Complete Genome Sequence of Escherichia coli BL21-AI.

Escherichia coli BL21-AI is a commercially available strain possessing a phage T7-based protein-expression system. A combination of tight regulation and high yield makes it widely used for high-level expression of toxic recombinant proteins. Here, we present the complete genome sequence of BL21-AI and provide insights on its genome.

ANNOUNCEMENT

Escherichia coli B is widely used for the induced expression of recombinant proteins. It lacks certain proteases and has an additional type II secretion system and different cell wall and outer membrane composition, making it more favorable for protein secretion (1). Examples of commonly used E. coli B strains are BL21 and BL21(DE3) (2). However, the leaky nature of these expression systems makes them less suitable for the induction of deleterious or toxic proteins. One strain developed specifically to combat leaky expression is E. coli BL21-AI. In this strain, expression of plasmid-borne genes is driven by a chromosomally encoded bacteriophage T7 RNA polymerase (T7-RNAP), which in turn is controlled by the tightly regulated arabinose-inducible P promoter (3, 4). The strain is produced by Invitrogen (Thermo Fisher Scientific), but its precise construction has not been disclosed due to proprietary considerations. Nevertheless, it is derived from strain BL21, and the two genome sequences are expected to be similar except for araB, the insertion locus of the T7-RNAP cassette and linked tetracycline-resistance marker. To aid in future investigations using BL21-AI, including the construction of related strains, we deemed it necessary to determine its exact genome sequence. Here, we describe the complete genome sequence of BL21-AI and compare it with that of BL21 (GenBank accession number CP010816).

BL21-AI competent cells were purchased from Invitrogen (catalogue number C607003, lot number 2067410) and streaked onto Luria agar plates containing tetracycline (15 μg/ml). Following incubation at 37°C for 16 h, two single colonies were used to inoculate LB liquid and were grown to saturation. Genomic DNA was extracted using Qiagen Genomic-tip 100/G, and whole-genome sequencing was performed at the National Human Genome Research Institute of the NIH. PacBio barcoded libraries were generated (SMRTbell Express template prep kit v2.0, Sequel II 16-overhang adapter barcode kit), pooled, and run on a PacBio Sequel II single-molecule real-time (SMRT) cell for 20 h using the continuous long-read sequencing mode. Demultiplexed subreads (825,284 and 931,154 reads with mean lengths of 7,908 bp and 8,354 bp, respectively, for two isolates) were assembled into a single contig with the PacBio SMRTLink v7.0 HGAP4 pipeline (5). Quality control of the sequence reads was performed using the default input filtering parameters in the pipeline. The contig was circularized using Circlator v1.5.3 (6) (“all” mode, “threads 8”) followed by two rounds of consensus polishing with the SMRTLink v7.0 resequencing pipeline (https://github.com/PacificBiosciences/GenomicConsensus). Two independent colonies of BL21-AI were processed in parallel, yielding identical genome sequences with average coverage depths of 1,341× and 1,600×, respectively. Genome annotation was carried out using the NCBI Prokaryotic Genome Annotation Pipeline (PGAP) v4.10 (7). Default parameters were used for all software unless otherwise specified.

The genome of BL21-AI is 4,530,564 bp long with a G+C content of 50.8%. Aligning the sequence (using the progressiveMauve algorithm) (8) with BL21 showed one large block of correspondence, confirming the overall relatedness of the two genomes, with the exception of a 4,084-bp deletion/insertion segment at BL21-AI coordinates 71213 to 75296. This region of nonconformity results from deletion of the BL21 araB gene (BL21 coordinates 71213 to 72853) and replacement by the T7-RNAP gene along with a tetA cassette, accounting for the observed tetracycline resistance of BL21-AI. In addition, the Mauve alignment revealed a total of 231 additional single-base substitutions (BL21-AI 68837 to 88634) flanking either side of the ara-tet-T7pol insert. These substitutions match the corresponding sequence of E. coli K-12 strain MG1655 (GenBank accession number U00096.3) spanning a total of 14 genes from araD to leuA, with the exception of a G→A mutation in sgrR (coordinate 81500) changing a Gln codon (CAA) to a termination codon (TAA). Overall, we infer that to generate BL21-AI from BL21, a ΔaraB::T7RNApol-tetA cassette was transferred from a K-12 strain by P1 transduction, resulting in the transfer of a surrounding ∼20-kb chromosomal donor DNA segment.

Data availability.

The complete genome sequence of Escherichia coli BL21-AI has been deposited at GenBank under accession number CP047231. The raw sequence data are available under SRA accession number SRR10851446.