![[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}
Danish Archaea Centre, Institute for Molecular Biology and Physiology, Copenhagen University, Sølvgade 83H, DK-1307 Copenhagen K, Denmark
Received 13 January 2006/ Accepted 26 March 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
80°C and pH 2 to 3), have provided many new insights into the basic molecular processes of both crenarchaea and the Archaea in general. One of the organisms that has received particular attention is Sulfolobus solfataricus P2, for which the genome sequence is available (24). The genome is exceptional in that it contains a large number of potentially mobile elements, including over 200 copies of intact insertion sequence (IS) elements of at least 25 different types and more than 140 copies of miniature inverted-repeat transposable elements (MITEs), which have been grouped into four main classes (SM1 to SM4), as well as numerous fragmented elements (4, 5, 22, 24). Some types of IS elements and MITEs are present in multiple copies of identical or near-identical sequence, consistent with their having recently transposed within the genome (4, 22). In total, they constitute more than 11% of the 3.0-Mb S. solfataricus genome and are more prevalent than in any other sequenced archaeal or bacterial genome (4, 7). Each of the IS elements of S. solfataricus P2 encodes a transposase, and many carry inverted terminal repeats to which the transposase attaches, while MITEs lack a transposase gene but exhibit inverted terminal repeats similar to those of an IS element (4, 22). The latter similarity is probably sufficient to ensure that the MITEs are mobilized by the corresponding IS element-encoded transposase. However, the internal sequences of the S. solfataricus P2 MITEs (type II) are different from those of the corresponding IS elements, suggesting that they evolved convergently (14, 21, 22).
The detailed transpositional mechanisms of the Sulfolobus elements, and those from other archaea, have received little attention experimentally, and they have generally been considered to be similar to those of their bacterial or eukaryal counterparts (4, 17). For the last two, transposition generally involves a double-strand excision from a given genomic position, with a subsequent nucleophilic attack by both 3' ends at another target site, which results in the insertion of the element, all catalyzed by the encoded transposase, which binds as a multimer to the inverted terminal repeats. Attack on the two DNA strands often occurs a few base pairs apart and thereby produces a direct repeat (DR), constituting a short (2- to 14-bp) duplication of the target site, which borders the newly inserted element (17).
Genome comparison studies have suggested that S. solfataricus P2 readily undergoes genomic rearrangements (5). Moreover, since it carries substantially more IS elements and MITEs than Sulfolobus tokodaii, or than Sulfolobus acidocaldarius, which probably lacks active mobile elements (4, 9, 11), it was proposed that the genomic rearrangements are driven primarily by mobile-element-dependent mechanisms (5). This proposal complements the earlier hypothesis that the megaplasmid pNRC100 of the euryarchaeon Halobacterium halobium arose via a multistep IS element-mediated process (10, 20).
The circumstantial evidence for high transpositional activity, and the possibility of mobile-element-induced genomic rearrangements occurring in S. solfataricus P2, raises the question as to how the organism remains viable. The complexity of this issue is compounded by the observation that a large number of IS element fragments are present in the genome, including some element types which are not present as intact copies. This suggests that they are difficult to delete cleanly (3, 4). However, although they are hard to remove, it was recently proposed that expression of the transposases is rigorously regulated via antisense RNAs (26), and several small putative antisense RNAs were detected in cell extracts of S. solfataricus using different experimental approaches (26, 27). Such a regulatory mechanism has also been shown to operate for some bacterial IS elements of the IS4 family, to which some of the Sulfolobus elements also belong (4, 25).
In order to gain a clearer picture of the transpositional events and mobile-element-induced genomic rearrangements, as well as other types of mutational change that occur in S. solfataricus P2, we performed a comprehensive mutational study of the pyrEF locus under selective conditions. Mutants were generated which were resistant to 5-fluoroorotic acid (FOA), which is converted into the toxic compound 5-fluoro-UMP by the gene products of pyrE and pyrF (16). Earlier studies, using a similar approach, detected only a few IS element insertions, mainly in the pyrE gene or the promoter region for the related strain Sulfolobus solfataricus P1 (18), while for the more distantly related S. acidocaldarius, only point mutations, small insertions, and deletions were observed at this locus (11).
The large number of spontaneous knockout mutants isolated for S. solfataricus P2 were analyzed using a combination of PCR amplification and sequencing. For those mutations which involved transpositions, the donor site of the transposed element was always examined. Furthermore, the locations and sizes of genomic rearrangements were identified using a method developed specifically for determining the sequences beyond the junctions of the rearranged region.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
S. solfataricus P2 was grown at 80°C in liquid culture for 2 days to an optical density at 600 nm (OD600) of 0.19 to 0.67 in medium supplemented with 2 g/liter Tryptone Peptone (Difco, Sparks, MD). The culture was then diluted to an OD600 of 0.0004 to 0.005 with fresh liquid medium, and 400-µl aliquots were spread onto Gelrite plates (8 g/liter; Kelco, San Diego, CA) containing the Sulfolobus medium and 1 mM MgCl2, 0.3 mM Ca(NO3)2, 50 mg/liter 5-fluoroorotic acid (Apollo, Whaley Bridge, United Kingdom), and 20 mg/liter uracil (Sigma, St. Louis, MO). Control samples were diluted to an OD600 of 2 x 10–7 to 1 x 10–6 and plated onto nonselective Gelrite plates. After 6 to 8 days at 75°C, the colonies were counted, and from each selective plate, 30 to 60 were streaked onto selective solid medium. After 2 to 3 days, the surviving strains were transferred to 5 ml selective medium and grown for 2 days at 80°C. Strains that failed to grow were discarded. One milliliter of the final culture was stored at –80°C, and 20 µl of the liquid culture was added to 400 µl buffer (10 mM Tris-HCl, pH 8, 1 mM EDTA) and mixed thoroughly on a vortex before being used as a template for PCR amplification.
PCR amplification. Primers were designed with pDRAW32 (ACACLONE Software) and Oligo Analyzer 1.0.2 (Teemu Kuulasma; see the primer list in the supplemental materials; software is available at http://molbiol-tools.ca/molecular_biology_freeware.htm). Fifteen microliters of reaction mixture contained 1 µM of each primer, 2 µl template, 1.5 µl 10x ThermoPol buffer (New England Biolabs), 1 mM deoxynucleoside triphosphates, 1.25 mM MgCl2, and 1 unit Taq DNA polymerase (New England Biolabs). Reactions were generally performed in a TRIO-thermoblock (Biometra, Goettingen, Germany) for 5 min at 96°C; 45 s at 96°C, 45 s at 50°C, and 4 min at 72°C for 35 cycles; and 5 min at 72°C, and then maintained at 4°C. For pyrEF locus amplification, four sets of PCR primers were used: Pf and Pr, Ef and Er, F1f and F1r, and F2f and F2r (Fig. 1).
|
DNA sequencing. Most sequencing reactions consisted of 4 µl DYEnamic ET Dye Terminator reagent premix (Amersham Biosciences, Little Chalfont, United Kingdom), 1 µl 10 pM primer, and 5 µl purified PCR product. The Peltier Thermal Cycler PTC-225 (MJ Research) was set to 2 min at 94°C; 30 s at 93°C, 30 s at 50°C, 1 min at 60°C for 30 cycles; then 30 s at 93°C, 30 min at 50°C, and 2 s at 60°C; and 10 min at 60°C, and then maintained at 4°C. The sequence reaction products were then purified with Sephadex G50 and sequenced with a MegaBACE 1000 (Amersham Biosciences). The remaining sequences were performed on an ABI Prism 310 Genetic Analyzer (Applied Biosystems, Foster City, CA), where 10 µl of sequencing reaction mixture consisted of 3 to 5 µl purified PCR product, 1.6 pM primer, and a 2-µl BigDye Terminator v. 1.1 cycle-sequencing kit (Applied Biosystems). The reaction was run on a TRIO-thermoblock (Biometra) (30 s at 96°C, 15 s at 50°C, 4 min at 60°C for 25 cycles, and then maintained at 4°C), after which it was ethanol precipitated and redissolved in 12.5 µl Template Suppressing Reagent (Applied Biosystems). The sequences were analyzed with Sequencher (Gene Codes, Ann Arbor, MI); BLAST searches used the Sulfolobus Database (http://dac.molbio.ku.dk/dbs/Sulfolobus/cbin/mutagen.pl), and sequence visualizations and manipulations were carried out with pDRAW32 (ACACLONE Software).
"In vitro library" with NlaIII. Genomic DNA was isolated with a DNeasy Tissue Kit (QIAGEN) using the bacterial protocol. A palindromic 42-bp library template oligonucleotide (300 pmol) in 20 µl H2O was heated to 98°C for 5 min and then cooled slowly to room temperature. To this was added 20 units NlaIII, 1x NEB4 restriction buffer, 0.01% bovine serum albumin (all from New England Biolabs), and 2–16 mol (400 ng) genomic DNA (corresponding to approximately 2.6 x 10–12 mol cut 3' overhangs) in a 60-µl final volume, which was maintained for 3 h at 37°C. NlaIII was heat inactivated for 20 min at 65°C; 3 µl 20 mM ATP and 1,000 units T4 DNA ligase (New England Biolabs) were added and incubated at 16°C overnight. The ligase was heat inactivated for 25 min at 65°C, and a QIAquick PCR purification kit (QIAGEN) (wash, two times with 700 µl; elution, 50 µl elution buffer after 1 min of incubation) was used to remove excess cut (and uncut) oligonucleotides, as well as ligations to small NlaIII fragments. The library was used as a template in 30 cycles of PCR (50°C annealing) with 1 µM library primer, together with 1 µM homing primer, and the relevant bands were excised from an agarose gel and sequenced using the homing primer. The library template was as follows (the NlaIII site is underlined): CGCCCGTCCGCTCCTGTCCCATGGGACAGGAGCGGACGGGCG; the library primer was as follows: CGCCCGTCCGCTCCTGTCC.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pyrEF locus was examined by PCR for 130 FOA-resistant mutants. Four primer sets were designed to cover a region extending from 371 bp upstream of the pyrE start codon to 14 bp downstream from the pyrF stop codon (genome positions 535561 to 537188) (Fig. 1). If any of the E, F1, or F2 PCR products (Fig. 1) migrated differently from the control PCR product on an agarose gel, they were sequenced. When no size change was detected, all three PCR products were sequenced, and if no mutation was detected, then the P PCR product, covering the sequence upstream of the promoter region, was generated and sequenced (Fig. 1). The identified mutations are presented schematically (Fig. 1) and summarized (Tables 1 and 2). (Details of each mutant are given in the supplemental materials.)
|
|
|
IS element-based mutations. About two-thirds of all the mutations were caused by IS element insertions, and their locations and properties are summarized in Fig. 1 and Table 3. Many of them were caused by the IS element ISC1058, and the insertions of this element were concentrated at three different sites, 9 in pyrE and 20 and 37 insertions at sites separated by only 2 bp within the promoter region (Fig. 1). At least 24 of these insertions did not correspond to preselection mutations because they (i) were isolated in different experiments, (ii) occurred in both orientations, and (iii) derived from different donor IS element copies (see below and Table 3). Similarly, for ISC1225, all three observed insertions occurred at one position, but they (i) originated from different donor copies, (ii) were present in different orientations, and (iii) exhibited two different DR sizes (Table 3). Since this target site shows sequence similarity to the inverted repeats of ISC1225 (Fig. 2B), the conserved sequence probably facilitates transposase recognition. A broader target specificity was observed for the seven ISC1439 mutants, but the six different target sites (Fig. 1) still exhibited high pairwise sequence similarity (Table 3).
|
First, the ISC1359 copy at position 1300529 was the only exact match to elements inserted in two of the mutants (strains P2A-003 and -102). However, this putative donor was not PCR amplifiable from either the mutant strains or the parent P2 strain. Further investigation showed that a large region, constituting about 4% of the published genome and extending from positions 1201319 to 1325406, was absent. In the published genome sequence, the region is bordered by two almost identical copies of ISC1439, which were reduced to one copy in the parent strain. Therefore, it was inferred that a deletion had occurred as a result of an intramolecular recombination between the two ISC1439s (Fig. 3A). Since the large deletion demonstrated a radical difference between the S. solfataricus P2 stock used for genome sequencing and that used for mutant isolation, we refer subsequently to the latter as strain P2A.
|
Second, the ISC1439 copies inserted into mutant strains P2A-121, -147, and -165 were identical, but this sequence was absent from the published genome sequence. Therefore, we assumed that either a more recent mutation had occurred in a genomic ISC1439 copy or there was an error in the genome sequence. In order to locate the donor element, one end (about 500 bp) was PCR amplified and sequenced for 22 of the 33 genomic copies of ISC1439 that exhibited the sequences most similar to the element inserted in the three mutants. None of the sequences obtained matched perfectly, and moreover, two of the 22 copies and their DRs were absent from the mutant and parent P2A strains (Fig. 2C). It was then discovered that the interrupted ISC1439 copy at position 454065 and the full-length copy at 480919 had apparently recombined (or had possibly been misassembled in the genome sequence) to produce two different ISC1439 copies, one of which was identical to the donor copy for all three mutants. This was confirmed by sequencing PCR products generated across the junctions of the rearranged region (Fig. 3B).
The dynamic complexity of the S. solfataricus P2 genome was further emphasized by the finding that one of the three identical copies of the potential donor for the inserted ISC1078 in strain P2A-009 was absent from S. solfataricus P2A, as were two of the three possible donors for ISC1078 in strain P2A-110. Sequencing revealed the absence of 2-bp DRs at the three expected ISC1078 sites (positions 1088787, 1903184, 1959047), and we infer, therefore, that the P2A strain never carried ISC1078 copies at these positions (Fig. 2C). Thus, the inserted copies of ISC1078 must have originated from the copies at position 35310 or 2734629 for the P2A-009 strain and at 1711152 for the P2A-110 strain (Table 3).
In conclusion, the donor for each IS element inserted into the pyrEF region could be accounted for, and each one was maintained at its original position in the strain P2A genome.
Genomic rearrangement at the pyrEF locus. In a single strain, P2A-003, no PCR products were obtained across the pyrEF region with either Pf plus Pr or Ef plus Er primer pairs (Fig. 1). PCR products of parental-strain length were obtained when primer pairs were annealed upstream of position 535728 or downstream of position 536105, but a combination of these primers yielded no product, and therefore, one could exclude a simple deletion of the priming sites (Fig. 1 and 3C). It was inferred that a genomic rearrangement had occurred, and by using the novel "in vitro library" strategy, sequences across the rearrangement junctions were obtained (see Materials and Methods). The method was developed to determine the sequence of an unknown region of a genome, with the only requirement being that it lie adjacent to a known sequence (in this case, the upstream region of pyrE). The genome was digested with NlaIII, and a 19-bp synthetic "library template" was then ligated to each cohesive end, thereby tying a defined 19 bp to the restriction site (Fig. 4). PCR with a "library primer" annealed to the library template and a "homing primer" annealed (homed in) to the known sequence then amplified the DNA between the homing primer and the closest NlaIII restriction site (Fig. 4). Sequencing of the product using the homing primer then yielded the sequence of the unknown region (Fig. 4).
|
5-kb inversion that exhibited a copy of ISC1359 at the "downstream" end and a 9-bp deletion (positions 535919 to 535927) in the "upstream" region (Fig. 3C). Undetected mutations. Nineteen mutant strains showed no sequence changes within the pyrEF locus for any of the PCR products (Fig. 1 and Table 1). Thus, the phenotype must result from a mutation in another genomic region. Each of these strains was streaked onto selective medium to confirm its mutant phenotype. In contrast to the strain P2A control, each of them grew and exhibited an FOA-resistant phenotype (data not shown). Since all 19 strains were from the same isolation experiment, one cannot readily estimate the mutation frequency. However, if they are independent, then the high number suggests a gene-size target, possibly that of an unidentified orotate transporter.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The identified nontranspositional mutations show some similarities to mutations observed in S. acidocaldarius, which probably lacks active mobile elements (9). For example, the 44-bp deletion had no recognizable direct repeats, inverted repeats, or secondary DNA structure at the endpoints, consistent with the deletions observed in S. acidocaldarius (12). Moreover, we observed two ±1-bp mutations in runs of single bases (T3 and A6 in strains P2A-045 and -139, respectively) (Table 2), although there was no strong preference for –1 bp mutations in long runs of the same base, as occurs in S. acidocaldarius, which is richer in homopolynucleotide sequences (11). However, in contrast to the pyrE bias observed for mutations in the pyrEF region of S. acidocaldarius (11), mutations were distributed fairly evenly throughout the region (Table 2).
Transposition of mobile elements. Although there is some bias, especially for ISC1058, in the frequency of transposition, the low transpositional activities observed for some classes of IS element may partly reflect the limited availability of sequence-specific sites for transposition in the pyrEF region. However, the discovery of putative antisense RNAs complementary to transposase mRNAs in S. solfataricus (26, 27) may also contribute to the relatively low number of transpositions observed for ISC1217, ISC1225, ISC1359, and ISC1439 (Table 1).
Surprisingly, given the large number of MITEs present in the strain P2 chromosome (140 copies), only one MITE transposition was observed. This MITE was predicted to be mobilized by the transposase of ISC1217, which was also transposed in this study, and both recognized the shared target site aanHHcgDDwtt (capital letters indicate a higher degree of conservation) (Table 3) (22). Although half of the mutants carried ISC1058 insertions, none were observed for the multiple-copy SM3 MITEs, which are considered to be mobilized by the ISC1058-encoded transposase. Possibly, as has been proposed for SM4, SM3 is an older, more degenerate element in strain P2, at least within its inverted terminal repeat, in contrast to SM1 and SM2 (22). Nevertheless, a Sulfolobus MITE which was previously shown to be mobile in Sulfolobus strain Y99 9-19 was proposed to be mobilized by the transposase of ISC1057, a close relative of ISC1058 (3). Although no genome data are available for this strain, it reinforces the fact that the SM3-like elements can be active.
Each of the mobile-element insertions observed in the pyrEF locus could be correlated with a donor element within the genome of S. solfataricus P2 (24). However, during our sequence analyses, we discovered examples of multicopy elements that are present in the published strain P2 genome sequence but were absent from our parent strain P2A stock. These included three copies of ISC1078 and two copies of ISC1439 elements and their DRs. The most likely explanation is that during assembly of the S. solfataricus P2 genome, occasionally overlapping clones occurred, with or without an IS element, and if most clones contained the element, it was generally assembled into the genome sequence (24; Q. She, personal communication). Thus, strain P2A corresponds to the clones that lacked these IS elements.
Each transposition event observed in this study, except the second transposition in strain P2A-003 (see below), resulted in an increased copy number of the element. The mechanism remains unclear. Bacterial-type replicative transposition can be ruled out, as it involves the donor-carrying replicon inserting between the two daughter IS elements (17), and this was not observed in the present study, where each inserted element was flanked by pyrEF sequences. Moreover, an earlier bacterial study demonstrated that persistent genomic proliferation of nonreplicative IS10 elements did not occur by "cut and paste" from one replicon to another with the subsequent loss of the donor replicon (1). They demonstrated instead that proliferation was most likely to occur by one of two possible mechanisms involving homologous repair: (i) when transposition occurs coincidentally with DNA replication and formation of hemimethylated DNA or (ii) when two or more sister chromosomes coexist postreplication (1, 19).
For Sulfolobus species, which exhibit a G2 phase constituting about 60% of the cell cycle (2), solution ii is the more likely, involving homologous repair of the donor chromosome using the sister chromosome as a template. However, to distinguish definitively between these mechanisms would require more sophisticated genetic tools than are currently available for Sulfolobus.
Genomic rearrangements. A model is proposed for the mechanism of genomic rearrangement observed in strain P2A-003 (Fig. 3C and 5). It involves two transposition steps. In the first, ISC1359 is inserted at position 535928 just upstream from pyrE, as also occurs in strains P2A-102 and -150 (Fig. 5A and B). In the second step, the multimeric ISC1359-encoded transposase binds to one terminal repeat of the inserted ISC1359 and to another similar sequence located just upstream of the 9-bp segment that is absent from strain P2A-003 (Fig. 5C). It then catalyzes excision of the genome as a giant mobile element. This results in the loss of the 9-bp segment, as well as one of the 4-bp DRs, from the first ISC1359 insertion. The genome-size mobile element then inserts into itself at position 539558, just 5 kb from the excision site, creating a second DR (Fig. 5C). This mechanism produces the sequence found in strain P2A-003 and accounts for the location of ISC1359 and the inversion event, as well as the absence of the 9-bp segment (Fig. 5D).
|
Two other genomic rearrangements were discovered during this study. One was probably caused by homologous recombination between two almost identical copies of ISC1439 in the same orientation, at positions 1201319 and 1325406, that produced a 124-kb deletion that included the ISC1359 copy at position 1300529 (Fig. 3A). The deletion was traced back to the plating of a culture of S. solfataricus P2 from which two single colonies, A and B, were grown for retention as stock at –80°C (Q. She, personal communication). Colony A had this deletion, whereas colony B did not, indicating that rearrangements occurred under normal laboratory conditions with no known selection. The other 25-kb inversion (Fig. 3B) was present in all S. solfataricus P2 strains available in our laboratory, all of which originated from the Deutsche Sammlung von Mikroorganismen und Zellkulturen, in contrast to the strain P2 used for genome sequencing, which was obtained from Wolfram Zillig.
Conclusion. The results show that IS elements can facilitate large genomic changes in Sulfolobus both by transpositional mechanisms catalyzed by their transposases and by providing similar sequences for homologous recombination. Moreover, the copy number increase by transposition observed for both IS elements and MITEs continually creates new sites for potential rearrangement and change. Thus, it is extremely important to use mutation-minimizing strategies, such as purifying strains through plating and avoiding long-term cultivation, when working with S. solfataricus. Although in a natural environment this genomic variability may be a disadvantage for an individual cell, it does not seem to be a disadvantage for the cellular population, albeit a continually changing one. This is illustrated by the fact that S. solfataricus strains, and the closely related Sulfolobus islandicus, which is also rich in mobile elements (K. Brügger, personal communication), occur widely in natural environments.
|
ACKNOWLEDGMENTS |
|---|
P.R. was funded by a Ph.D. scholarship from the Faculty of Science, Copenhagen University. The research was supported by a grant from the Danish Research Council for Natural Science.
|
FOOTNOTES |
|---|
Supplemental material for this article may be found at http://jb.asm.org/.
|
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 |
