Conditions for gene disruption by homologous recombination of exogenous DNA into the Sulfolobus solfataricus genome

Strains

Electroporation of Sulfolobus solfataricus cells

Electroporation conditions were essentially as described by Schleper et al. (1992). PBL2025 cells (50 ml) were grown overnight to an OD₆₀₀ ~ 0.1–0.3 in tryptone/sucrose medium. Cells were collected by centrifugation and resuspended in 50 ml of cold 20 mM sucrose. From this point on, cells were kept constantly on ice. The washing step was repeated with 10 ml of cold 20 mM sucrose. Finally, cells were resuspended in 20 mM sucrose at approximately 10¹⁰ cells ml^–1. An aliquot (50 µl) of the cell suspension was used for each electroporation. Electroporation conditions were 1.5 kV, 25 µF, 400 W with 2-mm cuvettes (Thermotron) and a Genepulser II (Bio-Rad) electroporator. The various modifications of the treatments of the cells after the electroporation step are listed in Table 1. The best post-electroporation procedure was to resuspend cells in 1 ml of demineralized water, leave them briefly (1–3 min) on ice and then incubate them for 10 min at 75 °C. The cells were then transferred to 50 ml of prewarmed lactose minimal medium.

Methylation of plasmid DNA

Plasmid DNA was methylated with HaeIII methylase according to the manufacturer’s recommendations. After methylation, the DNA was purified with the GFX Purification Kit (GE Healthcare, the Netherlands).

Construction of the various plasmids

The up- and downstream flanking regions of Sso0120 were amplified by PCR from genomic DNA isolated from S. solfataricus P2. The upstream flanking region was amplified with primers UP-F-KpnI and UP-R-NcoI (5′-CGGTACCGTGCGTATTATCTACGTTA-3′ and 5′-CCCCCATGGCAGTGTTTATTTAAAGAA-3′ respectively). The PCR product was digested with KpnI and NcoI and ligated into pET2268, which contains the lacS cassette with its own promoter and terminator region (Figure 1). The downstream flanking region was amplified with primers DWN-F-BamHI and DWN-R-NotI (5′-CCCGGATCCGGAGAATATTCATGATAC-3′ and 5′-CCCCCCCCCGCGGCCGCCGAGTGCAAAGATACTTG-3′, respectively), subsequently digested with BamHI and NotI and ligated into pET2268 already containing the upstream flanking region yielding pSVA37. The pSVA50 plasmid was constructed by inserting an Sso3017-containing PCR product into the lacS cassette of pET2268. The primers, 3017-F (5′ -CCCCCCCATGGTCTCTTTAAACCAGACG-3′) and 3017-R (5′-CCATAGAGGTAATGGCCAATGATACATG-3′), contained an NcoI and MscI restriction site, respectively, and were used to digest both the PCR product and pET2268 prior to ligation. pSVA63 was constructed by inserting the same up- and downstream flanking regions as for pSVA37 into pSVA50.

PCR analysis of deletion mutants

We performed PCR on the Sso0120 deletion mutant with primers 120-F (5′-CACCAGTAGCCTTTACAG-3′) and 120-R (5′-TTTGTTCGATACGGCGGTTG-3′), yielding a product of 2100 bp with the host strain PBL2025 and a 2900 bp product upon successful integration of the lacS gene into the Sso0120 locus (see Figure 2). LacS specific primers were LacS-F (5′-GGGGGCCATGGACTCATTTCCAAATAGCTTTAGG-3′) and LacS-R (5′-GGGGGGATCCGTGCCTTAATGGCTTTACTGGAGGTACGC-3′). Primers for the detection of the successful integration of the knockout plasmid into the Sso0120 gene locus were 3017-R and M3 (5′-CCCCCGGATCCAATAGCTTTAGTTTCAACTTTATCACC-3′). For the analytical reactions, PCR Mastermix was employed (Promega) with 30 cycles of 30 s at 94 °C, 30 s at 52.5 °C and extension at 72 °C, where the duration depended on the specific PCR product. Genomic DNA for PCR analysis was isolated from 2 ml of S. solfataricus cells with the QuickPick SML gDNA Kit (Bionobile).

X-gal treatment of plates

The X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) was dissolved to 25 mg ml^–1 in dimethylformamide. Plates were sprayed with a solution containing 5 mg ml^–1 X-gal diluted in demineralized water, and subsequently incubated for 30 minutes at 78 °C. Finally, the blue colonies were scored.

Selection of gene disruption strains after electroporation of plasmid DNA

PBL2025 has been described as a suitable host for the directed deletion of target genes in Sulfolobus solfataricus (Schelert et al. 2004). The strain is a natural deletion mutant lacking a chromosomal region of about 50 genes (Sso3004–Sso3050). The missing genes include lacS that encodes a β-galactosidase that is essential for growth of S. solfataricus on lactose minimal medium and therefore can be used as a selective marker.

Exogenous DNA is introduced into S. solfataricus by means of electroporation, but to maximize the yield of integrants, the post-electroporation conditions must be optimized to prevent the accumulation of false positives that slowly grow on the selective media. For this optimization, genes Sso2684–Sso2681 were targeted. These genes encode putative pilin proteins (Zolghadr et al. 2007). To test different post-electroporational conditions, PBL2025 cells were prepared as described in the Materials and methods and electroporated with 300 ng of pSVA78 plasmid DNA (Table 2). This plasmid harbors the lacS gene with flanking regions of the Sso02684 and Sso2681 genes (upstream flanking region was 733 bp, downstream flanking region was 756 bp). Upon transformation of the cells, the deletion casette will integrate into the genome resulting in an insertion of the lacS gene into the Sso2684–Sso2681 structural genes. Cells with a successful integration grow with lactose as the sole energy and carbon source, exhibit β-galactosidase activity and show, by PCR analysis, an inserted lacS at the targeted gene locus. After the 10-min incubation at 75 °C, 100 µl of cells was directly spread on lactose minimal medium plates and incubated at 78 °C for 7 days, but no individual colonies were retrieved. A sample of 900 µl of the electroporated cells was transferred to 50 ml of liquid lactose minimal medium, including the control cells that were electroporated without plasmid DNA. After about twelve days, cells of all conditions (Table 1) reached an OD_600nm of 0.1 whereupon 3 ml of each culture was used to inoculate 50 ml of fresh lactose minimal medium. From this stage, no growth could be detected with the control cells, whereas plasmid electroporated cells reached an OD_600nm of about 0.4 after 7 to 14 days. Subsequently, cells were spread on lactose and tryptone plates and sprayed with 5 mg ml^–1 X-gal after 7 days of growth. As shown in Table 1, in only two of the tested conditions were blue colonies obtained, indicating the presence of lacS in these cells on either lactose or tryptone plates. These blue single colonies were inoculated in tryptone medium.

Genomic DNA of the cells growing on tryptone medium was isolated and subjected to PCR analysis. The PCR analysis with the lacS primers generally yielded the fragment of the expected size when performed with cells derived from the blue colonies. The PCR analysis with primers specific for the 3′- or 5′-end of lacS and the flanking region of the targeted gene yielded products of the expected size, indicating that recombination had occurred at the correct site. However, a PCR analysis with the gene specific primers mostly indicated that the target gene was still present in the analyzed strains. An example is shown in Figure 2, where PCR yielded the expected 2100 bp product for the undisrupted gene and the larger 2900 bp product including the lacS gene for the disruption mutant (Figure 2). This indicates that either the cultures are mixtures of the wild type and gene disruption strain even though single blue colonies were picked from the plates, or the cells contain more than one chromosome. To single out the deletion strain, cells were again plated on tryptone plates and single blue colonies were selected, and this was repeated until the PCR analysis no longer indicated the presence of the wild type allele. Usually two to three rounds of single colony picking and replating was necessary to isolate the deletion mutant strain.

Optimization of growth and electroporation conditions for strain selection

For growth of S. solfataricus cells to occur on sugar minimal medium, it is necessary to adapt the cells to the sugar of choice by pre-growth on the respective sugar in the presence of tryptone or casamino acids before transfer of the cells to sugar minimal medium. After electroporation, successful recombinants are selected on lactose minimal media, therefore it was assumed that it would be beneficial to grow the cells before electroporation on tryptone medium with a sugar that induces lactose metabolism. Although the presence of sugar was necessary, no difference in growth after electroporation was observed when either sucrose or lactose was added to the tryptone medium.

Previously, methylated DNA has been used for transformation in S. solfataricus (Worthington et al. 2003). We compared the yield of blue colonies after selection on lactose minimal medium using S. solfataricus cells electroporated with either methylated or non-methylated plasmid DNA. However, this did not affect blue colony yield, whether the amount of plasmid DNA used was 100, 300 or 500 ng. Cells electroporated with 300 ng of plasmid DNA started to grow first (after about 8 days), whereas electroporation with 500 and 100 ng resulted in slower growth, which extended the selection protocol by 2 and 5 days, respectively.

As it has been reported that S. acidocaldarius can be effectively transformed with oligonucleotides, we tested whether variations in the length of flanking regions of the targeted gene affect the recovery of integrants. Gene fragments from 500 to 1200 bp were used, but no differences in the yield of blue colonies were observed.

Co-transformation of the lactose transporter in recombinant strain selection

For S. solfataricus strain MT4, a natural mutant has been described that, like PBL2025, lacks a 13 kb fragment of genomic DNA, which includes lacS and the gene upstream that is homologous with Sso3017 (Bartolucci et al. 2003). The latter represents a lactose transporter which, together with lacS, is needed to restore growth of the natural selected mutant on lactose selective medium (Bartolucci et al. 2003). To determine whether the co-introduction of the transport gene Sso3017 together with the lacS gene improves the recombination yield in PBL2025, plasmid pSVA50 was constructed containing both genes under the control of their natural promoters. Based on this plasmid, gene disruption plasmids were prepared for gene Sso0120, which encodes a secretion ATPase (Albers and Driessen 2005), that contained either only the lacS gene (pSVA37) or both the transporter and the lacS gene (pSVA63) (Figure 1). After transformation of both plasmids into PBL2025 cells, no significant advantage of the presence of the transporter was noticed with the recovery of transformed cells on lactose minimal medium and subsequent blue-white screening with the indicator X-gal on tryptone plates. However, blue colonies picked from cells containing pSVA63 were able to grow in 3 days on liquid lactose minimal medium, whereas only slow growth was observed in cells containing the deletion plasmid with only lacS as a marker. The PCR analysis showed that the blue colonies of pSVA37 transformed cells contained both lacS and a PCR product showing the insertion of lacS at the site of the target gene (Figure 3). However, mixed cultures were also obtained (Figure 2). All cultures from colonies containing pSVA63 showed a product for lacS and the target gene, but were negative for the PCR product of the flanking region indicating correct insertion (see Figure 3). This indicated that in, all these cases, the lacS-transporter cassette integrated at another site in the genome. Therefore the co-introduction of the transporter did not improve the selection procedure for deletion mutants on lactose minimal medium.