acmi.0.000736.v1-html.html

Erysipelothrix

spp. and other

Erysipelotrichales

detected by 16S rRNA microbial

community profiling in samples from healthy conventionally reared chickens and their

environment

Eva Wattrang

, Tina Sørensen Dalgaard

, Helena Eriksson

and Robert Söderlund

1.2 Affiliations

1 Department of Microbiology, National Veterinary Institute, Uppsala, Sweden

2 Department of Animal and Veterinary Sciences, Aarhus University, Tjele, Denmark

3 Department of Animal Health and Antimicrobial Strategies, National Veterinary Institute, Uppsala,

Sweden

1.3 Corresponding author:

Eva Wattrang, eva.wattrang@sva.se

1.4 Keywords:

Erysipelotrichales

;

Erysipelothrix

spp.; chicken

1.5 Repositories:

European Nucleotide Archive (https://www.ebi.ac.uk/ena) under project number

PRJEB67586

2. Abstract

Outbreaks of erysipelas, a disease caused by infection with

Erysipelothrix rhusiopathiae

(ER), is a re-

emerging problem in cage-free laying hen flocks. The source of ER infection in hens is usually unknown

and serological evidence has indicated the presence of ER or other antigenically related bacteria also in

healthy flocks. The aim of the present study was to evaluate sample collection, culture methods and DNA-

based methodology to detect ER and other

Erysipelotrichales

in samples from healthy chickens and their

environment.

We used samples from a research facility with conventionally reared chickens with no history of erysipelas

outbreaks where hens with high titers of IgY recognising ER previously have been observed. Microbial

DNA was extracted from samples either directly or after pre-culture in nonselective or ER-selective

medium. Real-time PCR was used for detection of

Erysipelothrix

spp. and high-throughput amplicon

sequencing of 16S rRNA sequencing was used for detection of

Erysipelotrichales

. A pilot serological

analysis of some

Erysipelotrichales

members with IgY from unvaccinated and ER vaccinated high

biosecurity-chickens as well as conventionally reared chickens was also performed.

All samples were negative for ER,

E. tonsillarum

and

E. piscisicarius

by PCR analysis. However, 16S rRNA

community profiling indicated the presence of several

Erysipelotrichales

genera

in both environmental

samples and chicken intestinal samples including

Erysipelothrix

spp. that were detected in environmental

samples. Sequences from

Erysipelothrix

spp. were most frequently detected in samples pre-cultured in

ER-selective medium. On species level the presence of

E. anatis

and/or

E. aquatica

was indicated.

Serological results indicated that IgY raised to ER showed some cross-reactivity with

E. anatis

Hence, environmental samples pre-cultured in selective medium and analysis by 16S rRNA sequencing

proved a useful method for detection of

Erysipelotrichales

including

Erysipelothrix

spp. in chicken flocks.

The observation of such bacteria in environmental samples offers a possible explanation for the

observation of high antibody titres to ER in flocks without a history of clinical erysipelas.

6. Data summary

All results from the presented experiments are included in this published article, sequence data with

corresponding sample metadata are available from the European Nucleotide Archive

(https://www.ebi.ac.uk/ena) under project number PRJEB67586.

The authors confirm all supporting data and protocols have been provided within the

article or through supplementary data files.

Prepr

int

Preprint DOI:

Posted on November 8, 2023

https://doi.org/10.1099/acmi.0.000736.v1

© 2023 The Authors. This is an open-access article distributed under the terms of the Creative Commons
Attribution License.

7. Introduction

The family

Erysipelotrichaceae

in the order

Erysipelotrichales

currently comprises several genera with

Erysipelothrix

as the family type-genus (1).

Erysipelothrix rhusiopathiae

(ER), a gram-positive facultative

anaerobic rod, is probably the most well-known bacterium of the family

This bacterium was first described

in the late nineteenth century and is known to infect a large variety of species including mammals, birds

and fish with or without causing clinical disease (2). Within the genus

Erysipelothrix, E. tonsillarum

was

described in 1987 (3) and is considered apathogenic for pigs and chickens (3-5) but a potential pathogen in

dogs (6). Among more recently described species in this genus,

E. piscisicarius

has been associated with

disease in pigs, turkeys and ornamental fish (7, 8) while

E. inopinata

(9),

E. larvae

(10),

E. anatis

aquatica

and

E. urinaevulpis

(11) have not yet been reported to be associated with disease.

In animals, the disease caused by ER is termed erysipelas and among livestock this disease is most well

recognised as a problem in pig and turkey production (2). However, erysipelas is also considered an

emerging disease in modern egg production (12-20). The emergence of the disease in laying hens has been

associated with the move from cage to floor housing and housing systems with outdoor access for hens

have an increased risk of erysipelas outbreaks (14-16). On flock level, erysipelas in laying hens manifests as

acute onset and rapid progression outbreaks with increased, up to 60%, mortality and sometimes

decreased egg production (13, 16, 17, 21). Many aspects of this disease still remain unclear, for instance are

the source and route of infection most often unknown. The clinical picture of outbreaks in laying hen flocks

would suggest introduction of a pathogen in a previously naïve population. On the other hand, several

reports have shown antibodies that recognise ER in conventionally reared laying hens with no history of

erysipelas (15, 22-24), which suggests that ER or antigenically similar bacteria could be common in these

chickens or in their environment. However, there are only a few reports on isolation of ER from clinically

healthy chickens (25-27) and additionally a study that fails to isolate ER by culture methods in healthy

laying hen flocks (18). By whole genome sequencing we have found (manuscript in preparation) that ER

from laying hens affected by clinical erysipelas belong to all described (28) genetic clades except clade 1.

Moreover, we have observed that clade 1 was the most common ER in samples from healthy pigs and wild

boar (29). This ER clade has previously been associated with disease in marine mammals (28) and it is

possible that ER of clade 1 is less pathogenic for terrestrial mammals and chickens. One may hypothesise

that ER of less pathogenic genotypes may be the cause of ER recognising antibodies in healthy laying hens.

It is known that under favourable conditions, ER may survive in the environment for weeks (30) but an

important role of carriers as reservoirs and source of ER for clinical outbreaks of erysipelas has also been

put forward (31). To better understand the pathogenesis of erysipelas in laying hens as well as to be able to

prevent this disease it is important to have the means to monitor the presence of ER and other

Erysipelotrichales

in healthy hens and their environment. This has proven to be challenging by traditional

culture methods because low-level presence of ER is easily overlooked in samples where faster-growing

bacterial species are prevalent. Hence, DNA based methodologies might prove beneficial for this purpose.

To gain more insight into the presence of ER and related bacteria in healthy chickens and their

environment this study was conducted to evaluate sample collection, culture methods and DNA-based

methodology to detect

Erysipelotrichales

. For this purpose, we collected samples from an experimental

poultry facility with a conventional biosecurity level. The facility has no history of erysipelas outbreaks but

healthy chickens with high IgY titers recognising ER have previously been identified (unpublished

observation). In addition, we performed a pilot serological analysis using some

Erysipelotrichales

members with IgY from specific pathogen free (SPF) and conventionally reared chickens to monitor the

presence of IgY recognising these bacteria.

8. Methods

100

101

Prepr

int

8.1 Sample collection for bacterial DNA analysis

102

103

Samples were collected at the experimental poultry facility at Foulum research centre, Aarhus University

104

(AU), Denmark. At this facility research chicken lines are bred and maintained, with chicks hatched from

105

eggs produced in-house and raised to adulthood. Adult chickens are group housed, seven hens and one

106

rooster, in furnished cages. Other poultry are kept at the facility for temporary experiments, and other

107

livestock including pigs are kept in nearby buildings at the same research centre. The facility biosecurity is

108

rated as „conventional

production

‟

, i.e., not SPF. No outbreaks of erysipelas have occurred at the facility,

109

but high antibody titres to

ER among adult hens have been observed (authors‟ unpublished observations).

110

A total of 56 samples were collected (Table 1.). Four 10-12 weeks old chickens were sacrificed, and

111

intestinal segments with content were collected from the ileum, caecum and cloaca (total n=12).

112

Environmental samples were collected with gauze wipes for surfaces and gauze overshoes for floor and

113

ground samples, in both cases the gauze was moistened with phosphate buffered saline (PBS) before use.

114

Samples were collected from cages (n=12), manure conveyor belts (n=12), floors in the animal rooms

115

(n=8), facility staff shoes used in the animal room (n=2), the ground outside the poultry facility (n=4) and

116

the ground outside the pig house (n=4). Pooled samples of poultry red mites (

Dermanyssus gallinae

), a

117

suspected ER vector (12), were also collected (n=2).

118

119

8.2 Enrichment and DNA extraction

120

121

All samples were divided into three portions used for the following protocols: 1) no enrichment; suspension

122

in PBS, 2) nonselective enrichment; 48h incubation at 37°C in tryptic soy broth (TSB), and 3) selective

123

enrichment; 48h incubation at 37°C in beef extract broth containing 0.2 mg/mL sodium-azide and

124

g/mL crystal-violet (SACV). Growth media were prepared at the National Veterinary Institute (SVA;

125

Uppsala, Sweden). Supernatants from the three protocols, a total of 168 samples, were frozen at -70°C.

126

DNA was later extracted from thawed supernatants using the IndiMag Pathogen kit (Indical) on a

127

Maelstrom-9600 automated system (TANBead).

128

129

8.3 Real-time PCR detection of

Erysipelothrix

spp.

130

131

A previously described fluorescent probe-based real-time PCR method (32) was used to detect ER,

132

tonsillarum

and

piscisicarius

(previously designated

sp. „strain 2‟

(8)). ER and

tonsillarum

positive

133

controls were used in the assay, however no positive control for

piscisicarius

was available for the

134

present study. The assay was run using PerfeCTa Multiplex qPCR ToughMix reagents (Quantabio) on a

135

CFX96 Touch Real-Time PCR instrument (Bio-Rad). To deliberately reduce the assay specificity in order to

136

detect any other closely related

Erysipelothrix

spp. in the samples a second PCR reaction was performed

137

for each sample excluding the probes and instead running the assay with PerfeCTa SYBR Green SuperMix

138

(Quantabio) and evaluating any PCR products by melting curve analysis and Sanger sequencing.

139

140

8.4 16S rRNA sequencing detection of

Erysipelotrichales

141

142

Partial 16S sequencing for community profiling was performed using the Illumina MiSeq 16S rRNA

143

protocol (33), with libraries verified on a 2100 Bioanalyzer instrument with a High Sensitivity DNA kit

144

(Agilent) and sequenced as 2x300 bp paired-end reads on a with a V3 run kit on a MiSeq instrument

145

(Illumina). The produced sequences from chicken and environmental samples were separately analysed

146

using the QIIME2 16S pipeline (34) implemented on the NIH/NIAID Nephele microbiome analysis

147

platform v. 2.20.2 (35) using the SILVA (RRID:SCR_006423

)

reference database with open OTU strategy

148

and otherwise default settings. The output was processed and visualised with the phyloseq package (36) in

149

R 4.0.4 (r-project.org). To avoid spurious detections due to index jumping (37), all OTU counts below five

150

in a sample were considered negative for that sample. A phylogenetic tree was inferred in Mega X (38)

151

Prepr

int

using the Neighbor-Joining method with pairwise deletion of missing positions, comparing all sequences

152

from putative

Erysipelotrichales

detected in the present study with selected sequences from the reference

153

database, all trimmed to the length of the Illumina protocol PCR product. To find any matches in less well-

154

curated sources all sequences from putative

Erysipelothrix

spp. as identified by QIIME2 were also

155

compared to the full NCBI

„

‟

nucleotide collection with blastn (blast.ncbi.nlm.nih.gov). Genomes of

156

interest identified in the 16S analysis were screened for the presence of ER proteins with known or

157

suspected immunogenicity using local Blast+ (39).

158

159

8.5 Samples for serological analyses

160

161

Serum samples from chickens reared at either of two types of biosecurity levels were included in the study;

162

SPF-reared chickens and chickens conventionally reared at the AU facilities described above. The SPF-

163

chickens were female Dekalb White Leghorn-type layer hybrids purchased from a commercial hatchery

164

and raised from day old at high biosecurity at SVA (Uppsala, Sweden) as earlier described (40). In the

165

current study we used sera collected from six unvaccinated chickens, two 45 days old (chickens #2 and #3)

166

and four 52 days old (chickens #21, 26, 33 and #68), and six vaccinated chickens, two 45 days old

167

(chickens #29 and #32) and four 52 days old (chickens #7, 16, 34 and #60). The vaccinated chickens were

168

injected with 0.5 mL of a commercial inactivated erysipelas vaccine containing the ER strain M2 of

169

serotype 2, belonging to clade 2, (Porcilis ERY Vet, MSD Animal Health) in the breast muscle 28 days prior

170

to sample collection (chickens #29 and #32) or 25 days prior to sample collection (chickens #7, 16, 34 and

171

#60).

172

Conventionally reared chickens were mixed sexes inbred L10 (41, 42) layer hybrid type hatched and reared

173

indoors without outdoor access at the AU animal facilities. After hatch, the chickens were kept in floor pens

174

on wood shavings until 16 weeks of age and subsequently transferred to enriched cages. In the current

175

study, sera collected at 10 weeks of age (male, n=4) and at 48 weeks of age (1 female and 3 male, n=4) were

176

used.

177

178

8.6 ELISA methodology for quantification of antibodies in chicken serum

179

180

A previously described in-house ELISA for quantification of IgY titers to ER in chicken serum samples was

181

used (40). In brief, a sonicated whole-bacteria lysate was used as target coating antigen in the assay and

182

serum samples were titrated in 2-fold dilutions to achieve a dilution curve. For each sample the A450

–

183

A650 values were plotted against the sample dilution and the equation for the linear part of the curve was

184

determined by regression analysis. Antibody titers were then calculated as the dilution that would achieve

185

an A450

–

A650 value of 1. In the current study either antigens prepared from ER,

Holdemania filiformis

186

E. anatis

, respectively, were used as coating antigens. The ER antigen was prepared as earlier described

187

(24) from strain 15-ALD003475 derived from an outbreak of erysipelas in a Swedish laying hen flock in

188

2015 that was classified as belonging to “intermediate” lineage

(28) and according to in silico application of

189

multiplex PCR serotyping (43) it is of serotype 1B. A reference strain of

H. filiformis

(CCUG 39501T) was

190

purchased from Culture collection University of Gothenburg and cultured anaerobically according to a

191

protocol from the supplier. A reference strain of

E. anatis

(DZM 111258) was purchased from Leibniz

‐

192

Institut DSMZ and cultured at microaerophilic conditions according to a protocol from DSMZ. The

193

H. filiformis

and

E. anatis

antigens were subsequently prepared according to the same protocol described

194

for the ER antigen (40). The protein concentrations of all antigens were determined by Bradford protein

195

assay and for coating a final concentration of 5 ug/ml was used for all antigens.

196

197

9. Results

198

199

All chicken intestinal samples as well as all environmental samples were negative for

Erysipelothrix

200

rhusiopathiae

tonsillarum

and

piscisicarius

by real-time PCR, and no positive reactions in the SYBR

201

Prepr

int

Green assay indicated the presence of closely related species. As expected from the heterogenous sample

202

material, 16S rRNA sequencing produced variable read counts from the different categories. The chicken

203

intestinal tract sampling resulted in a total of 38,405 - 171,517 high-quality sequences from each animal

204

combining the three pre-treatment protocols. The two poultry red mite samples yielded low sequence

205

counts of 7,807 and 8,923. In contrast, more sequence data were retrieved from all environmental sample

206

categories with read counts ranging from 109,283 to 414,161 per sample. A single sample from a manure

207

conveyor belt (#89, TSB treatment) failed to produce any sequence data. As expected, either of the two

208

enrichment protocols improved overall sequence recovery across all sample categories with the exception

209

of the poultry red mite samples. QIIME2 analysis revealed the presence of several

Erysipelotrichales

210

genera

in both environmental samples and chicken intestinal samples including

Massiliomicrobiota

211

Erysipelatoclostridium

Turicibacter

Merdibacter

Faecalitalea

and

Dielma

. The environmental samples

212

additionally contained representatives of

Longicatena

the „

Clostridium

‟

innocuum

group,

Holdemania

213

and

Erysipelothrix

(Table 1, Supp. Fig. 1). The sequences consistent with the genus

Erysipelothrix

were

214

found in a single shoe sample, four chicken cage samples and ten samples from manure conveyor belts.

215

The selective SACV enrichment was the most successful in producing

Erysipelothrix

-positive samples (14

216

samples), followed by PBS (7 samples) and TSB (1 sample). The fraction of observed sequences consistent

217

with the

Erysipelothrix

genus was ≤ 1% of the total sequence count for all these samples.

218

219

Consistent with real-time PCR results, no 16S sequences indicated the presence of the more well-known

220

Erysipelothrix

species (Supp. Fig. 1). However, BLAST searches against the nr sequence database revealed

221

close matches.

Erysipelothrix

sequence #2 in the present study is highly similar (99.6%) to the 16S

222

sequences of the two recently proposed novel species

E. anatis

sp. nov., and

E. aquatica

sp. nov. (11).

223

These species are closely related and cannot be differentiated by their 16S sequences (11). Matches in

224

GenBank to this genotype included isolates from pigs, geese, ducks and a turtle as well as two species of

225

beetle and a leech. No other putative

Erysipelothrix

sequences found in the present study matched any

226

bacterial isolate in the database. However, sequence #1 closely (99%) matches uncultured bacteria in swine

227

waste (GenBank GQ138120.1), and sequences #3 and #7 have a high similarity (99%) to uncultured

228

bacteria in a leafhopper (KC864783.1) and a manure sample (KP745023.1). For the remaining

229

Erysipelothrix

sequences, no relevant matches (>97%) were found.

230

231

The genomes of

E. anatis

sp. nov., and

E. aquatica

sp. nov. were investigated for the presence of sequences

232

similar to those encoding putative immunogenic proteins in ER. For both genomes, GAPDH was present

233

and similar to ER (100% coverage, 87% similarity in both). Genes similar to ER

Atsp

(92-98% coverage,

234

73-74% similarity) and to a lesser degree

dap2

(51-53%, 69-71%) were also found in both genomes,

235

whereas no close matches to the genes

spaA, rspA, CbpB, Plp, Neu, Bga, Bml, CwpA

, or

CwpB

were found.

236

237

Serological analysis of IgY antibodies to ER,

E. anatis

and

H. filiformis

was performed on samples from

238

SPF-reared chickens as well as from chickens reared at the AU research facilities (Figure 1). For

239

unvaccinated SPF-chickens reared under high biosecurity low levels of IgY that recognised either

240

bacterium were detected. However, the SPF-chickens that had received a dose of a commercial vaccine

241

against erysipelas showed high IgY titers to ER, >10-fold higher than those of the unvaccinated SPF-

242

chickens. For the vaccinated SPF-chickens a trend of slightly higher titers to

E. anatis

compared to those of

243

unvaccinated SPF-chickens was observed while titers to

H. filiformis

were low. For the conventionally

244

reared chickens from the AU research facilities variation between individuals in the levels of IgY titers to all

245

three bacteria was observed. In general, the conventionally reared chickens showed higher titers compared

246

to the unvaccinated SPF-chickens and some individuals showed titers at comparable levels to those

247

observed to ER for some of the vaccinated SPF-chickens. However, unlike the vaccinated SPF-chickens, for

248

all conventionally reared chickens except #2429 the titer levels to all three bacteria within chicken were

249

similar. Chicken #2429 on the other hand showed a clear bias with the highest titer to ER and low titers to

250

the other two bacteria.

251

252

Prepr

int

We have previously noted that chickens housed at the AU research facility have high serum titers of IgY

267

that recognise ER, even though the facility has never experienced any outbreak of clinical erysipelas in

268

chickens (unpublished observation). Given our recent observations of the occurrence of ER clade 1 in pigs

269

and wild boars without clinical signs of disease (29) and the increasing number of novel related species

270

discovered in a broad range of animal species in recent years (7, 9-11), our hypothesis was that the presence

271

of low-virulence

Erysipelothrix

sp. strains could explain the seropositive status of the chickens. Identifying

272

such strains would be of interest both for better understanding ER evolution and pathogenesis as well as

273

for improving management practices e.g., vaccination programs. Additionally, the authors have conducted

274

several experimental ER infection studies in recent years and found that even when using a strain

275

recovered from a severe outbreak in a production flock, infection under experimental conditions can be

276

associated with limited or no clinical signs (40, 44). As an alternative hypothesis, virulent ER could thus be

277

present at AU but not cause overt disease in the chickens due to host protective factors e.g., a robust health

278

status and low levels of stress. To discover ER shed only intermittently or by a small number of animals we

279

therefore performed extensive environmental sampling of the facility but could not detect ER DNA in any

280

of the samples with real-time PCR or with 16S rRNA sequencing. The presence of ER among the animals

281

therefore seems a more unlikely explanation, although the bacteria could of course have been present at an

282

earlier date and cleared from the environment when sampled.

283

284

To detect

Erysipelothrix

spp. other than ER we used amplicon-based high-throughput 16S rRNA

285

sequencing, amplifying a 460 bp fragment of the V3-V4 region with primers targeting conserved regions

286

(33). While unbiased, this approach produces only partial 16S sequences and can therefore not reliably

287

identify bacteria to species level; previous studies have also found that even complete 16S sequences do not

288

always provide good resolution between closely related

Erysipelothrix

species (8). Nonetheless, we note

289

that multiple species of

Erysipelothrix

appear to be present in environmental samples from the facility and

290

in particular in chicken cages and on the chicken manure conveyor belts. A near-perfect match was for

291

instance observed to

E. anatis

which has been observed in ducks (11), and there were several sequences

292

matching other undescribed and non-cultured bacteria related to the known species of

Erysipelothrix

293

These observations are consistent with transient or sporadic infection of the chickens with multiple species

294

of presumably low-virulence

Erysipelothrix

, or possibly introduction to the cages via bedding material,

295

feed, or arthropods. The highest number of samples positive for

Erysipelothrix

spp. were achieved with

296

SACV broth, which is a long-standing standard protocol for enrichment of ER (45). A tryptic soy variant of

297

SACV broth has also been found the most efficient pre-enrichment for detection of ER by PCR in seafood

298

samples (46) This suggests that SACV can be useful in future efforts to isolate and further characterize

299

some of the putative species of

Erysipelothrix

observed by us and in other studies. However, even with

300

SACV the read counts were low representing a minute fraction of the bacteria in the enrichment broth, and

301

isolation would likely have been challenging.

302

303

To assess potential antibody cross reactivity of chicken IgY between ER and other

Erysipelothrix

species

304

consistent with the observations made in the 16S rRNA analysis, we performed quantification of IgY to ER,

305

E. anatis

and

H. filiformis

in sera from chickens raised at the AU facility and from SPF-chickens raised

306

under high biosecurity.

E. anatis

was chosen as a representative of

Erysipelothrix

species that were

307

indicated at the AU facility.

H. filiformis

is the type species of

Holdemania

that were also present at the AU

308

facility.

Holdemania

is one of the genetically closest genera to

Erysipelothrix

(1) and contains the

309

uncommon type B cell wall murein in common with

Erysipelothrix

(47). Results showed that chickens

310

raised under high biosecurity had low titers to these three bacteria and under such circumstances

311

vaccination with inactivated ER induced a clear IgY response with a very strong bias towards ER. IgY from

312

ER vaccinated SPF-chickens showed a low but noticeable tendency to cross-react with

E. anatis

, which is

313

in accordance with the bioinformatic analysis that showed that

E. anatis

shared some immunogenic

314

proteins with ER but lacked for instance

spaA

that is considered a major immunogen in ER (48). Results

315

from the conventionally reared chickens were however more difficult to interpret. With one exception these

316

chickens showed titers of similar levels to all three bacteria. Several explanations to this observation are

317

Prepr

int

possible, e.g., that infection/colonisation with live bacteria might induce more cross-reactive antibodies

318

compared to vaccination with inactivated bacteria. It is also possible that natural, more polyreactive,

319

antibodies (49) may be more common in the conventionally reared chickens. Further research is thus

320

needed to understand the origin of the sometimes high titers of IgY recognising ER in conventionally

321

reared chickens.

322

323

12. Author statements

324

12.1 Author contributions

325

Conceptualization and funding: EW, RS and HE. Investigation: EW, TSD and RS. Methodology and formal

326

analysis: RS, EW and TSD. Data curation, visualization, writing

–

original draft and project management:

327

RS and EW. Writing

–

review and editing: all authors.

328

329

12.2 Conflicts of interest

330

The authors declare there are no conflicts of interest.

331

332

12.3 Funding information

333

This study received financial support from the Albert Hjärre Fund, the Swedish Research Council Formas

334

(grant numbers 942-2015-766 and 2019-01270) and the EryPoP-project financed by Animal Health

335

and Welfare ERA-Net (ANIHWA) under the European Union Seventh Framework Network (ID number

336

119, in Sweden grant number 221-2015-189). The funders had no role in study design, data collection and

337

interpretation, or the decision to submit the work for publication.

338

339

12.4 Ethical approval

340

Animals were housed at AU under protocols approved by the Danish Animal Experiments Inspectorate and

341

complied with the Danish Ministry of Justice Law number 382 (10th June 1987) and Acts 739

342

(6thDecember 1988) and 333 (19th May 1990) concerning animal experimentation and care of

343

experimental animals, following the described ethical guidelines. Licence number 2017-15-0201-01211.

344

SPF-chickens were housed at the animal facilities at SVA, approved for experimental animals by the

345

Swedish Board of Agriculture, and vaccination and blood sampling were approved by the Uppsala regional

346

Ethical Committee for Animal Experiments, permit no. C46/16, according to Swedish legislation and

347

directives (SJVFS 2019:9 L 150) based on European Union legislation directive 2010/63/EU.

348

349

12.6 Acknowledgements

350

The authors wish to thank Dr Harri Ahola, Karin Ullman, Tomas Jinnerot and Lene Rosborg Dal for expert

351

technical assistance and the staff at the animal facilities at AU and at SVA for excellent animal care.

352

353

References

354

Verbarg S, Göker M, Scheuner C, Schumann P, Stackebrandt E. The Families Erysipelotrichaceae

355

emend., Coprobacillaceae fam. nov., and Turicibacteraceae fam. nov. In: Rosenberg E, DeLong EF, Lory S,

356

Stackebrandt E, Thompson F, editors. The Prokaryotes: Firmicutes and Tenericutes. Berlin, Heidelberg:

357

Springer Berlin Heidelberg; 2014. p. 79-105.

358

Wang Q, Chang BJ, Riley TV. Erysipelothrix rhusiopathiae. Veterinary microbiology. 2010;140(3-

359

4):405-17.

360

Takahashi T, Fujisawa T, Benno Y, Tamura Y, Sawada T, Suzuki S, et al. Erysipelothrix

361

tonsillarum sp. nov. Isolated from Tonsils of Apparently Healthy Pigs. International Journal of Systematic and

362

Evolutionary Microbiology. 1987;37(2):166-8.

363

Harada K, Muramatsu M, Suzuki S, Tamura Y, Sawada T, Takahashi T. Evaluation on the

364

pathogenicity of Erysipelothrix tonsillarum for pigs by immunosuppression with cyclophosphamide or

365

dexamethasone. Res Vet Sci. 2011;90(1):20-2.

366

Takahashi T, Takagi M, Yamaoka R, Ohishi K, Norimatsu M, Tamura Y, et al. Comparison of the

367

pathogenicity for chickens of Erysipelothrix rhusiopathiae and Erysipelothrix tonsillarum. Avian Pathol.

368

1994;23(2):237-45.

369

Takahashi T, Tamura Y, Yoshimura H, Nagamine N, Kijima M, Nakamura M, et al. Erysipelothrix

370

tonsillarum isolated from dogs with endocarditis in Belgium. Res Vet Sci. 1993;54(2):264-5.

371

Prepr

int

Pomaranski EK, Griffin MJ, Camus AC, Armwood AR, Shelley J, Waldbieser GC, et al.

372

Description of Erysipelothrix piscisicarius sp. nov., an emergent fish pathogen, and assessment of virulence

373

using a tiger barb (Puntigrus tetrazona) infection model. Int J Syst Evol Microbiol. 2020;70(2):857-67.

374

Grazziotin AL, Vidal NM, Hoepers PG, Reis TFM, Mesa D, Caron LF, et al. Comparative genomics

375

of a novel clade shed light on the evolution of the genus Erysipelothrix and characterise an emerging species.

376

Scientific Reports. 2021;11(1):3383.

377

Verbarg S, Rheims H, Emus S, Frühling A, Kroppenstedt RM, Stackebrandt E, et al.

378

Erysipelothrix inopinata sp. nov., isolated in the course of sterile filtration of vegetable peptone broth, and

379

description of Erysipelotrichaceae fam. nov. International Journal of Systematic and Evolutionary Microbiology.

380

2004;54(1):221-5.

381

10.

Bang B-H, Rhee M-S, Chang D-H, Park D-S, Kim B-C. Erysipelothrix larvae sp. nov., isolated

382

from the larval gut of the rhinoceros beetle, Trypoxylus dichotomus (Coleoptera: Scarabaeidae). Antonie van

383

Leeuwenhoek. 2015;107(2):443-51.

384

11.

Eisenberg T, Muhldorfer K, Erhard M, Fawzy A, Kehm S, Ewers C, et al. Erysipelothrix anatis sp.

385

nov., Erysipelothrix aquatica sp. nov. and Erysipelothrix urinaevulpis sp. nov., three novel species of the genus,

386

and emended description of Erysipelothrix. Int J Syst Evol Microbiol. 2022;72(7).

387

12.

Chirico J, Eriksson H, Fossum O, Jansson D. The poultry red mite, Dermanyssus gallinae, a

388

potential vector of Erysipelothrix rhusiopathiae causing erysipelas in hens. Med Vet Entomol. 2003;17(2):232-4.

389

13.

Mazaheri A, Lierz M, Hafez HM. Investigations on the Pathogenicity of <span class="genus-

390

species">Erysipelothrix rhusiopathiae</span> in Laying Hens. Avian Diseases. 2005;49(4):574-6, 3.

391

14.

Kaufmann-Bart M, Hoop RK. Diseases in chicks and laying hens during the first 12 years after

392

battery cages were banned in Switzerland. Veterinary Record. 2009;164(7):203-7.

393

15.

Eriksson H, Nyman A-K, Fellström C, Wallgren P. Erysipelas in laying hens is associated with

394

housing system. Veterinary Record. 2013;173(1):18-.

395

16.

Fossum O, Jansson DS, Etterlin PE, Vågsholm I. Causes of mortality in laying hens in different

396

housing systems in 2001 to 2004. Acta veterinaria Scandinavica. 2009;51(1):3.

397

17.

Stokholm NM, Permin A, Bisgaard M, Christensen JP. Causes of Mortality in Commercial

398

Organic Layers in Denmark. Avian Diseases. 2010;54(4):1241-50, 10.

399

18.

Eriksson H, Bagge E, Båverud V, Fellström C, Jansson DS. Erysipelothrix rhusiopathiae

400

contamination in the poultry house environment during erysipelas outbreaks in organic laying hen flocks. Avian

401

Pathology. 2014;43(3):231-7.

402

19.

Schmitt F, Schade B, Böhm B, Shimoji Y, Pfahler C. Erysipelas in a free-range layer flock with

403

conjunctival oedema as an unusual clinical sign. Berl Munch Tierarztl Wochenschr. 2014;127(5-6):183-7.

404

20.

Crespo R, Bland M, Opriessnig T. Use of Commercial Swine Live Attenuated Vaccine to Control

405

an <i>Erysipelothrix rhusiopathiae</i> Outbreak in Commercial Cage-Free Layer Chickens. Avian Diseases.

406

2019;63(3):520-4, 5.

407

21.

Bisgaard M, Olsen P. Erysipelas in egg-laying chickens: clinical, pathological and bacteriological

408

investigations. Avian Pathol. 1975;4(1):59-71.

409

22.

Takahashi T, Takagi M, Yamamoto K, Nakamura M. A serological survey on erysipelas in

410

chickens by growth agglutination test. J Vet Med B Infect Dis Vet Public Health. 2000;47(10):797-9.

411

23.

Kurian A, Neumann EJ, Hall WF, Marks D. Serological survey of exposure to Erysipelothrix

412

rhusiopathiae in poultry in New Zealand. N Z Vet J. 2012;60(2):106-9.

413

24.

Wattrang E, Eriksson H, Albihn A, Dalgaard TS. Quantification of IgY to Erysipelothrix

414

rhusiopathiae in serum from Swedish laying hens. BMC veterinary research. 2021;17(1):111.

415

25.

Van Damme LR, Devriese LA. The Presence of Erysipelothrix rhusiopathiae in the Tonsils of

416

Swine and in the Larynx of Chidtens in Rwanda (Central Africa. Zentralblatt für Veterinärmedizin Reihe B.

417

1976;23(1):74-8.

418

26.

Nakazawa H, Hayashidani H, Higashi J, Kaneko K, Takahashi T, Ogawa M. Occurrence of

419

Erysipelothrix spp. in chicken meat parts from a processing plant. Journal of food protection. 1998;61(9):1207-

420

421

27.

Nakazawa H, Hayashidani H, Higashi J, Kaneko K, Takahashi T, Ogawa M. Occurrence of

422

Erysipelothrix spp. in broiler chickens at an abattoir. Journal of food protection. 1998;61(7):907-9.

423

28.

Forde T, Biek R, Zadoks R, Workentine ML, De Buck J, Kutz S, et al. Genomic analysis of the

424

multi-host pathogen Erysipelothrix rhusiopathiae reveals extensive recombination as well as the existence of

425

three generalist clades with wide geographic distribution. BMC genomics. 2016;17:461.

426

29.

Söderlund R, Formenti N, Caló S, Chiari M, Zoric M, Alborali GL, et al. Comparative genome

427

analysis of Erysipelothrix rhusiopathiae isolated from domestic pigs and wild boars suggests host adaptation

428

and selective pressure from the use of antibiotics. Microbial Genomics. 2020;6(8).

429

30.

Wood RL. Survival of Erysipelothrix rhusiopathiae in soil under various environmental

430

conditions. Cornell Vet. 1973;63(3):390-410.

431

31.

Opriessnig T, Coutinho TA. Erysipelas. Diseases of Swine2019. p. 835-43.

432

32.

Pal N, Bender JS, Opriessnig T. Rapid detection and differentiation of Erysipelothrix spp. by a

433

novel multiplex real-time PCR assay. Journal of applied microbiology. 2010;108(3):1083-93.

434

33.

Illumina. 16S Metagenomic Sequencing Library Preparation. Rev. B2013.

435

Prepr

int

34.

Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible,

436

interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37(8):852-7.

437

35.

Weber N, Liou D, Dommer J, MacMenamin P, Quinones M, Misner I, et al. Nephele: a cloud

438

platform for simplified, standardized and reproducible microbiome data analysis. Bioinformatics.

439

2018;34(8):1411-3.

440

36.

McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and

441

graphics of microbiome census data. PloS one. 2013;8(4):e61217.

442

37.

Anonymous. Effects of Index Misassignment on Multiplexing and Downstream Analysis2017.

443

38.

Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular Evolutionary Genetics

444

Analysis across Computing Platforms. Molecular biology and evolution. 2018;35(6):1547-9.

445

39.

Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+:

446

architecture and applications. BMC bioinformatics. 2009;10:421.

447

40.

Wattrang E, Eriksson H, Jinnerot T, Persson M, Bagge E, Söderlund R, et al. Immune responses

448

upon experimental Erysipelothrix rhusiopathiae infection of naïve and vaccinated chickens. Veterinary research.

449

2020;51(1):114.

450

41.

Laursen SB, Hedemand JE, Nielsen, Thiel OE, Koch C, Jensenius JC. Serum levels, ontogeny and

451

heritability of chicken mannan-binding lectin (MBL). Immunology. 1998;94(4):587-93.

452

42.

Juul-Madsen HR, Munch M, Handberg KJ, Sorensen P, Johnson AA, Norup LR, et al. Serum

453

levels of mannan-binding lectin in chickens prior to and during experimental infection with avian infectious

454

bronchitis virus. Poultry Science. 2003;82(2):235-41.

455

43.

Shimoji Y, Shiraiwa K, Tominaga H, Nishikawa S, Eguchi M, Hikono H, et al. Development of a

456

Multiplex PCR-Based Assay for Rapid Serotyping of Erysipelothrix Species. Journal of clinical microbiology.

457

2020;58(6).

458

44.

Wattrang E, Jaderblom V, Jinnerot T, Eriksson H, Bagge E, Persson M, et al. Detection and

459

quantification of Erysipelothrix rhusiopathiae in blood from infected chickens - addressing challenges with

460

detection of DNA from infectious agents in host species with nucleated red blood cells. Journal of medical

461

microbiology. 2019;68(7):1003-11.

462

45.

Packer RA. The Use of Sodium Azide (NaN(3)) and Crystal Violet in a Selective Medium for

463

Streptococci and Erysipelothrix rhusiopathiae. Journal of bacteriology. 1943;46(4):343-9.

464

46.

Fidalgo SG, Wang Q, Riley TV. Comparison of methods for detection of Erysipelothrix spp. and

465

their distribution in some Australasian seafoods. Applied and environmental microbiology. 2000;66(5):2066-

466

70.

467

47.

Willems A, Moore WEC, Weiss N, Collins MD. Phenotypic and Phylogenetic Characterization of

468

Some Eubacterium-Like Isolates Containing a Novel Type B Wall Murein from Human Feces: Description of

469

Holdemania filiformis gen. nov., sp. nov. International Journal of Systematic and Evolutionary Microbiology.

470

1997;47(4):1201-4.

471

48.

Imada Y, Mori Y, Daizoh M, Kudoh K, Sakano T. Enzyme-linked immunosorbent assay

472

employing a recombinant antigen for detection of protective antibody against swine erysipelas. Journal of

473

clinical microbiology. 2003;41(11):5015-21.

474

49.

Reyneveld GI, Savelkoul HFJ, Parmentier HK. Current Understanding of Natural Antibodies and

475

Exploring the Possibilities of Modulation Using Veterinary Models. A Review. Front Immunol. 2020;11:2139.

476

477

Prepr

int