Type of article:

Original research article

Title:

Gene expression of AvBD6-10 in broiler chickens is independent of AvBD6, 9, and 10 peptide

potency

Authors:

Catherine A. Mowbray1, Sherko S. Niranji1a,b, Kevin Cadwell1a, Richard Bailey2, Kellie A Watson2 and

Judith Hall1*

1Institute for Cell and Molecular Biosciences, Newcastle University, Newcastle upon Tyne, NE2 4HH,

UK and 2Aviagen Ltd, Newbridge, Midlothian, EU28 8SZ, UK.

a These authors contributed equally to the work

b Present address: Garmian University Research Centre and Department of Biology, College of

Education, University of Garmian, KRG, Iraq

*Corresponding Author:

Judith Hall, Institute for Cell and Molecular Biosciences, Newcastle University, Newcastle upon Tyne,

NE2 4HH, UK. Tel:+44 (0) 191 208 8346; Fax :+44 (0) 191 208 7424 ; Email :Judith.Hall@ncl.ac.uk

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

Title: Gene expression of AvBD6-10 in broiler chickens is independent of AvBD6, 9, and 10 peptide

potency

Abstract

The Avian β-defensin (AvBD) gene cluster contains fourteen genes; within this, two groups (AvBD6/7

and AvBD8 -10) encode charged peptides of >+5 (AvBD6/7), indicative of potent microbial killing

activities, and +4 (AvBD8-10), suggestive of reduced antimicrobial activities. Chicken broiler gut

tissues are constantly exposed to microbes in the form of commensal bacteria. This study examined

whether tissue expression patterns of AvBD6-10 reflected microbial exposure and the encoded

peptides a functional antimicrobial hierarchy.

Gut AvBD6-10 gene expression was observed in hatch to day 21 birds, although the AvBD8-10

profiles were eclipsed by those detected in the liver and kidney tissues. In vitro challenges of chicken

CHCC-OU2 cells using the gut commensal Lactobacillus johnsonii (104 CFU) did not significantly affect

AvBD8-10 gene expression patterns, although upregulation (P<0.05) of IL-Iβ gene expression was

observed. Similarly, in response to Bacteriodes doreii, IL-Iβ and IL-6 gene upregulation were detected

(P<0.05), but AvBD10 gene expression remained unaffected. These data suggested that AvBD8-10

gene expression was not induced by commensal gut bacteria.

Bacterial time-kill assays employing recombinant (r)AvBD6, 9 and 10 peptides (0.5µM - 12µM),

indicated an antimicrobial hierarchy, linked to charge, of AvBD6>AvBD9>AvBD10 against Escherichia

coli, but AvBD10>AvBD9>AvBD6 using Enterococcus faecalis. rAvBD10, selected due to its reduced

cationic charge was, using CHCC-OU2 cells, investigated for cell proliferation and wound healing

properties, but none were observed.

2

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

These data suggest that in healthy broiler chicken tissues AvBD6/7 and AvBD8-10 gene expression

profiles are independent of the in vitro antimicrobial hierarchies of the encoded AvBD6, 9 and 10

peptides.

Key Words: Broiler; β-defensins; gene profile; antimicrobial killing; gut

Introduction

Defensins are small <5KDa cysteine-rich peptides that play key protective roles in the innate

defences of invertebrates and vertebrates. These molecules group, structurally, into three distinct

families called α, β and θ defensins, are characterised by six cysteines and collectively exhibit

antimicrobial activity against either bacteria or fungi, as well as some viruses and parasites. Unlike

the α or θ defensins, the β-defensins, in which the six cysteines form three disulphide bridges in a

Cys1-5, Cys2-4 and Cys3-6 pattern, are found in almost all vertebrate species including fish,

amphibians, birds and mammals (Jenssen et al., 2006; van Dijk et al., 2008). In chickens, 14 β-

defensin genes have been identified and cluster to a 86kb region of chromosome 3 (Cuperus et al.,

2013). While this number suggests regular refreshment of the gene family (Cheng et al., 2015), it is

small compared to humans and cattle where as many as 30 and 57 β-defensin genes, respectively,

cluster on four different chromosomes (Hollox et al., 2003; Meade et al., 2014; Rodriguez-Jimenez et

al., 2003).

Evolutionary analyses have shown that the extensive defensin gene numbers have been driven by

gene duplication events and this is supported by significant regions of peptide sequence homology.

Phylogenetic linkage between β-defensin genes of distant species predicts that the vertebrate β-

defensin gene families arose from a common ancestral gene or genes that expanded through gene

duplication, but after the bird-mammal split (Cheng et al., 2015). Factors driving the divergence have

3

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

been reported to include the changing and increasing microbial challenges that each species has

encountered throughout evolution, with increased numbers of host antimicrobials supporting an

increased chance of survival (Tu et al., 2015).

In relation to their antimicrobial properties it is recognized that the cationic properties exhibited by

β-defensins facilitate electrostatic interactions with the negatively charged lipids of microbial

membranes. This, plus their hydrophobicity, allows them to insert into such membranes and cause

killing (Cuperus et al., 2013). In primates, β-defensins appear to have evolved to include further roles

such as wound healing and signalling from the innate to the adaptive immune system (Lehrer, 2004).

This is illustrated particularly by human β-defensin 2 (BD2), which is shown to be chemotactic for

both dendritic and T cells, and able to induce mast cell migration (Otte et al., 2008). Roles distinct

from innate defence have also been identified. For example, defensin genes clustered on

chromosome 8 are strongly expressed in the male reproductive tract, with deletion studies

suggesting an influence on sperm motility and function, and hence overall fertility (Zhou et al.,

2013).

While the human β-defensins have been studied extensively, less is known about the avian peptides.

The AvBD genes comprise up to four exons encoding a signal peptide, a small pro-piece and the

mature peptide. Peptide structures of AvBD2 and the penguin AvBD103b (Spheniscin-2) suggest a

three-stranded β-sheet structure, characteristic of the human β-defensins, although the former

appears to lack an N-terminal α-helix often linked to membrane insertion and bacterial killing

(Derache et al., 2012). A further set of genes encoding the ovodefensins has been described in birds,

with OvoDA1 (gallin) expressed in the oviduct and responsive to oestrogen and progesterone levels

(Gong et al., 2010). The encoded peptides, also characterised by a β-sheet structure and di-sulphide

bridges, have been shown to be antimicrobial and hence are presumed to play a role in egg defence

(Herve et al., 2014).

4

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

The chicken β-defensin gene cluster of chromosome 3 [3q3.5-q3.7] is flanked by the Cathepsin B

(CTSB) and translocation associated membrane protein 2 (TRAM2) genes, and the gene order is

highly conserved (Cheng et al., 2015; Xiao et al., 2004). The AvBD6 and 7 genes, and the AvBD8, 9

and 10 genes lie adjacent to each other in this cluster with the former group linked to gene

duplication and encoding peptides with an average net cationic charge of >+5, and the latter

encoding peptides carrying charge +4. Assuming net charge links to antimicrobial potency (Kluver

et al., 2005) this predicts the AvBD8, 9 and 10 peptides to exhibit reduced killing properties (Lee et

al., 2016). Focussing on young broiler chickens, which are dependent on their innate immune

defences for protection, this study explored whether the AvBD6-10 gene expression profiles of the

gut, liver and kidney tissues of hatch to day 21 birds reflected their potentially differing functions

such as antimicrobial killing, wound healing and signalling in different tissues. Additionally,

recombinant 6, 9 and 10 peptides were used to explore potential antimicrobial hierarchies between

the peptides encoded by the AvBD6/7 and AvBD8-10 gene groups.

Materials and Methods

Peptide Synthesis & Purification

pGEX-6P-1 plasmids (GE Healthcare) containing AvBD6, 9 and 10 cDNAs and mutated versions

thereof (Table 1) were each transformed into BL21 (DE3) pLysS, and colonies resistant to ampicillin

(50 µg/ml) and chloramphenicol (30µg/ml) selected. Single colonies were inoculated into and

cultured overnight in 10ml Luria Broth (LB) containing antibiotic. Each culture was added to 1L LB,

shaken to OD600 of 0.8-1.0, peptide synthesis induced using iPTG (1M), and following further 2-3h of

shaking the bacterial pellets were collected by centrifugation. Each pellet was resuspended in 20ml

PBS, sonicated, recentrifuged and the supernatants collected. The supernatants were passed

through glutathione sepharose (GS) columns, GS tags removed on the columns using preScission

protease and the pure peptides collected using a 10 kDa Vivaspin column (Vivaspin 20 columns, GE

5

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

Healthcare Life Sciences). Purified peptides were analysed using NUPAGE gel electrophoresis with

precast gels, prior to lyophilisation and storage at 4oC. All peptides were verified by MALDI-MS/MS

(York University Proteomics).

Bacterial Growth & Antimicrobial assays

Bacteria used in the study were isolated post-mortem from the gastrointestinal tracts of broiler

chickens reared on commercial farms as previously described (Cadwell et al., 2017). Antimicrobial

time-kill assays were conducted in LB (salt concentration 87.5mM) diluted with 1x phosphate

buffered saline (PBS: salt concentration 137mM). Bacterial culture, time-kill, radial diffusion and

calcein leakage assays were performed as described previously (Cadwell et al., 2017). Lyophilised

peptides were resuspended to a working stock of 100µg- 1mg/ml.

Synthetic Peptides

AvBD 6 and 9 synthetic peptides were synthesised by PeptideSynthetics (Hampshire, UK) with >95%

purity. Lyophilised peptide was stored at -20°C and a working stock of 1mg/ml prepared by

dissolving 1mg peptide in 20µl of 10% acetic acid and the volume increased to 1ml using Milli-Q

water.

Structure Modelling

The AvBD peptide structures were modelled using Raptor X online software

(http://raptorx.uchicago.edu/) (Peng et al., 2011). The 3D structure prediction used the Penguin

AvBD 103b (Spheniscin-2) structure, solved previously using two dimensional NMR, as the template

(Landon et al., 2004).

Birds & Rearing Environment

Pure line Ross 308 broiler chicks were placed at one day old on a commercial farm and managed

under standard commercial conditions. The birds were vaccinated and fed a standard broiler diet in

6

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

line with industry recommendations

(http://en.aviagen.com/assets/Tech_Center/Ross_Broiler/Ross308BroilerNutritionSpecs2014-

EN.pdf). Five birds were selected at random from the flock at day 0 (prior to placement), 7 and 21

days and euthanised by trained personnel. Tissues were removed and stored at -80oC in RNA later

(ThermoFisher).

Immunohistochemistry (IHC)

A rabbit polyclonal antibody to AvBD9 was produced by Cambridge Research Biochemicals

(Cleveland, U.K.) using the unique peptide antigen, LASRQSHGS-amide and used in the IHC analyses.

Avian tissues were prepared as described previously (Cadwell et al., 2017), with antigen retrieval

performed by pressure cooking in EDTA (pH8.0). Antibody was used at 1:70 dilution in TBS (pH7.6)

for 1 hour at room temperature, and staining protocol completed using Vectastain Elite ABC

Peroxidase Kit (Vector Laboratories, Peterborough, UK) as per manufacturer’s instructions. The

reaction was developed using the peroxidase chromogen DAB (3,3-diaminobenzedine

tetrahydrochloride) and the nuclei counterstained using Mayer’s Haematoxylin and Scot’s tapwater

substitute. A no primary antibody control was conducted in parallel.

In vitro Cell Culture/Challenge

Chicken CHCC-OU2 cells (Ogura et al., 1987) were cultured at 41oC in 5% CO2 in high glucose DMEM

medium containing 5% FCS, 1% chicken serum and 10% tryptose buffered phosphate solution

(Sigma). Cells were seeded into 12-well plates (1 x105 cells/well) and once confluent challenged with

heat killed bacteria (102- 104/well) for up to 24 hours. The cells were washed in phosphate-buffered

saline (PBS) and lysed for RNA extraction and qPCR.

RNA

RNA was prepared as described previously (Cadwell et al., 2017). Quantitative PCR was performed

using Sybr Green mastermix (Roche) with primers and annealing conditions specific for each gene of

7

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

interest (Table 1). Primers were designed to amplify over an exon-exon junction to eliminate

amplification of any remaining genomic DNA and products were verified by cloning and sequencing.

qPCR analysis was carried out using the LightCycler 480 (Roche) with the following program: 95 oC 10

mins, 45 cycles of 95oC 10 seconds, ToC 10 seconds, 72oC 5 seconds, followed by melt curve analysis

to confirm generation of a single product. Negative, without RT, and positive, cloned plasmid,

controls were included on each plate. GeNorm (Primer Design, UK)(Vandesompele et al., 2002) was

used to identify suitable reference genes – data indicated that two reference genes would be

suitable for analyses of these data, with the two most stable being SDHA and SF3A1. Standard curves

for each assay were applied to raw Ct values and the resulting data were normalised to the

geometric mean of reference genes SDHA and SF3A1 and presented as arbitrary units (AU). Data are

comparable within a single defensin regardless of day, e.g. AvBD8 expression is comparable between

day 0, day 7 and day 21 and between tissues. Expression is not comparable between defensins, e.g.

values for AvBD8 cannot be compared to AvBD9.

Wound Healing Assays

CHCC-OU2 cells were seeded into 6-well plates, cultured until confluent and a wound healing assay

performed to test cell migration. An injury line was made using a 2mm-wide plastic pipette tip. After

incubation, the excess liquid in the wells was removed, the wells were rinsed with PBS and covered

with medium. The cells were incubated in complete medium for up to 72 hours to allow the cells to

grow, along with PBS (control) or different concentrations of AvBD10 peptide (0.1, 0.5 and 1nM).

Photographs were acquired at different time points and cell migration areas analysed using ImageJ

software to quantify the area between the two sides of the scratch.

Cell Proliferation Assays

The CellTiter96AQueous cell proliferation assay (Promega) was used to measure metabolically active

cells following bacterial challenge, and the CellTitre-Blue assay (Promega) was used to assess the

8

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

number of viable cells following incubation with either PBS, Bovine Serum Albumin (10nM, generic

control), mitomycin C (10nM) or AvBD10 (10nM).

Statistical Analyses

Statistical analyses were performed using the Prism 6 Software package (GraphPad Software Inc, La

Jolla, California, USA). For analyses of data involving more than two groups, ANOVA followed by

either Tukey’s multiple comparison or Bonferroni post-tests, as appropriate, was used. The

significance level was set at 5% (P<0.05).

Results

Bird Tissue AvBD Gene Expression Patterns

The groups of AvBD genes, AvBD6/7 and AvBD8-10, lie adjacent to each other on chromosome 3. To

explore whether these groupings were associated with specific gene expression profiles, the

AvBD6/7 and AvBD8-10 gut tissue expression profiles (duodenum (D), jejunum (J), ileum (I) and

caecum (C)) of hatch, day 7 and day 21 broiler chickens predicted to be in direct contact with

microbes, namely commensal gut populations, and tissues (kidney (K) and liver (L)), exposed

primarily to microbe-associated molecular patterns, were analysed. At day 0, AvBD 8, 9 and 10 gene

expression was detected in all tissues, but the values observed in the gut were reduced compared to

those of the kidney and liver. These patterns were maintained at days 7 and 21 post hatch, with

duodenal expression becoming more prominent than in other gut tissues in the older birds (Fig 1A-

D). AvBD9 peptide was detected by IHC in day 7 duodenal tissues (Fig 2i-v), with antibody staining

visible in the villi, lamina propria, mucosa including duodenal glands, and in the muscle layer. The

gene expression of AvBD6/7, encoding peptides with cationic charge ≥+5, was detected in all the

tissues of the newly hatched birds (Fig 1E & F), with the caecal and caecal tonsil expression values

dominating the profiles of the older birds. Ranges of Ct values for each tissue and assay can be found

in Supplementary Table 1 and Supplementary Fig 1.

9

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

In vitro Microbial Challenge with Gut Bacteria

AvBD8-10 gene expression was detected in all broiler chicken gut tissues analysed, and apart from

the duodenal samples, the profiles remained comparable from hatch to day 21. The gastrointestinal

tracts of birds raised in a commercial rearing environment are colonised by commensals, including

Lactobacillus sp, in the upper gastrointestinal tract (Stanley et al., 2012), and a mix of bacteria

including Bacteroides sp in the caecum (Cadwell et al., 2017). Expression such as this may reflect a

mechanism by which the host is able to control the commensal microbial populations (Bevins et al.,

2011). To explore whether Lactobacilli could specifically regulate AvBD8, 9 and 10 gene expression,

in vitro experiments utilising the virus free CHCC-OU2 immortalised chicken cell line (Ogura et al.,

1987) and Lactobacillus johnsonii were performed. Challenging the cells with L. johnsonii did not

significantly affect either AvBD8, 9 or 10 gene expression (Fig 3A), suggesting that the encoded

peptides do not function in regulating gut commensal populations although it cannot be excluded

that the null response of the cell line to Lactobacilli was cell line specific. The CHCC-OU2 cells did not

express the AvBD6 and 7 genes. However the induction (P<0.05) of AvBD2 gene expression,

encoding a peptide with charge >+4 and down-regulation of AvBD1 and AvBD3 gene expression

(P<0.05) encoding peptides of +8 and +6 respectively, in response to a higher Lactobacilli dose (104

CFU), suggested potential roles for other encoded AvBDs in controlling the commensal populations.

These data also provided support for the use of CHCC-OU2 cells as an appropriate model in the in

vitro bacterial challenge experiments.

No significant changes in AvBD8-10 expression were detected in the bird caeacal tissues, and in vitro

challenges using Bacteriodes doreii (104 bacteria/105 cells) did not affect AvBD10 gene expression

(Fig 3B). However, significant induction of genes encoding the inflammatory proteins IL-1β and IL-6

was detected (P<0.001) (Fig 3C & D). Interestingly, challenging the cells with the gut pathogen

Salmonella enterica serovar Typhimurium 1344, resulted in IL-1β, IL-6 (P<0.001) and AvBD10

(P<0.05) gene upregulation (Fig 3B-D).

10

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

Recombinant AvBD6, 9 and 10 Anti-microbial Activities

AvBD10 gene upregulation in response to Salmonella suggested a killing role for the encoded

AvBD10 peptides. To investigate this further, the antimicrobial activities of recombinant (r)AvBD10

(net charge of +2) as well as AvBD9 (+4) and AvBD6 (+6) peptides (Hellgren et al., 2010) (Fig S2) were

compared against bacterial strains isolated from the gastrointestinal tracts of commercially raised

birds. Antimicrobial data from time-kill assays (Table 2) revealed that at lowest concentration

comparable (0.5µM), AvBD6 peptides exhibited antimicrobial activity against Escherichia coli (44%

killing), compared to 29% killing for AvBD9 and 0% for AvBD10. At 12.5µM AvBD6 remained the most

potent antimicrobial agent, exhibiting 100% E.coli killing compared to 78% for AvBD9 and 63% for

AvBD10. These data were further supported by membrane permeabilisation studies in which sAvBD6

(0.5µM) was associated with 47.2±6.4% cell leakage that increased to a maximum of 60.3±6.3% at 2

minutes (Fig 4A); leakage associated with sAvBD9 (0.5 and 2µM) remained below 25% (Fig 4B). These

data supported an antimicrobial hierarchy of AvBD6>AvBD9>AvBD10. However, when the Gram-

positive strain, Enterococcus faecalis, was used, the hierarchy changed to AvBD10>AvBD6>AvBD9.

However, radial diffusion analyses performed under anaerobic conditions suggested AvBD10 exhibits

a bacteriostatic rather than a killing effect (Fig 4C).

The AvBD6 and 9 peptides, but not AvBD10, contain a C terminal tryptophan (W) amino acid which

has been linked to enhanced antimicrobial activity (Bi et al., 2013). To explore this property further,

an AvBD9 peptide was synthesized in which the terminal tryptophan (W) was replaced with a glycine

(G) and the antimicrobial properties of the engineered peptide re-examined using E.coli and E.

faecalis isolates. The resulting data (Table 2) indicated that the loss of the tryptophan (W) further

impaired the killing potency of the AvBD9 peptide and this was linked, potentially, to its membrane

permeabilisation and hence pore forming abilities (Fig 4B). Interestingly, engineering the AvBD10

peptide to contain a C terminal tryptophan (W) did not enhance its antimicrobial properties against

E.coli.

11

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

Structure Modelling

To help explain the differing antimicrobial properties of the peptides, in silico modelling of the three

peptide primary sequences (Fig 5A-C) was performed. Predicted structural models all displayed

significant β-sheet content linked to three di-sulphide bonds and an N-terminal α-helix. The AvBD6

structure supported the positively charged arginine (R) amino acids R10, R38 and R40 forming a

cluster with the C-terminal tryptophan 41 (W41), and lysine K33 along with arginines R24 and R19

forming a contiguous bunch with tryptophan 20 (W20). The aromatic tyrosine residues, Y22 and Y23,

connected the two clusters resulting in a forearm hook-like structure, all of which were consistent

with its in vitro antimicrobial properties. A similar structure was observed for AvBD9. This included a

surface cluster of positively charged R7, R19, R29, histidine H10 and K32 residues. However, while

the structure predicted the hydrophobic aromatic residues tryptophan W38 and phenylalanine F15

to be exposed, no hook-like forearm structure was apparent, which probably explained its reduced

antimicrobial potency in vitro. In contrast to AvBD6 and 9, the predicted AvBD10 model suggested a

less compact structure with no distinctive clusters of positive charge and no hook-like structures, all

of which predicted weak antimicrobial activity.

Wound Healing

The distinctive kidney and liver tissue expression of AvBD8-10, lack of gene induction in vitro and the

variability of the commensal bacteria antimicrobial data hinted at roles for these peptides in addition

to microbial killing. To explore this further, rAvBD10, selected due to its non-compact structure and

reduced cationic charge, was investigated for cell proliferation and wound healing properties using

CHCC-OU2 cells. No significant effects were observed using concentrations of 0.1-1nM (Fig 6A & B).

However, there was a suggestion of a positive effect on wound closure at lower peptide

concentrations, but this lacked statistical significance due to the variability within the dataset.

12

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

Discussion

Gallus gallus and other avian species express an array of AvBD genes, with the encoded peptides

functioning as part of the innate defences and helping to protect the tissues against infection

(Cuperus et al., 2013). In vivo expression of the AvBD genes has been widely reported but the data

are often conflicting, with variability attributed to the different bird ages, breeds and/or the rearing

conditions studied. However, within the literature are consistent patterns of AvBD gene expression

in tissues that in healthy birds are not routinely exposed to whole microbes. One such pattern is the

expression of AvBD8, 9 and 10 mRNA in the liver and AvBD9 and 10 in the kidney tissues (Fig 1;

(Butler et al., 2016; Cuperus et al., 2016; Lee et al., 2016; Ma et al., 2012; Wang et al., 2010)). It has

been suggested that expression in such tissues links to the encoded peptides functioning to fight

systemic infections (van Dijk et al., 2008), but these encoded peptides are characterised by charges

of +4. This is suggestive of weak antimicrobial properties, and hence inconsistent with their having

key roles in fighting infection. In support, synthetic linear AvBD8 and AvBD10 peptides have been

reported to demonstrate weak lytic activity against E.coli (Lee et al., 2016). Additionally modification

of AvBD8 to carry an increased charge has been shown to enhance its killing of gram negative

bacteria (Higgs et al., 2007), which strongly suggests that AvBD8 has not naturally evolved as a

potent antimicrobial agent. However, it cannot be discounted that these peptides carrying reduced

cationic charge work synergistically with each other and other AvBD peptides, to potentiate defensin

killing potency (Milona et al., 2007).

Although reduced compared to the liver and kidney tissues, AvBD 8, 9 and 10 gene expression and

AvBD9 synthesis were detected in the gut, and particularly in the duodenal tissues. As the gut

epithelium is continuously challenged by microbes, such expression may reflect a mechanism by

which the host is able to control the commensal microbial populations (Bevins et al., 2011). Poultry

commensals include Lactobacilli in the upper GI tract and Bacteroides in the caecum. The lack of

13

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

AvBD8, 9 or 10 gene induction in CHCC-OU2 cells in response to increasing numbers of these

bacteria did not, however, support key roles for the encoded peptides in regulating these gut

commensal populations. The use of CHCC-OU2 cells to model the gut epithelium was supported in

that AvBD10 induction was detected in response to Salmonella, an observation also reported in vivo

following the challenge of poultry with the gut pathogen (Hong et al., 2012). While these data

indicated a role for AvBD10 peptides in the protection of the gut epithelium from potential gut

pathogens, the actual mechanism remains puzzling as AvBD10 carries a relatively low positive charge

(+2), and in vitro the peptide displayed comparatively poor bacterial killing properties against Gram

negative bacteria.

Salt concentrations have been reported to affect AvBD anti-microbial activity in vitro, with

concentrations >50mM shown to reduce AvBD4, 6 and 10 activity against Gram negative and

positive bacteria (Yacoub et al., 2015; Yang et al., 2016). However, in vitro salt concentrations linked

to potent antimicrobial activity do not reflect the physiological salt concentrations in vivo at an

epithelial surface. Hence all in vitro assays were performed in PBS to reflect the in vivo environment.

Moreover, the fact that microbial killing was observed in vitro using AvBD 6 and 9 peptide

concentrations of 12.5M compares favourably to mucosal AMP concentrations reported in vivo (Shi

et al., 1999). These data suggest that the avian defensin peptides, at such concentrations, contribute

significantly to maintaining a protective barrier that defends the epithelium from microbial assault.

Mammalian defensins can function in wound repair and cell growth following microbial damage

(Cuperus et al., 2013). While the in vitro wound healing data using CHCC-OU2 cells did not wholly

support such a role for AvBD10, this may have reflected the peptide concentrations utilised. The

concentrations used modelled normal mammalian mucosa defensin gut concentrations (Nishi et al.,

2005), but were up to 100 fold less than those reported in relation to the wound healing properties

of the human defensin BD2 (Cuperus et al., 2013). Interestingly, the mammalian defensin HD6, like

AvBD10, lacks broad spectrum antimicrobial activity but defends the mammalian gut by forming

14

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

oligomers or nanonets that trap pathogens in the gut lumen, preventing them from entering host

cells (Chairatana et al., 2017; Chu et al., 2012). These nanonets are then either excreted or

inactivated by other components of the innate immune system such as recruited neutrophils and

macrophages. Whether AvBD10 oligomerises and functions as a net to capture microbes such as

Salmonella is not known, but its structure and bacteriostatic, rather than killing, mode of action is

supportive of such a mechanism. Interestingly, AvBD6 is shown in vitro to be a potent killing agent

(Table 2) and is also induced in the avian gut tissues in response to Salmonella (Akbari et al., 2008;

Ramasamy et al., 2012); it has also been shown to be mildly chemotactic for chicken macrophages

(Yang et al., 2016). This suggests that regulated expression of the AvBD6 and 10 genes within the

AvBD cluster, in response to potential gut pathogens, results in peptides that could potentially

function by independent yet co-ordinated mechanisms to clear such microbes from the gut.

Despite marked expression in kidney and liver tissues, AvBD9 gene expression was also evident in

the duodenal tissues with the peptide easily detectable in the day 7 tissues. Although predicted a

charge of +4, the rAvBD9 peptide preparations exhibited relatively weak killing properties against

chicken gut microbes, which were further reduced by the loss of the C-terminal tryptophan. The

poor membrane permeabilisation properties of AvBD9 also supported a reduced killing role for the

peptide, although other methods of bacterial killing in addition to pore formation are known to

function in vivo (Brogden, 2005). For example, AvBD9 treatment has been shown to be linked to

irregular septum formation in Clostridial division (van Dijk et al., 2007), although this implies some

degree of bacterial specificity. Gut AvBD9 gene expression has also been reported to localise with

enteroendocrine (EEC) cells, leading to the suggestion that AvBD9 released by chicken EECs may

function in host defence, not through direct killing mechanisms specifically, but by inducing T cell

cytokine production and dendritic cell differentiation (Cuperus et al., 2016).

Other non-killing roles for the defensins have also been proposed. Development of the early chicken

embryo is associated with AvBD9 and 10 gene expression (Meade et al., 2009), and while an

15

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

immunological function cannot be dismissed for the encoded peptides, the observed differential and

coordinated expression of the genes suggests potential roles in cell migration and tissue growth.

Such roles are strengthened by observations in mammals; for example, identification of defensin

expression in the notochord of the developing mouse embryo, EMAGE: 23850; 23812; 21820; 21821,

suggests potential roles for the encoded peptides in cell signalling and coordinating development

(Richardson et al., 2010). This is further supported by gradients of defensin expression identified in

the developing zebrafish (Oehlers et al., 2011). Although in vitro the rAVBD9 peptide did not affect

CHCC-OU2 cell proliferation, additional studies using higher concentrations of peptide and different

cell types are required to explore this further. Mammalian defensins have also been shown to play

key roles in mammalian reproduction particularly in relation to sperm function (Dorin et al., 2014).

Interestingly AvBD8 and 10 gene expression have been identified in the ovaries of laying hens

(Yoshimura, 2015) and AvBD10 in testes (Anastasiadou et al., 2014) so while the encoded peptides

may function to provide resistance to infection, they may also be involved in the physiological

processes themselves.

Expression, antimicrobial and modelling data all support AvBD6 being a key host antimicrobial agent

synthesised by and functioning in protecting the epithelial tissues, including the gut, from microbial

assault. In contrast, complementary data relating to the AvBD8-10 gene group suggests the encoded

peptides, while displaying antimicrobial properties to help fight potential infections, may also

possess other cellular functions and further biological analyses are needed to help define such

functions.

Conclusion

These data showed that in healthy broiler chicken tissues AvBD6/7 and AvBD8-10 gene expression

profiles were independent of the in vitro antimicrobial hierarchies of the encoded AvBD6, 9 and 10

peptides. Data relating to the AvBD8-10 gene group suggests the encoded peptides, while displaying

limited antimicrobial properties, may also function in other physiological processes and further

16

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

biological analyses are needed to help define such functions. However, it cannot be discounted that

the AvBD8-10 peptides carrying reduced cationic charge may work synergistically with each other,

and with other AvBD peptides, to potentiate defensin killing potency. Further biological analyses are

required to fully extrapolate the killing roles of these peptides physiologically.

Acknowledgements.

We acknowledge the support of BBSRC through grants BB/H018603/1, BB/1532845/1,

BBS/S/M/13127 and a Biosciences KTN PhD Top-up Award to KC. SSN was supported by the

Kurdistan Regional Government.

References

Akbari, M.R., Haghighi, H.R., Chambers, J.R., Brisbin, J., Read, L.R., Sharif, S., 2008. Expression of antimicrobial peptides in cecal tonsils of chickens treated with probiotics and infected with Salmonella enterica serovar typhimurium. Clin Vaccine Immunol 15, 1689-1693.

Anastasiadou, M., Theodoridis, A., Michailidis, G., 2014. Effects of sexual maturation and Salmonella infection on the expression of avian beta-defensin genes in the chicken testis. Vet Res Commun 38, 107-113.

Bevins, C.L., Salzman, N.H., 2011. The potter's wheel: the host's role in sculpting its microbiota. Cell Mol Life Sci 68, 3675-3685.

Bi, X., Wang, C., Ma, L., Sun, Y., Shang, D., 2013. Investigation of the role of tryptophan residues in cationic antimicrobial peptides to determine the mechanism of antimicrobial action. J Appl Microbiol 115, 663-672.

Brogden, K.A., 2005. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol 3, 238-250.

Butler, V.L., Mowbray, C.A., Cadwell, K., Niranji, S.S., Bailey, R., Watson, K.A., Ralph, J., Hall, J., 2016. Effects of rearing environment on the gut antimicrobial responses of two broiler chicken lines. Vet Immunol Immunopathol 178, 29-36.

Cadwell, K., Niranji, S.S., Armstrong, V.L., Mowbray, C.A., Bailey, R., Watson, K.A., Hall, J., 2017. AvBD1 nucleotide polymorphisms, peptide antimicrobial activities and microbial colonisation of the broiler chicken gut. BMC Genomics 18, 637.

17

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409410411412413414415416417418419420421422423424425426427

Chairatana, P., Nolan, E.M., 2017. Human alpha-Defensin 6: A Small Peptide That Self-Assembles and Protects the Host by Entangling Microbes. Acc Chem Res 50, 960-967.

Cheng, Y., Prickett, M.D., Gutowska, W., Kuo, R., Belov, K., Burt, D.W., 2015. Evolution of the avian beta-defensin and cathelicidin genes. BMC Evol Biol 15, 188.

Chu, H., Pazgier, M., Jung, G., Nuccio, S.P., Castillo, P.A., de Jong, M.F., Winter, M.G., Winter, S.E., Wehkamp, J., Shen, B., Salzman, N.H., Underwood, M.A., Tsolis, R.M., Young, G.M., Lu, W., Lehrer, R.I., Baumler, A.J., Bevins, C.L., 2012. Human alpha-defensin 6 promotes mucosal innate immunity through self-assembled peptide nanonets. Science 337, 477-481.

Cuperus, T., Coorens, M., van Dijk, A., Haagsman, H.P., 2013. Avian host defense peptides. Dev Comp Immunol 41, 352-369.

Cuperus, T., van Dijk, A., Dwars, R.M., Haagsman, H.P., 2016. Localization and developmental expression of two chicken host defense peptides: cathelicidin-2 and avian beta-defensin 9. Dev Comp Immunol 61, 48-59.

Derache, C., Meudal, H., Aucagne, V., Mark, K.J., Cadene, M., Delmas, A.F., Lalmanach, A.C., Landon, C., 2012. Initial insights into structure-activity relationships of avian defensins. J Biol Chem 287, 7746-7755.

Dorin, J.R., Barratt, C.L., 2014. Importance of beta-defensins in sperm function. Mol Hum Reprod 20, 821-826.

Gong, D., Wilson, P.W., Bain, M.M., McDade, K., Kalina, J., Herve-Grepinet, V., Nys, Y., Dunn, I.C., 2010. Gallin; an antimicrobial peptide member of a new avian defensin family, the ovodefensins, has been subject to recent gene duplication. BMC Immunol 11, 12.

Hellgren, O., Ekblom, R., 2010. Evolution of a cluster of innate immune genes (beta-defensins) along the ancestral lines of chicken and zebra finch. Immunome Res 6, 3.

Herve, V., Meudal, H., Labas, V., Rehault-Godbert, S., Gautron, J., Berges, M., Guyot, N., Delmas, A.F., Nys, Y., Landon, C., 2014. Three-dimensional NMR structure of Hen Egg Gallin (Chicken Ovodefensin) reveals a new variation of the beta-defensin fold. J Biol Chem 289, 7211-7220.

Higgs, R., Lynn, D.J., Cahalane, S., Alana, I., Hewage, C.M., James, T., Lloyd, A.T., O'Farrelly, C., 2007. Modification of chicken avian beta-defensin-8 at positively selected amino acid sites enhances specific antimicrobial activity. Immunogenetics 59, 573-580.

Hollox, E.J., Armour, J.A., Barber, J.C., 2003. Extensive normal copy number variation of a beta-defensin antimicrobial-gene cluster. Am J Hum Genet 73, 591-600.

Hong, Y.H., Song, W., Lee, S.H., Lillehoj, H.S., 2012. Differential gene expression profiles of beta-defensins in the crop, intestine, and spleen using a necrotic enteritis model in 2 commercial broiler chicken lines. Poult Sci 91, 1081-1088.

Jenssen, H., Hamill, P., Hancock, R.E., 2006. Peptide antimicrobial agents. Clin Microbiol Rev 19, 491-511.

Kluver, E., Schulz-Maronde, S., Scheid, S., Meyer, B., Forssmann, W.G., Adermann, K., 2005. Structure-activity relation of human beta-defensin 3: influence of disulfide bonds and cysteine substitution on antimicrobial activity and cytotoxicity. Biochemistry 44, 9804-9816.

Landon, C., Thouzeau, C., Labbe, H., Bulet, P., Vovelle, F., 2004. Solution structure of spheniscin, a beta-defensin from the penguin stomach. J Biol Chem 279, 30433-30439.

Lee, M.O., Jang, H.J., Rengaraj, D., Yang, S.Y., Han, J.Y., Lamont, S.J., Womack, J.E., 2016. Tissue expression and antibacterial activity of host defense peptides in chicken. BMC Vet Res 12, 231.

Lehrer, R.I., 2004. Primate defensins. Nat Rev Microbiol 2, 727-738.Ma, D., Lin, L., Zhang, K., Han, Z., Shao, Y., Wang, R., Liu, S., 2012. Discovery and characterization of

Coturnix chinensis avian beta-defensin 10, with broad antibacterial activity. J Pept Sci 18, 224-232.

Meade, K.G., Cormican, P., Narciandi, F., Lloyd, A., O'Farrelly, C., 2014. Bovine beta-defensin gene family: opportunities to improve animal health? Physiol Genomics 46, 17-28.

18

428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477

Meade, K.G., Higgs, R., Lloyd, A.T., Giles, S., O'Farrelly, C., 2009. Differential antimicrobial peptide gene expression patterns during early chicken embryological development. Dev Comp Immunol 33, 516-524.

Milona, P., Townes, C.L., Bevan, R.M., Hall, J., 2007. The chicken host peptides, gallinacins 4, 7, and 9 have antimicrobial activity against Salmonella serovars. Biochem Biophys Res Commun 356, 169-174.

Nishi, Y., Isomoto, H., Mukae, H., Ishimoto, H., Wen, C.Y., Wada, A., Ohnita, K., Mizuta, Y., Murata, I., Hirayama, T., Nakazato, M., Kohno, S., 2005. Concentrations of alpha- and beta-defensins in gastric juice of patients with various gastroduodenal diseases. World J Gastroenterol 11, 99-103.

Oehlers, S.H., Flores, M.V., Chen, T., Hall, C.J., Crosier, K.E., Crosier, P.S., 2011. Topographical distribution of antimicrobial genes in the zebrafish intestine. Dev Comp Immunol 35, 385-391.

Ogura, H., Fujiwara, T., 1987. Establishment and characterization of a virus-free chick cell line. Acta medica Okayama 41, 141-143.

Otte, J.M., Werner, I., Brand, S., Chromik, A.M., Schmitz, F., Kleine, M., Schmidt, W.E., 2008. Human beta defensin 2 promotes intestinal wound healing in vitro. J Cell Biochem 104, 2286-2297.

Peng, J., Xu, J., 2011. RaptorX: exploiting structure information for protein alignment by statistical inference. Proteins 79 Suppl 10, 161-171.

Ramasamy, K.T., Verma, P., Reddy, M.R., 2012. Differential gene expression of antimicrobial peptides beta defensins in the gastrointestinal tract of Salmonella serovar Pullorum infected broiler chickens. Vet Res Commun 36, 57-62.

Richardson, L., Venkataraman, S., Stevenson, P., Yang, Y., Burton, N., Rao, J., Fisher, M., Baldock, R.A., Davidson, D.R., Christiansen, J.H., 2010. EMAGE mouse embryo spatial gene expression database: 2010 update. Nucleic Acids Res 38, D703-709.

Rodriguez-Jimenez, F.J., Krause, A., Schulz, S., Forssmann, W.G., Conejo-Garcia, J.R., Schreeb, R., Motzkus, D., 2003. Distribution of new human beta-defensin genes clustered on chromosome 20 in functionally different segments of epididymis. Genomics 81, 175-183.

Shi, J., Zhang, G., Wu, H., Ross, C., Blecha, F., Ganz, T., 1999. Porcine epithelial beta-defensin 1 is expressed in the dorsal tongue at antimicrobial concentrations. Infection and immunity 67, 3121-3127.

Stanley, D., Denman, S.E., Hughes, R.J., Geier, M.S., Crowley, T.M., Chen, H., Haring, V.R., Moore, R.J., 2012. Intestinal microbiota associated with differential feed conversion efficiency in chickens. Appl Microbiol Biotechnol 96, 1361-1369.

Tu, J., Li, D., Li, Q., Zhang, L., Zhu, Q., Gaur, U., Fan, X., Xu, H., Yao, Y., Zhao, X., Yang, M., 2015. Molecular Evolutionary Analysis of beta-Defensin Peptides in Vertebrates. Evol Bioinform Online 11, 105-114.

van Dijk, A., Veldhuizen, E.J., Haagsman, H.P., 2008. Avian defensins. Vet Immunol Immunopathol 124, 1-18.

van Dijk, A., Veldhuizen, E.J., Kalkhove, S.I., Tjeerdsma-van Bokhoven, J.L., Romijn, R.A., Haagsman, H.P., 2007. The beta-defensin gallinacin-6 is expressed in the chicken digestive tract and has antimicrobial activity against food-borne pathogens. Antimicrob Agents Chemother 51, 912-922.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A., Speleman, F., 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3, RESEARCH0034.

Wang, R., Ma, D., Lin, L., Zhou, C., Han, Z., Shao, Y., Liao, W., Liu, S., 2010. Identification and characterization of an avian beta-defensin orthologue, avian beta-defensin 9, from quails. Appl Microbiol Biotechnol 87, 1395-1405.

Xiao, Y., Hughes, A.L., Ando, J., Matsuda, Y., Cheng, J.F., Skinner-Noble, D., Zhang, G., 2004. A genome-wide screen identifies a single beta-defensin gene cluster in the chicken: implications for the origin and evolution of mammalian defensins. BMC Genomics 5, 56.

19

478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529

Yacoub, H.A., Elazzazy, A.M., Abuzinadah, O.A., Al-Hejin, A.M., Mahmoud, M.M., Harakeh, S.M., 2015. Antimicrobial activities of chicken beta-defensin (4 and 10) peptides against pathogenic bacteria and fungi. Frontiers in cellular and infection microbiology 5, 36.

Yang, M., Zhang, C., Zhang, X., Zhang, M.Z., Rottinghaus, G.E., Zhang, S., 2016. Structure-function analysis of Avian beta-defensin-6 and beta-defensin-12: role of charge and disulfide bridges. BMC Microbiol 16, 210.

Yoshimura, Y., 2015. Avian beta-defensins expression for the innate immune system in hen reproductive organs. Poult Sci 94, 804-809.

Zhou, Y.S., Webb, S., Lettice, L., Tardif, S., Kilanowski, F., Tyrrell, C., Macpherson, H., Semple, F., Tennant, P., Baker, T., Hart, A., Devenney, P., Perry, P., Davey, T., Barran, P., Barratt, C.L., Dorin, J.R., 2013. Partial deletion of chromosome 8 beta-defensin cluster confers sperm dysfunction and infertility in male mice. PLoS Genet 9, e1003826.

Conflict of Interest: RB and KAW are employees of Aviagen Ltd.

Figure Legends

Fig 1: In vivo tissue expression profiles of the AvBD6-10 genes. Expression profiles (Arbitary

Units (AU)) of the AvBD8 (A), 9 (B), 10 (C), 6 (D) and 7(E) genes in chicken kidney (K), liver (L),

duodenum (D), jejunum (J), ileum (I), caecum (C) and caecal tonsil (CT) tissues removed from

birds at hatch (D0), day 7 (D7) and day 21 (21). Data relates to 5 birds/group and is presented as

mean ± SEM.

Fig 2: AvBD9 IHC staining of duodenal tissues from day 7 birds. IHC analyses to show epithelial

localisation of AvBD9 in duodenal tissues from Day 7 birds using AvBD9 polyclonal antibody

diluted 1:70 and peroxidase staining (x40 (i, ii, iii) and x400 (iv, v) magnification). A no primary

antibody control (i) was conducted alongside all staining.

Fig 3: AvBD expression profiles of bacterially challenged CHCC-OU2 cells. AvBD gene

expression profiles of CHCC-0U2 cells (105) challenged for 24h with heat killed 102-104

Lactobacillus johnsonii (A). AvBD10 (B), IL-1β (C) and IL-6 (D) gene expression profiles in

response to Bacteriodes doreii (102-4 bacteria/105 cells), Salmonella enterica serovar

Typhimurium 1344 and L. johnsonii. (N=3 experiments; n=3 technical replicates; data presented

as mean ± SEM) * P < 0.05; ** P < 0.01: *** P < 0.001.

20

530531532533534535536537538539540541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

Fig 4: sAvBD9 membrane permeabilisation and rAvBD10 antimicrobial activities. Calcein

leakage from liposomes incubated with AvBD6 0.5uM (A) and AvBD9 variant peptides 0.5-2uM

(B). Melittin (0.5M) was used as the permeabilization control (Cadwell et al., 2017).

Experiments (AvBD6 N=4; n=4; mean ± SEM; AvBD9 N=1; n=3) performed at room temperature

in 50mM sodium phosphate buffer. Radial diffusion data in an anaerobic environment showing

inhibitory effects of AvBD10 against Lactobacillus johnsonii (C) and Bacteroides dorei (D).

Fig 5: Predicted 3D structures of AvBD peptides. AvBD6 (A), AvBD9 (B) and AvBD10 (C).

Fig 6: Effects of AvBD10 on CHCC-OU2 wound healing and cell proliferation. Wells containing

confluent CHCC-OU2 cell monolayers were scratched and incubated with either PBS, Bovine

Serum Albumin (10nM), mitomycin C (10nM) or AvBD10 (10nM) for up to 72h. Images were

taken at each time point, 0, 24, 48 and 72h, and % healing calculated using ImageJ software.

N=3 for each time-point; mean ± SEM. Cell proliferation is shown relative to PBS control. ** P

<0.01

Fig S1: In vivo tissue expression profiles of the AvBD6-10 genes. Data as in Figure 1 but

presented with comparative y axes to visualise gene expression with bird age and within genes.

Fig S2: NU-PAGE gels of purified recombinant AVBD6, 9 and 10 peptides. (i) rAVBD9:

M:molecular weight marker, 1 -3: elution buffer collections; (ii) rAVBD6: M:molecular weight

marker, 1-GST cleavage, 2-6 elution buffer collections with fraction 6 used in antimicrobial

activity studies; (iii) AvBD10: M:molecular weight marker, 1-GST cleavage, 2-4 elution buffer

collections with fraction 4 used in antimicrobial activity, cell proliferation and wound healing

studies.

21

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

Table 1: End-point PCR and qPCR primers AvBD1-10 primers and optimal annealing conditions

utilised for quantitative analyses and expression cloning studies

Table 2: Time-kill antimicrobial assay Time-kill assay data showing bacterial survival following 2h

incubation of E. coli and E. faecalis with rAvBD6, 9 and 10 peptides (Data presented as mean±SEM;

N=3 experiments and minimum of 6 replicates). ND – no data available

Supplementary Table 1: Ct values for AvBD7-10 qPCR expression Raw minimum and maximum Ct

values for each of the AvBD6-10 qPCR assays at each day tested (0, 7, 21) and for each tissue (kidney

(K), liver (L), duodenum (D), jejunum (J), ileum (I), caecum (C), caecal tonsil (CT)).

22

583

584

585

586

587

588

589

590

591

592