€¦ · web viewjudith hall, institute for cell and molecular biosciences, newcastle university,...
TRANSCRIPT
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 :[email protected]
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