Genome Sequences of Hemolytic and Nonhemolytic Listeria innocua Strains from Human, Food, and Environmental Sources

ABSTRACT

This report describes genome sequences for nine Listeria innocua strains that varied in hemolytic phenotypes on sheep blood agar. All strains were sequenced using Pacific Biosciences (PacBio) single-molecule real-time (SMRT) chemistry; overall, the average read length of these sequences was 2,869,880 bp, with an average GC content of 37%.

ANNOUNCEMENT

The Listeria genus represents a diverse group of Gram-positive organisms that are commonly found in environmental reservoirs. Public health focuses on Listeria monocytogenes, which can cause fatal infections in vulnerable human populations (1, 2). Unlike L. monocytogenes, Listeria innocua typically is nonpathogenic but in rare cases has been linked to human infections (3, 4). Genotypic characterization has identified L. innocua as a close relative of L. monocytogenes (2, 4–6). Although L. innocua is typically nonhemolytic, some isolates within the species exhibit β-hemolysis on blood agar (2, 5, 7, 8). These atypical strains also share virulence factors with L. monocytogenes (2, 5, 7). A genomic assessment was performed with nine L. innocua strains that had been isolated from various sources and were archived at the National Listeria Reference Laboratory at the Centers for Disease Control and Prevention (CDC). Here, we present Pacific Biosciences (PacBio) sequences of nine L. innocua isolates that were received between 1987 and 2015.

Genomic DNA was extracted from single-colony isolates using the DNeasy blood and tissue kit (Qiagen, Germany) after overnight growth on Trypticase soy agar II with 5% sheep blood (BD, Germany) at 36°C. Samples were assessed for quality (5 to 10 μg DNA in 100 μL buffer) for submission for PacBio sequencing. The Biotechnology Core Facility Branch at the CDC performed sequencing and raw read analysis.

Genomic DNA was sheared to 10 kb or 20 kb using needle shearing. Plasmids were identified as contigs of <0.5 MB, consistent with standard threshold cutoff values. Contigs from strains H0996, 2010L-1951, and 2010L-2059 were linear; all other contigs were circularized. Default parameters were used for all software unless otherwise specified. Libraries were generated with the SMRTbell template preparation kit v1.0 (PacBio, Menlo Park, CA). The 2012L-5520 and 2015L-6726 sample libraries were size selected to 8 kb with the BluePippin system (Sage Science, Beverly, MA). All libraries were bound to polymerase using the DNA/polymerase binding kit P6 v2 (PacBio), loaded on one single-molecule real-time (SMRT) Cell (PacBio), and sequenced with C4 v2 chemistry (PacBio) for 270- or 360-min movies, depending on the length, with an RS II instrument (PacBio). Sequence assembly was performed on the HGAP v3 platform (9), with an average yield of 2,869,880 bp.

A phylogenetic tree was inferred on the BioLinux command line using Parsnp v1.2 (10); it included 30 published sequences from the NCBI database annotated as Listeria innocua and was visualized in MEGA7 (6) (Fig. 1). The level of base alignment with the reference genome (H0996) was 73.8%, as assessed with Gingr v1.3 (10).

FIG 1.

Phylogenetic tree based on the full genome sequences for 39 L. innocua isolates. PacBio sequences for the 9 isolates featured in this study are shown in bold, and their sources are listed in parentheses. Strains exhibiting the atypical hemolysis phenotype are indicated by brackets. The phylogeny was constructed by Parsnp using H0996 as the reference genome. Bootstrap branch support values are depicted beside the root and branches.

All isolates that exhibited β-hemolysis were in two distinct clades (Fig. 1). Isolate 2012L-5520 (FSL J1-023) is of unknown origin, F6215 from a food source, and F6217, 2010L-2059, and 2010L-1951 from human infections (Table 1). The description of these atypical strains confirms that L. innocua can contribute to human illness. These genome sequences serve as useful resources for additional characterization of the evolutionary mechanisms of Listeria.

TABLE 1.

Summary of genomic characteristics from PacBio sequencing for the L. innocua isolates

Isolate	Host (source)	Depth of coverage (×)	No. of reads	Avg read length (bp)	No. of contigs (no. of plasmids)	Contig N50 (bp)	GC content (%)	Total genome size (bp)	SRA accession no.	BioSample accession no.	Nucleotide accession no.
F4078 (ATCC 33090)	Animal (cow brain)	79.51	15,820	16,676	2a	2,794,052	37.6	2,782,538	SRS12596339	SAMN13941777	NZ_JABXLC000000000
F6215	Food (meat)	178.65	35,563	18,169	2a	2,901,888	37.5	2,939,591	SRS12596348	SAMN13941779	NZ_JABXLE000000000
F6217	Human (peritoneum)	353.39	79,416	17,132	1a	2,884,312	37.4	2,855,969	SRS952935	SAMN03761682	NZ_CP095724.1
2015L-6714	Food (meat)	223.30	49,392	16,497	2a	2,930,612	37.38	2,984,334	SRR21942761	SAMN27548027	NZ_CP095723.1
H0996 (ATCC 51742)	Food (cabbage)	394.95	84,980	19,213	3 (1)	2,801,715	37.5	2,795,487	SRR21942762	SAMN10869156	NZ_CP095730.1
2010L-1951	Human (abscess [armpit])	431.68	72,490	21,319	3 (1)	2,818,897	37.4	2,810,256	SRS952976	SAMN03761736	NZ_CP095727.1
2010L-2059 (PNUSAL000003)	Human (knee fluid)	322.32	81,774	18,681	15 (2)	2,340,577	37.4	2,816,380	SRR21982767	SAMN10869157	NZ_CP095726.1
2012L-5520 (FSL J1-023)	Unknown	337.92	77,923	16,424	1a	2,826,835	37.5	2,826,835	SRR2157026	SAMN13941775	NZ_JABXLA000000000
2015L-6726	Food (meat)	249.48	66,858	12,899	2a	2,857,050	37.4	2,926,353	SRS1160884	SAMN04263682	NZ_CP095728.1

^aCircularized contig.

Data availability.

The GenBank accession numbers for all strains are provided in Table 1. PacBio sequences are available under BioProject accession number PRJNA212117. All isolates are available upon request.