Paenibacillus larvae and their phages; a community science approach to discovery and initial testing of prophylactic phage cocktails against American Foulbrood in New Zealand
, ... Abstract
Background: American foulbrood (AFB) is a devastating disease of the European honey bee (Apis mellifera) and is found throughout the world. AFB is caused by the bacterium Paenibacillus larvae (P. larvae). Treatment with antibiotics is strictly forbidden in many regions, including New Zealand. Safe and natural prophylactic solutions to protect honey bees from AFB are needed. Bacteriophages are a well-studied alternative to antibiotics and have been shown to be effective against P. larvae in other countries.
Methods: We employed a community science approach to obtaining samples from around New Zealand to discover novel bacteriophages. Standard isolation approaches were employed for both bacteria and bacteriophages. Host range testing was performed by agar overlay spot tests, and cocktail formulation and in vitro testing were performed in 96-well plate assays, followed by sub-sampling and CFU visualization on agar plates.
Results: Herein, we describe the discovery and isolation of eight P. larvae bacterial isolates and 26 P. larvae bacteriophages that are novel and native to New Zealand. The phage genomes were sequenced and annotated, and their genomes were compared to extant sequenced P. larvae phage genomes. We test the host ranges of the bacteriophages and formulate cocktails to undertake in vitro testing on a set of representative bacterial strains. These results form the basis of a promising solution for protecting honey bees in New Zealand from AFB.
Keywords
INTRODUCTION
The European honey bee (Apis mellifera) is a valuable livestock animal globally[1]. In New Zealand, this value comes from their role in the pollination of horticultural and agricultural crops, which contributes roughly 8.7 billion dollars to New Zealand’s current GDP per annum, assuming this ratio has maintained since 2013[2]. The export of apiculture products, including honey, beeswax and live bees, contributes a further $483 million NZD p.a.[3]. Since 2006, New Zealand has seen a steep increase in the number of beekeepers and apiaries; with these rising numbers, there has also been a rising number of colony losses observed[3].
Honey bees are under constant attack by abiotic and biotic factors, including but not limited to herbicides, pesticides, parasites, viruses and bacteria[4]. The two biggest biotic threats to honey bees today are the parasitic Varroa mite (Varroa destructor) and American foulbrood (AFB), which is caused by the spore-forming, bacterial pathogen Paenibacillus larvae (P. larvae). AFB is a serious and destructive disease that attacks honey bees in their larval and pupal stages[5,6]. AFB has detrimental consequences at both the larval and colony level[7-9].
AFB has been present in New Zealand for at least 146 years after first being discovered in 1877[10,11]. By 1887, AFB had caused significant damage around the country and led to a 70% reduction in honey production[12]. The use of antibiotics to treat or mask an AFB infection in New Zealand is strictly prohibited[13,14]. Current legislation stipulates that beekeepers must destroy hives infected with AFB within seven days of discovery, using petrol fumes and incineration to ensure all traces of AFB are removed[13,14]. This method is costly to both the beekeeping community and the New Zealand economy.
A potential solution to AFB infection in New Zealand is the prophylactic application of bacteriophages in a phage cocktail. Bacteriophages, or phages informally, are self-propagating viruses that are only able to infect and replicate within bacteria. Phages are ubiquitous and are the most numerous biological entity on Earth, with at least 1031 phages in existence globally at any point in time[15,16].
Work undertaken overseas has shown it is possible to protect honey bee larvae from AFB infection through the application of P. larvae phage cocktails both in vitro[17] and in honeybee colonies in an at-risk apiary[18]. In the latter, phage protection appeared to remain intact for at least four months after application of the cocktail. In addition, a recent genomic analysis was performed in which 48 completely sequenced P. larvae phages were described and classified into four clusters and one “singleton”[19]. These provide a rich resource of information on phage diversity. However, New Zealand biosecurity laws prohibit us from importing non-native phages for domestic release. Therefore, in order to create phage cocktails to protect honeybees in our domestic honey production sector, it is necessary that we isolate P. larvae phages that are native to New Zealand. While a collection of P. larvae bacterial isolates was previously reported[20], that collection was subsequently destroyed (P. Lester personal communication). Discovering novel New-Zealand-based P. larvae isolates was therefore necessary.
Honeybees are primarily kept in New Zealand for the production of high-value honey, like mānuka
Herein, we describe how we developed a collection of novel P. larvae bacterial isolates and used these to discover P. larvae phages native to New Zealand. We briefly describe their genome sequences, report phage host ranges, and design phage cocktails in addition to performing in vitro testing of several phage cocktails. This work forms the groundwork to develop an approach to protecting beehives using New Zealand native phages that can be applied to protect hives against infection by a devastating bacterial pathogen that is affecting this industry globally.
MATERIALS AND METHODS
Isolation of Paenibacillus larvae
P. larvae was isolated by swabbing suspected brood frames and wiping swabs on MYPGP[21] with Nalidixic acid (10 μg/mL) and Pipemidic acid (10 μg/mL) plates. Plates were incubated at 37 oC for 3-5 days until colonies had formed. Single colonies were picked and purified by single colony isolation on another MYPGP plate. Subsequently single colonies were picked and grown in liquid MYPGP for 48 h at 37 oC and shaken at 100 rpm, then frozen at -80 oC.
Sporulation of Paenibacillus larvae for microscopy
To produce bacterial spores for microscopy, a 10-fold dilution series of P. larvae PFR-Pl-2006 bacterial culture was spread onto several MYPGP agar plates. Plates were incubated at 37 oC for 6-7 days and plates exhibiting individual colonies were selected. After incubation, spores were removed from the plates by washing with 5 mL cold sterile water. Water was added to the plate, and the surface of the plate was gently scraped with a sterile inoculation loop to loosen spores. Water and spores were then removed from the plates using a syringe and transferred into Eppendorf tubes. The spore suspension was concentrated via centrifugation (12,000 × g, 15 min, 4 oC). After centrifugation, the supernatant was discarded, and the spore pellet was resuspended in 1 mL of cold water. This step was repeated three times. The final spore pellets from all tubes were resuspended in a total volume of 2 mL cold water. Spores were stored at 4 oC[22].
Bacterial DNA extraction and 16S rRNA PCR
Bacterial DNA was extracted from overnight cultures in mBHI (Oxoid CM1135B) broth using the commercially available Promega Wizard Genomic DNA Purification kit (www.promega.com/protocols/). The protocol for gram-positive bacteria was followed. A PCR mix was prepared with each tube containing a final volume of 50 μL. The amplification conditions were 95 oC (3 min) followed by 30 cycles of 93 oC
Primers used were[23]:
AFB-F 5’-CTT-GTG-TTT-CTT-TCG-GGA-GAC-GCC-A-3’
AFB-R 5’-TCT-TAG-AGT-GCC-CAC-CTC-TGC-G-3’
Bacterial DNA sequencing and assembly
DNA was either sent for sequencing at MicrobesNG (Birmingham, UK) or MiGS (Pittsburgh, PA, USA) for complete genome Illumina sequencing. Sequencing was performed by MicrobesNG by preparing genomic DNA libraries with Nextera XT Library Prep Kit (Illumina, San Diego, USA). Libraries were then sequenced on an lllumina NovaSeq 6000 (Illumina, San Diego, USA) using a 250 bp paired-end protocol
Processing soil/hive samples for phages
Soil samples were processed as previously described[17]. Only one pass through a 0.45 μm sterile syringe filter was performed. The resulting filtrate was used as a starting material for enrichment. Enrichments were a combination of 1 mL of starting material, 100 μL of each of eight P. larvae bacterial isolates, 8 mL mBHI and 0.4% glucose. These were incubated for 48 h at 37 oC, shaken at 100 rpm. After 48 h, enrichments were centrifuged at 3,200 g for 15 min, and filter sterilized to 0.45 μm. The resulting supernatants were assayed for phage presence by 3 μL spots on double-layer agar containing one of the P. larvae bacterial isolates.
Phage plaque purification
Phages underwent three rounds of purification. Plaques were picked off a double-agar plate using a 200 μL pipette tip; the tip was put in 100 μL of BHI and pipetted up and down to remove phage particles. This lysate was used to inoculate the next double-agar plate.
Creation of lysates
To create phage lysates, 10 plaque plates with the highest number of individual plaques were flooded with
Phage DNA extraction and sequencing
Phage DNA was extracted using a modified zinc chloride precipitation method[31]. Modifications included the addition of 1 µL Proteinase K (20 mg/mL), incubated at 37 °C for 10 min after the resuspension in TES buffer (0.1M Tris-HCl, pH 8, 0.1M EDTA, 0.3% SDS) step. Tubes were left overnight on ice after isopropanol was added. 1 µL of pure glycogen was added to each tube at the beginning of Day 2 before the centrifugation step to aid in pelleting of DNA. DNA pellets were resuspended in 50 µL nuclease-free water.
Phage genomes were sequenced and annotated as previously described[30]. Briefly, phage genomes were assembled using Geneious 9.05 (Auckland, New Zealand) (https://www.geneious.com); assembled genomes were then run through Phage Commander[32] to identify all genes. Genomes were manually checked using DNA Master[33] and as previously described[34].
Host range testing
The ability of phages to infect each isolate was assessed by 3 μL spots of each phage lysate on double-layer agar containing 500 μL of bacterial lawn. Each P. larvae bacterial isolate was tested separately. The majority of spot tests showed the presence of individual plaques owing to low phage titers during this testing.
In vitro cocktail assays
Phage titers were normalized to 1 × 108 PFU/mL and 50 μL of each phage selected was combined into a cocktail. Bacterial cultures were grown in BHI for 48 h at 37 oC to ~1 × 108 CFU/mL. The bacteria were serially diluted up to a 10-6 dilution and 20 μL was aliquoted into 96-well plates containing 90 μL 2 × BHI and 90 μL BHI. The phage cocktails were also serially diluted such that each row contained from 107 to 103 PFU total phage. 20 μL of the phage cocktail was added to each well of the plate. Plates were incubated, shaking at 37 oC for 24 h. Aliquots of 3 μL were spotted onto BHI plates and incubated for three to four days at 37 oC to observe CFU.
RESULTS
Isolating P. larvae from infected colony material
Previous work suggested that a curated collection of P. larvae isolates from New Zealand had been characterized[20]. Further investigation revealed that the existing collection had been destroyed (P. Lester private communication). Therefore, a new collection of representative P. larvae strains was needed. AsureQuality, a New Zealand government-approved testing facility, provided us with swabs of brood frames or infected larvae material and whole brood frames from beehives suspected of AFB infection. Potential isolates were cultured on semi-selective MYPGP agar plates in order to obtain single colonies. Ultimately, eight P. larvae strains were isolated from around New Zealand, with each isolate coming from a different location [Figure 1A and Table 1]. P. larvae is a filamentous (2.5-5 μm by 0.5-0.8 μm), spore-forming, gram-positive bacterium [Figure 1B and C][6]. Isolates were confirmed to be P. larvae by positive amplification with 16S rRNA PCR primers[23].
Figure 1. Paenibacillus larvae bacterial strains (A) Locations of the eight isolated Paenibacillus larvae (P. larvae) bacterial strains;
Paenibacillus larvae bacterial strains isolated from New Zealand
| P. larvae strain | Isolation location | MLST ST | GC% | No. of contigs | Size range contigs (Kbp) | Accession No. |
| Pl-WAI | Wellington | 18 | 44.2% | 219 | 0.128-218 | JARDRH000000000 |
| Pl-TP | Rotorua | 18 | 44.1% | 157 | 0.128-250 | JARDRJ000000000 |
| Pl-CHCH | Christchurch | 18 | 44.1% | 163 | 0.128-218 | JARDRI000000000 |
| PFR-Pl-2017 | Auckland | 18 | 44.1% | 176 | 0.128-218 | JARDRG000000000 |
| PFR-Pl-2006 | Hamilton | 18 | 44.1% | 185 | 0.128-218 | JARDAI000000000 |
| Pl-F1A | North Canterbury | 18 | 44.1% | 175 | 0.5-191 | JARDRL000000000 |
| Pl-F2B | South Canterbury | 18 | 44.0% | 167 | 0.5-191 | JARDRM000000000 |
| Pl-P1627 | Queenstown | 23 | 44.1% | 171 | 0.5-195 | JARDRK000000000 |
DNA was extracted from each of the bacterial isolates and submitted for genome sequencing. Genomes were assembled using SPAdes 3.15.3[24,25] and then annotated using either RAST 1.073 [26-28] or Prokka 1.14.5[29]. The resulting genome assemblies had a range of 157 to 219 contigs, with sizes varying from 0.12 to 218 Kbp. These P. larvae strains had GC contents of 44.0% to 44.2% [Table 1].
Multilocus sequence typing (MLST) was undertaken using PubMLST[35]. MLST for P. larvae consists of the following seven housekeeping genes: ftsA (cell division protein), clpC (catabolite control protein A), glpT (glycerol-3-phosphate permease), glpF (glycerol uptake facilitator protein), rpoB (RNA polymerase beta subunit), Natrans (forward sodium dependant transporter), and sigF (sporulation sigma factor F) as these offered the most diversity between genomes tested[36]. Seven of the New Zealand isolates belonged to the 18 MLST ST and one belonged to 23 MLST ST [Table 1]. MLST can also be used to distinguish between the ERIC I genotype and the ERIC II genotype. MLST 18 and MLST 23 both belong to the ERIC I genotype[36-38].
We used CRISPRFinder to look for detectable CRISPR systems in these eight isolates[39]. Seven of the isolates contained four CRISPR arrays and one isolate contained five. The total number of spacers within the CRISPR arrays for each isolate varied from 15-25 spacers [Table 2]. Across all eight isolates, 29 unique spacers were observed, Pl-P1627 contained 12 unique spacers that were not found in any of the other isolates.
CRISPR array and spacer details of the eight Paenibacillus larvae isolates
| P. larvae strain | No. of CRISPR arrays | No. of spacers | No. of unique spacers |
| Pl-WAI | 4 | 15 | 0 |
| Pl-TP | 4 | 17 | 0 |
| Pl-CHCH | 4 | 17 | 0 |
| PFR-Pl-2017 | 4 | 17 | 0 |
| PFR-Pl-2006 | 4 | 16 | 0 |
| Pl-F1A | 4 | 17 | 0 |
| Pl-F2B | 4 | 17 | 0 |
| Pl-P1627 | 5 | 25 | 12 |
We also used DefenseFinder[40,41] to search for known anti-phage systems in our bacterial strains. All eight isolates contained the same seven anti-phage systems: both a type I and II restriction-modification system[42], a Gao_let system[43], two Cas systems (CAS_Class1-Subtype-III-B and CAS_Class1-Subtype-I-B)[44], a Wadjet_III system[45], and a Mokosh_TypeII system[46].
Finally, we used Phaster[47,48] to identify prophages contained within the genomes. Phaster designates prophages as either intact, questionable or incomplete by comparing them to a NCBI database of complete viral genomes. Potential prophage regions are then given a completeness score; this score is calculated on the proportion of phage genes in the identified region. An intact prophage has a score > 90, a questionable prophage has a score between 70-90, and an incomplete prophage has a score < 70. All isolates contained at least one intact prophage, with six containing two intact prophages. The intact prophages were genomically similar to P. larvae Phage Harrison and Phage Vegas[49]. All genomes also contained 3-4 questionable prophages and 6-9 incomplete prophages [Table 3].
Prophages found in the eight Paenibacillus larvae isolates
| P. larvae strain | Total prophages | Intact | Name of intact phage | Size of prophage (Kb) | Total proteins # | GC content (%) | Questionable | Incomplete |
| Pl-WAI | 14 | 2 | Vegas | 41.7 | 60 | 43.06 | 3 | 9 |
| Harrison | 15.2 | 19 | 43.30 | |||||
| Pl-TP | 14 | 2 | Vegas | 39.6 | 61 | 43.29 | 3 | 9 |
| Harrison | 15.2 | 18 | 43.29 | |||||
| Pl-CHCH | 14 | 2 | Vegas | 39.6 | 61 | 43.30 | 3 | 9 |
| Harrison | 15.2 | 18 | 43.31 | |||||
| PFR-Pl-2017 | 13 | 2 | Vegas | 39.6 | 60 | 43.29 | 4 | 7 |
| Harrison | 15.2 | 18 | 43.30 | |||||
| PFR-Pl-2006 | 15 | 2 | Vegas | 39.6 | 61 | 43.30 | 4 | 9 |
| Harrison | 15.2 | 18 | 43.30 | |||||
| Pl-F1A | 12 | 2 | Vegas | 35.7 | 55 | 43.56 | 3 | 7 |
| Harrison | 31.5 | 52 | 42.07 | |||||
| Pl-F2B | 13 | 1 | Harrison | 41.5 | 81 | 41.95 | 3 | 9 |
| Pl-P1627 | 10 | 1 | Vegas | 43.1 | 66 | 43.68 | 3 | 6 |
Phage discovery
A community science approach to national sample collection
Bee hives are distributed throughout the country in out-of-the-way locations and often on private property. In order to isolate phages from around New Zealand, we used a community science approach to engage the assistance of New Zealand beekeepers. An infographic [Supplementary Figure 1] was developed and distributed widely in beekeeping circles via social media, beekeeping magazines, in-person apiculture conferences and posted on our website (http://www.hendricksonlab.co.nz/ABATE/). Beekeepers were encouraged to take samples of soil or hive/bee debris and return them to be processed for the presence of phages in a prepaid and addressed envelope. As part of the community science, beekeepers were able to name any phages that were discovered within a sample they had provided us. A total of 720 sample tubes were distributed, out of which 430 samples were returned and processed, with a return rate of 60%. Samples were taken from a wide distribution of locations in New Zealand [Figure 2A].
Figure 2. Sourcing samples from beehives across the nation. (A) Locations of samples of soil, bee debris, or wax that were provided by beekeepers; (B) The locations and names of Paenibacillus larvae (P. larvae) phages discovered as a result of these efforts.
Twenty-six of the samples contained a novel phage able to infect at least one of our bacterial isolates of
Details of 26 Paenibacillus larvae phages discovered, sequenced and annotated
| Geographic region | Bacteria isolated on | Genome length (bp) | No. of genes | GC content (%) | Cluster | Accession No. | |
| ABAtENZ | Hamilton | Pl-PFR-2017 | 44,419 | 82 | 42.97 | Vegas | OP503968 |
| AJG77 | Wanaka | Pl-PFR-2017 | 44,417 | 82 | 42.98 | Vegas | OP503969 |
| ApiWellbeing | Masterton | Pl-F1A | 44,429 | 82 | 43.01 | Vegas | OP503970 |
| BarryFoster_Benicio | Whangarei | Pl-F1A | 44,421 | 82 | 42.98 | Vegas | OP503543 |
| Bloomfield | Haast | Pl-PFR-2017 | 44,419 | 82 | 42.98 | Vegas | OP503971 |
| Bob | Matakana Island | F2B | 43,553 | 80 | 43.03 | Vegas | OP503972 |
| Callan | West Taratahi | Pl-PFR-2006 | 44,768 | 77 | 39.69 | Harrison | OP503989 |
| Carlos | Carterton | Pl-F1A | 44,430 | 83 | 42.98 | Vegas | OP503973 |
| Dante | Elgin | Pl-WAI | 44,420 | 82 | 42.98 | Vegas | OP503974 |
| Dash | West Taratahi | Pl-PFR-2006 | 44,599 | 79 | 39.39 | Harrison | OP503990 |
| FutureBee | Hamilton | Pl-TP | 44,417 | 83 | 42.98 | Vegas | OP503975 |
| GaryLarson | Willowby | Pl-F2B | 44,420 | 82 | 42.98 | Vegas | OP503976 |
| GIW2016 | Wanaka | Pl-PFR-2017 | 43,555 | 80 | 43.01 | Vegas | OP503977 |
| Jacinda | Haast | Pl-PFR-2017 | 44,419 | 82 | 42.97 | Vegas | OP503978 |
| Lena | Rotorua | Pl-PFR-2017 | 44,420 | 82 | 42.97 | Vegas | OP503979 |
| Lilo | Pukekawa | Pl-F1A | 40,941 | 70 | 40.33 | Harrison | OP503991 |
| Logan | Tolaga Bay | Pl-PFR-2017 | 44,419 | 82 | 42.99 | Vegas | OP503980 |
| LunBun | Gisborne | Pl-F1A | 44,421 | 82 | 42.97 | Vegas | OP494865 |
| NHScienceFair | Albany | F1A | 44,419 | 82 | 42.98 | Vegas | OP503981 |
| Ollie | Marton | Pl-PFR-2017 | 44,420 | 83 | 42.98 | Vegas | OP503982 |
| Rae.2Bee1 | Fairton | Pl-TP | 44,420 | 82 | 42.97 | Vegas | OP503983 |
| Rosalind | Westport | F1A | 43,556 | 80 | 43.00 | Vegas | OP503984 |
| Ted | Napier | Pl-PFR-2017 | 44,419 | 82 | 42.99 | Vegas | OP503985 |
| TonyLawson77 | Palmerston North | Pl-F1A | 44,420 | 82 | 42.96 | Vegas | OP503986 |
| UtuhinaGold_Zacery | Rotorua | Pl-PFR-2017 | 44,420 | 82 | 42.97 | Vegas | OP503987 |
| WildCape | Gisborne | Pl-F1A | 44,430 | 82 | 43.00 | Vegas | OP503988 |
Upon sequencing, we discovered the New Zealand phages were between 40-44 kbp in length with 70-83 genes per genome. The phages belong to two of the four major genomically determined clusters of P. larvae phages; three belong to the Harrison cluster and 23 belong to the Vegas cluster [Table 4]. Clusters were determined by average nucleotide identity (ANI); if two phages have ANI greater than or equal to 60%, they are placed in the same cluster[19]. All New Zealand P. larvae phages are linear and use the 3’ cohesive end DNA packaging mechanism, similar to the majority of previously described P. larvae phages[19]. The New Zealand P. larvae phages are lytic in vitro, despite the presence of annotated integrases in their genomes, the presence of which suggests the capacity for a temperate lifestyle. Phage Dash (Harrison Cluster) has an integrase at GP38 and phage ABAtENZ (Vegas Cluster) has an integrase at GP32 [Figure 3]. The presence of integrases, and the absence of evidence of a temperate lifecycle in the laboratory, is consistent across the majority of known P. larvae phages[50]. All New Zealand phage genomes encode a conserved
Figure 3. Representative phage genome maps for Dash (Harrison cluster) and ABAtENZ (Vegas cluster) generated with Phamerator.org, showing pairwise sequence similarity (shading) and the homologous genes (matching-colored boxes). The shades of color indicate a combination of length and significance of nucleotide identity, and the large purple blocks are the strongest regions of similarity observed between phages Dash and ABAtENZ.
Isolating phages from soil/hive material
The 430 samples received were processed using an enrichment technique followed by three rounds of plaque purification [Figure 4A and B]. The plaque morphologies of all isolated phages were tiny, pin-prick plaques that appeared clear. When tested by a standard spot titer plate method, all of our phages had very low effective titers ranging from 6.7 × 102 to 2.7 × 105. A RAMP-UP technique was used to increase the titer of all phages in order to extract DNA and send it for sequencing[30]. Once these titers were raised by this method to > 1 × 108, we proceeded with visualization and complete genome sequencing.
Figure 4. Phage discovery (A) Schematic of phage enrichment and isolation process (created with BioRender.com); (B) Positive spot tests after enrichment; (C) Representative TEM image of Phage Lilo (Harrison cluster); (D) Representative TEM image of Phage Ollie (Vegas cluster). Scale bars = 50 nm. TEM: Transmission electron microscopy.
Transmission electron microscopy of isolated phage
Electron microscopy was undertaken on Phage Lilo (Harrison Cluster) Figure 4C and Phage Ollie (Vegas Cluster) [Figure 4D]. These two phages were selected as they each represented one of the two clusters of phages found in New Zealand. These revealed phages with long, filamentous, non-contractile tails; phages with these types of tails are classified as having Siphoviridae morphology[53]. All the new phages reported here are Gochnauervirinae, a subfamily in the Caudoviricetes class. All but one of the known P. larvae phages have this morphotype[19]. Phage Lilo had a tail of approximately 148 nm in length, with a prolate head measuring approximately 105 nm by 41 nm. Phage Ollie had a tail approximately 156 nm in length, with a prolate head measuring approximately 106 nm by 43 nm.
Host range testing
Specificity of the New Zealand isolated phages on each of the eight native P. larvae isolates identified in this paper, as well as 22 native P. larvae isolates provided by the ApiWellbeing team[54] was carried out using standard spot test assays. Phages were scored as positive or negative for cell lysis. Nine distinct infection patterns were identified [Figure 5].
Figure 5. Host range of Paenibacillus larvae (P. larvae) phages on P. larvae bacterial isolates from New Zealand. Grey boxes indicate cell lysis and white boxes indicate no cell lysis has occurred. NB: In some instances, spot clearing was observed, but plaques were not. This is explicit in Supplementary Figure 2.
None of the phages were capable of lysing all 30 bacterial isolates, but they were able to lyse between 57% to 87%. Bacterial isolatets Pl-P1627 and W19_08094, both isolated from the Otago region, are not lysed by any of the phages found in New Zealand to date. Pl-P1627 belongs to a different multilocus sequence type than the other seven bacterial strains identified and sequenced in this paper, as well as having 12 unique spacer sequences within its CRISPR arrays.
Phages Dash, Lilo, and Callan can infect P. larvae strains W19_08078, W19_08082, W19_08091,
Cocktail formulation and in vitro testing
Initially, four cocktails were formulated based on the host range of the phages to ensure coverage of as many bacterial strains as possible [Table 5]. We note that at this time, the phage genome sequences were not yet known. The incidence of American foulbrood is low and must be reported by law in New Zealand; the likelihood of a colony being infected with P. larvae in New Zealand is 0.0032[55], so the chance of more than one strain infecting a single hive in New Zealand is extremely low (~0.00001024). Our phage cocktail design, therefore, focused on covering the breadth of strains that could infect a colony [Table 4].
The phages contained within the four cocktails
| Cocktail one | Callan | Logan | ApiWellbeing | Freya |
| Cocktail two | Dash | AJG77 | UtuhinaGold_ Zacery | NHScienceFair |
| Cocktail three | Callan | Ted | FutureBee | BarryFoster_ Benicio |
| Cocktail four | Dash | LunBun | Carlos | Bloomfield |
The initial set of cocktails, One to Four, were capable of lysing 93% of all P. larvae strains in our collection (28/30). Ultimately, each cocktail had a 70%-73% breadth of activity across our 30 bacterial isolates. The breadth of activity is calculated as the susceptibility of a pathogen to at least two phages in the cocktail[56]. Higher breadth is an indication of the cocktail's ability to mitigate phage resistance in the pathogen.
Once the four cocktails were decided upon, they were tested against four bacterial strains chosen to represent the types of P. larvae present in New Zealand; the bacterial strains selected were Pl-2017, Pl-2006, W19-08100, and Pl-P1627 [Figure 6A]. P. larvae Pl-P1627 was chosen as a negative control as it is not infected by the P. larvae phages in our collection to date.
Figure 6. In vitro testing of phage cocktails. (A) Four cocktails were tested for their effectiveness against each of four bacterial strains: PFR-Pl-2017, PFR-Pl-2006, W19-08100, and Pl-P1627. Strain Pl-P1627 was known to be resistant to all phages in this study; (B) Tables to summarize the results of cocktail testing and the substitution of Phage Callan for Phage Dash in Cocktails 2 and 4. Table colours represent the host range observations. Black: plaques; grey: full clearings; white: no infection. The symbols within the table record the overall effectiveness of the cocktail that phage is in on that host strain: +: effective cocktail; -/+: partially effective cocktail; -: ineffective cocktail.
All four cocktails showed good activity on strain Pl-2017, with cocktails Two and Four showing slightly better lysis potential. Cocktails Two and Four were very effective on strain Pl-2006, while cocktails One and Three only showed adequate lysis at the highest concentration of phages. On strain W19-08100, cocktails One, Two and Four all showed some lysis, while cocktail Three showed very little lysis potential on the susceptible P. larvae isolates. P. larvae Pl-P1627 was not infected by the cockails.
Ultimately, we sequenced the phage genomes and found that despite their excellent activity, phage cocktails Two and Four contained phage Dash, a phage with a large Plx1 toxin encoded in the genome. The host range of phage Dash was, however, very similar to that of phage Callan. These two Harrison cluster phages shared 91% of their genes and Callan does not contain the Plx1 toxin. We therefore tested the capacity of cocktails Two and Four with phage Callan on the same P. larvae strains [Supplementary Figure 3 and Figure 6B]. To our surprise, this seemingly small substitution did not retain the activity of the original cocktails Two and Four. Replacing phage Dash with the safer phage, phage Callan reduced the capacity of the cocktails to lyse these strains. This phenomenon warrants further investigation and suggests that phage-phage interactions are coming into play.
DISCUSSION
In this study, we have set out to lay the groundwork for using phages as a prophylactic against the devastating pathogen P. larvae in the New Zealand apiculture industry. Our work began with the isolation and preliminary sequencing of a set of eight novel P. larvae strains from across New Zealand. These
To better understand how these isolates might interact with phages, we screened these preliminary genomes for positive signs of phage defense mechanisms. We found each of our isolates contained CRISPR arrays. Each strain contained four to five CRISPR arrays with a total of 15 to 25 spacers. P. larvae strains isolated previously have been found to contain CRISPR arrays as phage defense mechanisms. P. larvae ERIC I strains ATCC 9545 and DSM 7030 contained four CRISPR arrays with 17 spacers[57].
We also used PHASTER to evaluate the sequenced contigs for prophages, and we found each isolate had one or two intact prophages. In another study, P. larvae ERIC type I strains DSM 25719, MEX14,
Seven anti-phage systems were also discovered within our eight bacterial strains. These data suggested to us that these isolates are encountering an active population of phages in nature and are maintaining a suite of defense systems to counter infection when they meet. This is common as previous reports suggest that 50% of bacteria have CRISPR systems[59] and other defense mechanisms such as restriction-modification systems are widely found within prokaryotes[60].
The ApiWellbeing project, an initiative of the New Zealand Ministry for Primary Industries, generously gifted us with 22 additional P. larvae isolates from their own recent collection efforts, which brought our collection of hosts to 30[38]. The sequencing and annotation of this collection are underway and will provide a valuable asset in the future.
Since the discovery of the first P. larvae phage in 1953[61], 69 P. larvae-specific phages have been found[49,62-69]. Due to the strict biosecurity laws in New Zealand, it is unlikely that non-native phages would be permitted in the apiculture industry here. We therefore sought to discover a suite of native New Zealand P. larvae phages to combat AFB.
Previous hunts for P. larvae phages have included samples from soil, bee debris, cosmetics, and bee wax[62,65,69]. A large-scale hunt across New Zealand was a daunting task for our small team; we, therefore, approached beekeepers from around New Zealand and received 430 samples of bee debris and soil from both the North and South Islands. These types of community science phage hunts have been used previously and there are ongoing community science projects to isolate new phages for Pseudomonas aeruginosa[70]. These samples were processed and led to the discovery of 26 independent phages. Unlike similar efforts overseas[17,62] in which phages have been isolated from infected hives, the phages discovered herein were reported to have been isolated only from hive material or soil associated with healthy hives[62].
The phage genomes were sequenced to completion, their genes were identified and annotated, and the genomes were made available publicly [Table 4]. Sequencing and annotation of phage genomes is particularly important for identifying gene functions that would make the phages unsuitable or unsafe for therapeutic use. In this study, we found that two phages, Dash and Lilo, contain a dangerous toxin that confers virulence to P. larvae, thereby ruling out these two phages for future therapy applications. Unfortunately, at the time the cocktail testing was carried out, these phages had not yet been sequenced, and the presence of this toxin was therefore unknown.
To determine whether our phage genomes were distinct, we used criteria previously described by Stamereilers et al.[19]. Phages are usually phenotypically identical if they have an ANI greater than 99.975%. We had several groups of phages that had ANI greater than this cut-off value. However, further analysis showed they all contained at least one amino acid difference, so in these instances, the phages were classed as phenotypically different. A more thorough analysis of these genomes and their relationships to the global P. larvae phages is in preparation.
Host range experiments revealed nine distinct infection patterns, which included a subset of six bacterial strains that were only able to be infected by three phages and two recalcitrant bacterial strains that were not infected by any of the phages discovered in New Zealand to date. Overall, there was a 93% host range coverage for our collection of P. larvae phages. In similar host-range experiments with P. larvae phages, Yost et al. found a 100% host-range coverage when testing 29 phages on 11 P. larvae bacterial strains[17]. In another experiment, Brady et al. tested 39 P. larvae specific phages on 59 bacterial strains and also found a 100% host range coverage[18]. Efforts are ongoing to find phages that can lyse the final resistant strains that are present in New Zealand. We do not currently know if phages found overseas have the ability to lyse the resistant P. larvae strains.
Four cocktails were formulated and tested against each of the four P. larvae isolates before the genomes were known. Cocktails One and Three both contained Phage Callan and Cocktails Two and Four both contained Phage Dash, as these were two of the phages that were able to infect six bacterial strains resistant to all other phages.
In vitro testing using these four cocktails varied, with two cocktails standing out as the most effective at killing three of the bacterial strains. One P. larvae strain, Pl-P1627, was completely resistant to all cocktails. This was to be expected as this strain was not infected by any of our individual phages. In this limited instance, we did not see any evidence of emergent infectivity above and beyond that of the individual phages present in the cocktail.
In our study, cocktails Two and Four were more effective than cocktails One and Three against P. larvae strains Pl-2006, Pl-2017 and W19-08100. In New Zealand, the likelihood of P. larvae infections is 0.0032[55]. Therefore, an ideal cocktail should be highly effective against each of the prevalent P. larvae strains, but it need not be effective against infection by multiple strains simultaneously at the level of a hive.
Interestingly, cocktails One and Four had a predicted Breadth1 of activity of 50% (2/4 phages were able to infect at least 2/4 bacterial strains), while cocktails Two and Three had a 75% Breadth1 of activity. This suggests that the quality of the phage cocktails tested here cannot be attributed to the Breadth1 of activity alone. The cause of the difference in outcomes of these phage cocktails remains to be investigated.
We observed host-dependent phage antagonism in our cocktails. When phage Callan was used in place of phage Dash on P. larvae W19-08100. This was surprising because phage Callan is not able to form plaques on this host. There are at least two possible explanations for this phenomenon. It may imply that the effect of the presence of Callan in the cocktail is the result of a direct host response preventing cocktail members from infection as a result of phage Callan DNA in the cytoplasm. Another possible mechanism is that the phage Callan lysate contains some effector which changes the susceptibility of this host to other phage cocktail members. Several studies have shown that phages within phage cocktails can have an antagonistic relationship. Forti et al found combining six phages into a cocktail to lyse Pseudomonas aeruginosa showed a lower host range than what had been predicted based on individual phage host ranges[71]. Another study testing different phage cocktails on Escherichia coli O157 showed that not all combinations of phages were as effective as others and phage antagonism was common in certain cocktails[72]. Our results suggest that there can be both host dependence on these antagonistic effects and that non-plaque forming phages can induce host resistance to infection. These preliminary results warrant further investigation.
These results, taken together with previous studies, show the importance of testing a variety of phage cocktails to find the most effective combination of phages, regardless of their individual host ranges. All
This study shows promising results and forms the beginning of the work needed to find a solution to prophylactically protect honey bees in New Zealand from the destructive disease known as AFB. Further work will need to be undertaken to completely understand the characteristics of the phages, their persistence in the environment, cross-resistance of P. larvae strains to phages, and appropriate delivery mechanisms. In order to bring about a prophylactic solution for beekeepers, we will also need to undertake in vitro testing on honeybee larvae and large-scale field trials[11]. This work has been made possible by the collective efforts of the beekeepers of New Zealand, and we will continue to honour their contributions by pursuing this project further.
DECLARATIONS
AcknowledgmentsWe wish to thank the many beekeepers in New Zealand who contributed soil and bee debris samples which made this work possible. We thank Richard Hall and Hayley Pragert from the ApiWellBeing project, and the many helpful people from both AsureQuality, Apiculture NZ, and the National American Foulbrood Pest Management Plan for support of this project. Last but not least, our sincerest thanks to Barry Foster for his support and kindness.
Authors’ contributionsConceptualization: Hendrickson HL, Kok DN
Methodology: Kok DN
Formal analysis: Kok DN
Investigation: Kok DN, Zhou D
Writing - original draft preparation: Kok DN
Writing - review and editing: Kok DN, Zhou D, Tsourkas PK, Hendrickson HL
Visualization: Kok DN
Supervision: Hendrickson HL
Project administration: Hendrickson HL
Funding acquisition: Hendrickson HL
Availability of data and materialsThe authors confirm that the genomes used for this work are publicly available, and their accession numbers are listed in Tables 1 and 4.
Financial support and sponsorshipThis research was funded by AGMARDT, grant number AIGITINQ-000301, The Sustainable Food & Fibre Futures Fund, grant number 405604, and Apiculture NZs Honey Trust.
Conflicts of interestAll authors declared that there are no conflicts of interest.
Ethical approval and consent to participateNot applicable.
Consent for publicationNot applicable.
Copyright© The Author(s) 2023.
Supplementary MaterialsREFERENCES
1. Klein AM, Vaissière BE, Cane JH, et al. Importance of pollinators in changing landscapes for world crops. Proc Biol Sci 2007;274:303-13.
2. Newstrom-Lloyd LE. Pollination in New Zealand. In: Dymond JR, editor. Ecosystem services in New Zealand : conditions and trends. Lincoln: Manaaki Whenua Press; 2013. p. 408-431. Available from: http://www.mwpress.co.nz/__data/assets/pdf_file/0008/77057/2_11_Newstrom.pdf. [Last accessed on 26 Jul 2023].
3. Ministry for Primary Industries. 2021 apiculture monitoring data. Available from: https://www.mpi.govt.nz/dmsdocument/48793-2021-Apiculture-monitoring-report-data. [Last accessed on 26 Jul 2023].
4. Li G, Zhao H, Liu Z, Wang H, Xu B, Guo X. The wisdom of honeybee defenses against environmental stresses. Front Microbiol 2018;9:722.
5. Genersch E. American Foulbrood in honeybees and its causative agent,
6. Genersch E.
7. Alippi AM, Reynaldi FJ, López AC, De Giusti MR, Aguilar OM. Molecular epidemiology of
8. Rauch S, Ashiralieva A, Hedtke K, Genersch E. Negative correlation between individual-insect-level virulence and colony-level virulence of
9. Alippi AM, López AC, Aguilar OM. Differentiation of
10. Lester P. Healthy bee, sick bee: the influence of parasites, pathogens, predators and pesticides on honey bees. Victoria University of Wellington Press; 2021. Available from: https://play.google.com/books/reader?id=fY4bEAAAQBAJ&pg=GBS.PP1&hl=en. [Last accessed on 26 Jul 2023].
11. Kok DN, Hendrickson HL. Save our bees: bacteriophages to protect honey bees against the pathogen causing American Foulbrood in New Zealand. N Z J Zool 2023;1-16.
13. The management agency national american foulbrood pest management plan. Available from: https://afb.org.nz/wp-content/uploads/2018/07/BRIEFING-DOCUMENT-MPI-Government-01112017.pdf. [Last accessed on 31 Jul 2023].
14. Biosecurity (National American Foulbrood Pest Management Plan) Order 1998. 1998. p. 1-21. Available from: https://afb.org.nz/wp-content/uploads/2018/10/Biosecurity-National-American-Foulbrood-Pest-Management-Plan-Order-1998.pdf. [Last accessed on 25 Jul 2023].
15. Mushegian AR. Are there 1031 virus particles on earth, or more, or fewer? J Bacteriol 2020;202:e0052-20.
16. Hendrix RW, Smith MC, Burns RN, Ford ME, Hatfull GF. Evolutionary relationships among diverse bacteriophages and prophages: all the world’s a phage. Proc Natl Acad Sci USA 1999;96:2192-2197.
17. Yost DG, Tsourkas P, Amy PS. Experimental bacteriophage treatment of honeybees (
18. Brady TS, Merrill BD, Hilton JA, Payne AM, Stephenson MB, Hope S. Bacteriophages as an alternative to conventional antibiotic use for the prevention or treatment of
19. Stamereilers C, Fajardo CP, Walker JK, et al. Genomic analysis of 48
20. Graham SAM. American foulbrood and its causative agent, Paenibacillus larvae, in new zealand’s registered hives and apiaries. Available from: https://openaccess.wgtn.ac.nz/articles/thesis/American_foulbrood_and_its_causative_agent_Paenibacillus_larvae_in_New_Zealand_s_registered_hives_and_apiaries/17013008. [Last accessed on 26 Jul 2023].
21. Dingman DW, Stahly DP. Medium promoting sporulation of
22. de Graaf DC, Alippi AM, Antúnez K, et al. Standard methods for American Foulbrood research. J Apic Res 2013;52:1-28.
23. Dobbelaere W, de Graaf DC, Peeters JE. Development of a fast and reliable diagnostic method for American Foulbrood disease
24. Bankevich A, Nurk S, Antipov D, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing.
25. Prjibelski A, Antipov D, Meleshko D, Lapidus A, Korobeynikov A. Using SPAdes De Novo assembler. Curr Protoc Bioinformatics 2020;70:e102.
26. Overbeek R, Olson R, Pusch GD, et al. The SEED and the Rapid Annotation of microbial genomes using Subsystems Technology (RAST). Nucleic Acids Res 2014;42:D206-14.
27. Brettin T, Davis JJ, Disz T, et al. RASTtk: a modular and extensible implementation of the RAST algorithm for building custom annotation pipelines and annotating batches of genomes. Sci Rep 2015;5:8365.
28. Aziz RK, Bartels D, Best AA, et al. The RAST Server: rapid annotations using subsystems technology. BMC Genomics 2008;9:75.
30. Kok DN, Turnbull J, Takeuchi N, Tsourkas PK, Hendrickson HL.
31. Santos MA. An improved method for the small scale preparation of bacteriophage DNA based on phage precipitation by zinc chloride. Nucleic Acids Res 1991;19:5442.
32. Lazeroff M, Ryder G, Harris SL, Tsourkas PK. Phage commander, an application for rapid gene identification in bacteriophage genomes using multiple programs. Phage 2021;2:204-13.
33. Pope WH, Jacobs-sera D. Annotation of bacteriophage genome sequences using DNA master: an overview. In: Clokie MR, Kropinski AM, Lavigne R, editors. Bacteriophages. New York: Springer; 2018. p. 217-29.
34. Salisbury A, Tsourkas PK. A method for improving the accuracy and efficiency of bacteriophage genome annotation. Int J Mol Sci 2019;20:3391.
35. Jolley KA, Bray JE, Maiden MCJ. Open-access bacterial population genomics: BIGSdb software, the PubMLST.org website and their applications. Wellcome Open Res 2018;3:124.
36. Morrissey BJ, Helgason T, Poppinga L, Fünfhaus A, Genersch E, Budge GE. Biogeography of
37. Papić B, Diricks M, Kušar D. Analysis of the global population structure of
38. Binney BM, Pragert H, Foxwell J, et al. Genomic analysis of the population structure of
39. Grissa I, Vergnaud G, Pourcel C. CRISPRFinder: a web tool to identify clustered regularly interspaced short palindromic repeats. Nucleic Acids Res 2007;35:W52-7.
40. Tesson F, Hervé A, Mordret E, et al. Systematic and quantitative view of the antiviral arsenal of prokaryotes. Nat Commun 2022;13:2561.
41. Abby SS, Néron B, Ménager H, Touchon M, Rocha EP. MacSyFinder: a program to mine genomes for molecular systems with an application to CRISPR-Cas systems. PLoS One 2014;9:e110726.
42. Oliveira PH, Touchon M, Rocha EP. The interplay of restriction-modification systems with mobile genetic elements and their prokaryotic hosts. Nucleic Acids Res 2014;42:10618-31.
43. Gao L, Altae-Tran H, Böhning F, et al. Diverse enzymatic activities mediate antiviral immunity in prokaryotes. Science 2020;369:1077-84.
44. Bernheim A, Bikard D, Touchon M, Rocha EPC. Atypical organizations and epistatic interactions of CRISPRs and cas clusters in genomes and their mobile genetic elements. Nucleic Acids Res 2020;48:748-60.
45. Doron S, Melamed S, Ofir G, et al. Systematic discovery of antiphage defense systems in the microbial pangenome. Science 2018;359:eaar4120.
46. Millman A, Melamed S, Leavitt A, et al. An expanded arsenal of immune systems that protect bacteria from phages. Cell Host Microbe 2022;30:1556-69.
47. Arndt D, Grant JR, Marcu A, et al. PHASTER: a better, faster version of the PHAST phage search tool. Nucleic Acids Res 2016;44:W16-21.
48. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. PHAST: a fast phage search tool. Nucleic Acids Res 2011;39:W347-52.
49. Tsourkas PK, Yost DG, Krohn A, et al. Complete genome sequences of nine phages capable of infecting
50. Tsourkas PK.
51. Ebeling J, Fünfhaus A, Genersch E. The buzz about ADP-ribosylation toxins from
52. Fünfhaus A, Poppinga L, Genersch E. Identification and characterization of two novel toxins expressed by the lethal honey bee pathogen
54. Ministry for Primary Industries. ApiWellbeing. Available from: https://www.mpi.govt.nz/biosecurity/how-to-find-report-and-prevent-pests-and-diseases/bee-biosecurity/apiwellbeing/. [Last accessed on 26 Jul 2023].
55. King C. American foulbrood. Surveillance 2020;47:42. Available from: https://www.mpi.govt.nz/dmsdocument/43978. [Last accessed on 31 Jul 2023]
56. Abedon ST, Danis-Wlodarczyk KM, Wozniak DJ. Phage cocktail development for bacteriophage therapy: toward improving spectrum of activity breadth and depth. Pharmaceuticals 2021;14:1019.
57. Stamereilers C, Wong S, Tsourkas PK. Characterization of CRISPR spacer and protospacer sequences in
58. Ribeiro HG, Nilsson A, Melo LDR, Oliveira A. Analysis of intact prophages in genomes of
59. Hille F, Richter H, Wong SP, Bratovič M, Ressel S, Charpentier E. The biology of CRISPR-Cas: backward and forward. Cell 2018;172:1239-59.
60. Loenen WA, Raleigh EA. The other face of restriction: modification-dependent enzymes. Nucleic Acids Res 2014;42:56-69.
61. Gochnauer T A. Some properties of a bacteriophage from
62. Merrill BD, Fajardo CP, Hilton JA, et al. Complete genome sequences of 18
63. Jończyk-Matysiak E, Owczarek B, Popiela E, et al. Isolation and characterization of phages active against
64. Carson S, Bruff E, DeFoor W, et al. Genome sequences of six
65. Walker JK, Merrill BD, Berg JA, et al. Complete genome sequences of
66. Beims H, Wittmann J, Bunk B, et al.
67. Oliveira A, Melo LD, Kropinski AM, Azeredo J. Complete genome sequence of the broad-host-range
68. Ribeiro HG, Melo LDR, Oliveira H, et al. Characterization of a new podovirus infecting
69. Yost DG, Chang C, LeBlanc L, et al. Complete genome sequences of
70. Citizen Phage Library. Available from: https://www.citizenphage.com. [Last accessed on 26 Jul 2023].
71. Forti F, Roach DR, Cafora M, et al. Design of a broad-range bacteriophage cocktail that reduces
Cite This Article
Export citation file: BibTeX | RIS
OAE Style
Kok DN, Zhou D, Tsourkas PK, Hendrickson HL. Paenibacillus larvae and their phages; a community science approach to discovery and initial testing of prophylactic phage cocktails against American Foulbrood in New Zealand. Microbiome Res Rep 2023;2:30. http://dx.doi.org/10.20517/mrr.2023.16
AMA Style
Kok DN, Zhou D, Tsourkas PK, Hendrickson HL. Paenibacillus larvae and their phages; a community science approach to discovery and initial testing of prophylactic phage cocktails against American Foulbrood in New Zealand. Microbiome Research Reports. 2023; 2(4): 30. http://dx.doi.org/10.20517/mrr.2023.16
Chicago/Turabian Style
Kok, Danielle N., Diana Zhou, Philippos K. Tsourkas, Heather L. Hendrickson. 2023. "Paenibacillus larvae and their phages; a community science approach to discovery and initial testing of prophylactic phage cocktails against American Foulbrood in New Zealand" Microbiome Research Reports. 2, no.4: 30. http://dx.doi.org/10.20517/mrr.2023.16
ACS Style
Kok, DN.; Zhou D.; Tsourkas PK.; Hendrickson HL. Paenibacillus larvae and their phages; a community science approach to discovery and initial testing of prophylactic phage cocktails against American Foulbrood in New Zealand. Microbiome. Res. Rep. 2023, 2, 30. http://dx.doi.org/10.20517/mrr.2023.16
About This Article
Special Issue
Copyright
Data & Comments
Data
Cite This Article 8 clicks
Like This Article 5
likes
Comments
Comments must be written in English. Spam, offensive content, impersonation, and private information will not be permitted. If any comment is reported and identified as inappropriate content by OAE staff, the comment will be removed without notice. If you have any queries or need any help, please contact us at support@oaepublish.com.