New reference genomes of honey bee-associated bacteria Paenibacillus melissococcoides, Paenibacillus dendritiformis, and Paenibacillus thiaminolyticus

ABSTRACT We sequenced the genomes of recently discovered Paenibacillus melissococcoides (CCOS 2000) and of the type strains of closely related P. thiaminolyticus (DSM 7262) and P. dendritiformis (LMG 21716). The three genomes set the basis to unambiguous diagnostic of these honey bee associated Paenibacillus bacteria.

S everal bacteria species of the Paenibacillus genus are associated with the honey bee, Apis mellifera. Among them, P. larvae (1) causes American foulbrood, a highly contagious disease, which impedes colony development and can lead to its death. P. melissococcoides (2) and P. dendritiformis (3) were isolated from colonies affected by European foulbrood (4). P. thiaminolyticus was isolated from hive material (5,6). Because of their close 16S rRNA genetic relatedness (2), P. melissococcoides could have been wrongly identified as P. dendritiformis or P. thiaminolyticus in previous studies.
Here, we present the genome sequences of P. melissococcoides CCOS 2000 and of the type strains P. dendritiformis LMG 21716 and P. thiaminolyticus DSM 7262. The latter bacteria were obtained from BCCM and DSMZ culture collections, respectively. P. melissococcoides was found in worker jelly droplets cultured on EFB Basal medium (7,8) under anaerobic conditions for 4 days at 36°C. The sampled colony was located near Reutigen (46° 41′ 39″ N, 7° 37′ 13″ E), Switzerland.
The three bacteria were grown in 10 mL liquid Basal medium at 36°C overnight. High-molecular-weight genomic DNA was recovered using the GES method of DNA extraction (9) and assessed for quantity, quality, and purity using a Qubit 4.0 fluorometer (dsDNA HS Assay kit; Q32851, Thermo Fisher Scientific, Waltham, MA, USA), an Advanced Analytical FEMTO Pulse instrument (Genomic DNA 165 kb Kit; FP-1002-0275, Agilent, Santa Clara, CA, USA), and a Denovix DS-11 UV-Vis spectrophotometer. Multiplexed SMRTbell libraries were prepared according to PacBio guidelines, Part Number 101-696-100 Version 06 (March 2020). One microgram of gDNA in 100 µL was sheared in a g-TUBE (Covaris, Woburn, MA, USA), concentrated, and cleaned using AMPure PB beads. Samples were quanti fied and qualified to be in the range of 12-15 kb using the Qubit and the FEMTO instruments, respectively. Libraries were pooled using the PacBio microbial multiplexing calculator. Prior to and after size selection, the library pool was purified using AMPure PB beads. Size selection was performed with a BluePippin instrument (BLU0001; Sage Science, Beverly, MA, USA) using BluePippin with dye free, 0.75% Agarose Cassettes, and S1 Marker (Sage Science; BLF7510) wherein the selection cut-off was set at 6,000 bp. Library pool concentration and size was again assessed using the Qubit and FEMTO instruments, respectively. The final library pools were on average 11.4 kb in size.
PacBio Sequencing primer v4 and Sequel DNA Polymerase 3.0 were annealed and bound, respectively, to the DNA template libraries. Libraries and Spike-In internal control were diffusion loaded at an on-plate concentration of 10 or 11 pM. Sequencing was performed in continuous long read (CLR) mode on the Sequel System with Sequencing kit 3.0, SMRTCells 1M v3, and a 2 h pre-extension followed by 600 min movie time.
Raw data and assembly statistics are summarized in Table 1.

ACKNOWLEDGMENTS
This work was funded by the Agroscope and the Gottfried and Julia Bangerter-Rhyner foundation.
We thank the Next Generation Sequencing Platform of the University of Bern for the high-throughput sequencing.