regulatory activities.
|
In the absence of sequences involved in the repression of conjugal transfer, TraG is expressed constitutively from pLYL72, while in the presence of the enhancer and repressor (pKSO7), TraG expression is repressed in the absence of tetracycline and enhanced in the presence of tetracycline (Fig. 4C). Western blot analysis of a strain containing pKSO4 (Fig. 4A) had already suggested that the repressor of transfer was located near the 3' end of exc, and so Western analysis was used to further localize the repressor of transfer. When rteR is truncated (pGW59), TraG production is no longer negatively regulated in the absence of tetracycline, and instead, TraG is produced constitutively. In contrast, when sequences upstream of rteR are deleted and the 250 bp downstream of rteR is still present, the repressor is active (pGW60) and the expression of the pLYL72 transfer protein, TraG, is repressed in both the presence and absence of tetracycline (Fig. 4C).
Since the repression of transfer protein gene expression occurs in the absence of tetracycline, it was possible that the repressor of pLYL72 transfer is expressed in the absence of tetracycline. Similarly, the higher level of transfer proteins only occurs in the presence of tetracycline, and so the positive effector of transfer is only produced under induced conditions. These results are consistent with two differentially regulated factors being involved in the modulation of conjugal transfer. Our results also showed that an increase in conjugal transfer is associated with an increase in the level of transfer proteins detectable by Western blotting. Whether the increase in transfer protein levels detected by Western blotting is attributable to an increase in the stability of the transfer proteins, to more efficient localization and therefore protection of the membrane-associated transfer proteins, or to an increase in expression at the transcriptional or translational level remains to be determined.
RteC is required for production of the enhancer of pLYL72 transfer.
Previous results suggested that RteC plays an important role in conjugal transfer (13). To determine whether RteC is also required for enhancement of conjugal transfer, pKSO1 was introduced into a strain of Bacteroides that contains pLYL72 and tetQ, rteA, and rteB from the central regulatory region but no rteC [BT4100N1S1QAB(pLYL72)] (G.-R. Wang, unpublished). In the absence of RteC, no enhancement of pLYL72 transfer or increase in transfer protein production was detected. Repression of transfer and transfer protein expression, however, was constitutive. In the absence of pKSO1, there is no repression of constitutive pLYL72 transfer (Table 3). These results show that RteC is required for the positive regulation of conjugal transfer.
In the presence of sequences required for enhancement, transfer of pLYL72 between Bacteroides strains is detectable.
Prior to this work, pLYL72 had been shown to contain sequences sufficient for self-transfer from Bacteroides to E. coli (13). However, pLYL72 appeared not to transfer at a detectable level from Bacteroides to Bacteroides (<10-9). The reason for this was not known. In light of results presented in this work, in which a region of CTnDOT shown to enhance pLYL72 transfer has been identified, the ability of pLYL72 to transfer from a Bacteroides donor to a Bacteroides recipient (BT4100) was reassessed. In the presence of the central regulatory region alone [BT4001QABC(pLYL72)], a 10- to 100-fold increase in the transfer frequency of pLYL72 was observed (Table 4). This suggests that for conjugal transfer between Bacteroides strains, the central regulatory region alone is able to elicit a positive effect that we were not able to detect in Bacteroides-to-E. coli matings. In the presence of sequences known to enhance pLYL72 transfer from Bacteroides to E. coli [BT4001QABC(pLYL72, pKSO7)], transfer of pLYL72 was increased at least 1,000- to 10,000-fold (Table 4). These results suggest that the enhancement of transfer is important for transfer between Bacteroides strains.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although these two factors have been localized to the same region of the CTn, they are clearly separate from each other. The enhancer is induced in the presence of tetracycline, is transcribed from a promoter region located upstream of orf2C, and includes Exc plus another trans factor or factors encoded within a 1.0-kb region upstream of orf3. In contrast, the repressor of transfer is expressed in the absence of tetracycline and encoded by a gene that is not dependent on the Exc promoter region for expression.
It is perhaps not surprising that the regulations of excision and transfer functions, both positive and negative, are linked, because these processes need to be tightly coordinated to ensure that the excision of the integrated CTn takes place before transfer. If transfer was able to occur prior to excision, this might, as in the case of E. coli strains containing an integrated F plasmid, result in the formation of Hfr-like strains of Bacteroides. This is because after the internal oriT of CTnDOT is nicked, the entire 6-Mb Bacteroides chromosome would have to be transferred before the part of the element attached to the 3' end of the oriT. Consequently, premature nicking of oriT could result in deletions of part of the CTn.
The gram-positive CTn Tn916 solves this coordination problem structurally, since excision and circularization of the transfer intermediate are required for the expression of transposon-encoded transfer functions (5). In Tn916, the expression of tra functions in the circular intermediate is due to transcription from a promoter that runs through the attachment site. For the larger Bacteroides CTns of the CTnDOT/CTnERL family, such a coordination strategy is not feasible. The transfer genes of CTnDOT are not located near an end of the transposon; in fact, the first transfer gene in the tra operon, traA, is 40 kb from the attachment site in the circular intermediate from which such a promoter would be derived (Fig. 1A). These results suggest that for the CTnDOT/CTnERL family of transposons, multiple regulatory circuits have been recruited to ensure that excision and transfer are coordinated appropriately.
One explanation for the apparent increase in the amount of transfer protein detected by Western blot analyses after induction with tetracycline might be that the accumulation is due to an increase in the copy number of circular forms of the CTn. Such a time-dependent accumulation of the circular form of the mobilizable transposon NBU1 was observed recently (33). However, in the transfer system [BT4001QABC(pLYL72)] used in these studies, the CTn is essentially already in a circular form, since the region being transferred is located on a plasmid. This plasmid contains a pB8-51 replicon that is maintained at a copy number of approximately 8 to 10 per copy of the Bacteroides chromosome (19).
Results also show that the enhancement of transfer is important for detection of pLYL72 transfer between Bacteroides strains. The frequency of transfer for a single copy of a CTnDOT/CTnERL element between Bacteroides strains is 10-5 to 10-6 transconjugants per recipient, the same frequency at which pLYL72 transferrs from Bacteroides to Bacteroides in the presence of the enhancer sequences. It should be noted, however, that an integrated CTn differs from pLYL72 in that transfer involves three steps. The element must excise, transfer, and integrate into the recipient chromosome. In contrast, transfer of pLYL72 involves only two steps, transfer and the establishment of a plasmid. With the omission of the excision and integration steps, one might expect that the multiple-copy plasmid pLYL72 would transfer at a higher frequency than the CTnDOT element. Our results show that this is not the case.
Also, the transfer frequency of pLYL72 from Bacteroides to E. coli is about 1,000- to 10,000-fold higher than the transfer frequency of pLYL72 between Bacteroides spp. There are many possible explanations for this phenomenon, one of which could be that the oriV (RSF1010-base) utilized in E. coli is not the same as the one utilized in Bacteroides (pB8-51). Consequently, the efficiency with which the plasmid can be established and the stability of the plasmid in the recipient strain may be influencing the observed transfer frequency.
The previous report that the protein encoded by exc had significant homology to proteins encoded by some conjugal plasmids and a CTn, Tn1549 (10), and to chromosomally encoded topoisomerases (6) is of significant interest. This is because there are several examples of topoisomerase alleles being implicated in the global modulation of gene expression by altering DNA topology (7, 8). Consequently, it is possible that exc, like other topoisomerase alleles, is able to enhance conjugal transfer and transfer protein production by altering local supercoiling. Hence, investigations are in progress to characterize the Exc protein to determine whether it is a topoisomerase and whether this topoisomerase activity is required for the enhancement of conjugal transfer and of transfer protein expression. The nature of the interaction between Exc and the other trans factors encoded near the exc promoter region is an area that we also intend to investigate in the future. Similarly, the mechanism by which the repressor factor modulates conjugal transfer will also be investigated.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grant AI 22383 from the National Institutes of Health.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Appl. Environ. Microbiol. | Infect. Immun. | Eukaryot. Cell |
|---|---|---|
| Mol. Cell. Biol. | J. Virol. | Microbiol. Mol. Biol. Rev. |
| ALL ASM JOURNALS |
