Genome Sequences of 11 Brucella abortus Isolates from Persistently Infected Italian Regions

Brucella abortus was recognized as the causative agent of bovine brucellosis by Benhard Bang in 1895, distinguishing it as a separate strain from Brucella melitensis, the causative agent of Malta fever (1). The number of species in the Brucella genus has greatly expanded since then, but B. abortus remains among the world's most important livestock pathogens (2). In the past 40 years, concerted eradication efforts have reduced brucellosis prevalence in cattle in several European countries, yet the disease remains endemic in some Mediterranean countries, causing economic losses for agricultural producers and zoonotic concerns for consumers of contaminated dairy products and animal handlers (3). Despite a countrywide eradication program in Italy, the disease-free status has been achieved only in the central and northern parts of the country. In a previous study, we conducted variable-number tandem-repeat (VNTR) analysis of isolates to understand the genetic relatedness among Italian strains (4). Based on these results and recommended guidelines, we selected a subset of genetically and geographically diverse isolates for whole-genome sequencing (5).

Nine B. abortus biovar 3 and two B. abortus biovar 1 strains were selected for sequencing, 10 from cattle and one from a water buffalo (strain 8979). Approximately 1 to 5 µg of genomic DNA extracted from each isolate was sheared in a SonicMan microplate sonicator (Brooks Automation, Chelmsford, MA, USA) to produce fragments averaging 600 bp in length. Libraries for Illumina sequencing were prepared with the Kapa high-throughput library preparation kit “with bead” (Kapa Biosystems, Wilmington, MA). The libraries were labeled with 8-bp indices for multiplex sequencing (6). Library quantification was conducted with the Kapa library quantification kit on an ABI Prism 9600 real-time PCR system (Life Technologies, Grand Island, NY). The fragment size distribution was also confirmed with an Agilent DNA high-sensitivity kit for the 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). An Illumina MiSeq was used to sequence the libraries, producing 250-bp paired-end reads. The reads were assembled de novo with SPAdes 3.0.0 (7) and improved with Pilon 1.8 (8) and SSPACE 3.0 (9). Single-nucleotide polymorphisms (SNPs) were detected relative to the genome of reference strain B. abortus 2308 (GenBank accession numbers NC_007618.1 and NC_007624.1) using MUMmer 3.23 (10). The genomes were annotated by NCBI with PGAP 2.10.

The genome assemblies covered an average of 99.61% of the reference genome, with a mean of 16 contigs among assemblies and an N₅₀ of 434,000 bp. The assemblies averaged 3,161 genes (3,062 coding sequences [CDSs]), representing >99% of the genes in the reference sequence. Further genome annotation results are available in Table 1. For variable sites, 2,116 SNP positions were orthologous to the reference and all of the samples from Italy. In conclusion, our data set will hopefully form the basis of new studies of B. abortus in Italy. As a result, whole-genome sequencing is providing new insights into the phylogeography and molecular evolution of Italian isolates of Brucella.

We thank the Italian Network Istituti Zooprofilattici Sperimentali for their outstanding diagnostic capabilities and for the support of the National and OIE Brucella Reference Laboratory.

This work was funded by grants from the Italian Ministry of Health, ricerca corrente 2012, project MSRCTE0212.

We declare no conflicts of interest.