Genome Sequence of Zymomonas mobilis subsp. mobilis NRRL B-1960

ABSTRACT Zymomonas mobilis subsp. mobilis is an efficient ethanol producer with application for industrial production of biofuel. To supplement existing Z. mobilis genomic resources and to facilitate genomic research, we used Oxford Nanopore and Illumina sequencing to assemble the complete genome of the beer spoilage isolate Z. mobilis subsp. mobilis strain NRRL B-1960.

Z ymomonas mobilis subsp. mobilis is a Gram-negative alphaproteobacterium capable of highly efficient ethanol production through the Entner-Doudroff pathway (1). Z. mobilis has garnered interest for its prospective biofuel application, and it is utilized in industrial-scale fermentations to produce bionic acid, levan, and sorbitol (2,3). The genomes of several natural isolates of Z. mobilis from North America, South America, Europe, and Africa have been previously sequenced (4)(5)(6)(7)(8). Here, we sequenced the genome of Zymomonas mobilis subsp. mobilis NRRL B-1960 (NCIMB 8227), originally isolated from "bad beer" in the United Kingdom (1), to facilitate population and comparative genomics research of this industrially important species.
Z. mobilis NRRL B-1960 was grown in 0.5% yeast extract and 2% glucose liquid medium at 30°C for 24 h. DNA was extracted using the PureLink microbiome DNA purification kit (Thermo Fisher Scientific), and DNA integrity was evaluated via gel electrophoresis. We constructed an Oxford Nanopore (ONT) rapid 1D library from 500 ng of DNA, according to the manufacturer's instructions (kit version RAD002) and performed sequencing on two flow cells (R9 version) of the MinION device. Poretools version 0.6.0 (9) was used to convert ONT data from fast5 to fastq format. We performed error correction of ONT reads using Canu version 1.4 (10) and generated a total of 12,302 corrected reads, with an average read length of 4,152 bp, representing~25ϫ coverage. A paired-end 101-bp Illumina library was constructed and sequenced at Macrogen (Rockville, MD) from 1 g of DNA. Illumina sequence data were deduplicated using Tally (11), adapter and quality trimmed using Trim Galore version 0.4.3 (https:// www.bioinformatics.babraham.ac.uk/projects/trim_galore/), and errors corrected using SPAdes version 3.10.0 (12), resulting in 14,838,803 paired-end reads, representing 1,400ϫ coverage. We incorporated both sequence data types in SPAdes to produce an assembly using k-mer sizes of 27, 37, 47, 57, 67, 77, 87, and 97 (12). Last, preassembly-improved Illumina data was used for genome polishing via Pilon version 1.21 (13).
We assembled the genome into a single scaffold consisting of 2,045,798 bp, with a GC content of 46.09%. Two additional scaffolds were assembled which correspond to plasmid sequences. The larger plasmid sequence, pZMO1960_1, is 34,463 bp, with a GC content of 45.77%, and has an~5.5-kb region nearly identical to a Z. mobilis CP4 plasmid (GenBank accession no. CP006894) (6). The other plasmid sequence, pZMO1960_1A, is 1,743 bp, with a GC content of 38.15%, and it is identical to the Z. mobilis NCIMB 11163 plasmid pZMO1A (GenBank accession no. GQ293074) (5). The RAST server (14, 15) was used for genome annotation, and clustered regularly inter-spaced short palindromic repeat (CRISPR) arrays were predicted using CRISPRFinder (16). The genome contains 1,971 protein-coding genes, 3 rRNA gene loci, 51 tRNAencoding genes, and 3 CRISPR arrays.
Phylogenetic analysis of 74,852 polymorphic sites across the sequenced natural isolates of Z. mobilis indicates that the NRRL B-1960 genome is most closely related to the NCIMB 11163 strain, which was isolated from spoiled beer in England (5). We conservatively identified 11 genes unique to the NRRL B-1960 genome, including 2 TonB-dependent receptor-encoding genes, via BLAST (17) searches of the NRRL B-1960 genes against the 5 sequenced Z. mobilis genomes (4)(5)(6)(7)(8).
Accession number(s). The Zymomonas mobilis subsp. mobilis NRRL B-1960 genome was assigned GenBank accession number CP021053 for the chromosome and accession numbers CP021791 and CP021792 for the plasmids.

ACKNOWLEDGMENTS
We thank the Clark University Biology Department for funding in support of this project, James Swezey of the U.S. Department of Agriculture for the Z. mobilis NRRL B-1960 isolate, and Matt Essig, Aaron Bennett, Arnab Banik, and Juan Park for their maintenance of Clark University's High Performance Computing Cluster and their assistance with software installation. This work was conducted in part using the Clark University High Performance Computing Cluster.