Pathogens and Disease Advance Access published November 24, 2015
1
A novel experimental platform for toxigenic and nontoxigenic
2
Corynebacterium ulcerans infection in mice
3 4
Yu Mochizuki1,3‡, Honami Saeki1,3¶, Masaaki Iwaki1*, Hitrotaka Takagi2,
5
Keigo Shibayama1, Hiromi Amao3 and Akihiko Yamamoto1§
6
1
7
Research, National Institute of Infectious Diseases, Tokyo, Japan, 3 School
8
of Animal Science, Nippon Veterinary and Life Science University, Tokyo,
9
Japan * Corresponding Author: Tel: +81-42-561-0771 Fax: +81-42-561-7173 e-mail: [email protected] Running title: Experimental Corynebacterium ulcerans infection in mice ‡ Present address: Vaccine Research Department, Denka Seiken Co. Ltd. Niigata, Japan
20
¶ Present address: KAC Inc. Kyoto, Japan
21 22 23 24 25 26 27 28 29 30
§ Present address: Division of Biosafety Control and Research, National Institute of Infectious Diseases, Tokyo, Japan Keywords: zoonosis, diphtheria, respiratory infection, animal model, Corynebacterium, intranasal infection One-sentence summary: An intranasal experimental model for Coryebacterium ulcerans was constructed in mice to provide a platform for the analysis of its virulence.
1
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
10 11 12 13 14 15 16 17 18 19
Department of Bacteriology II, 2 Division of Biosafety Control and
Abstract
32
Corynebacterium ulcerans is a zoonotic pathogen that can produce
33
diphtheria toxin and causes an illness categorized as diphtheria in the
34
European Union, because its clinical appearance is similar to that of
35
diphtheria caused by Corynebacterium diphtheriae. Despite the importance
36
of the pathogen for public health, the organism’s mechanism of infection
37
has not been extensively studied, especially in experimental animal models.
38
In the present study, we thus constructed an intranasal infection system for
39
mice, which are insensitive to diphtheria toxin and advantageous in
40
excluding the cytotoxic effect of the toxin that might interfere the analysis
41
of the early stage of infection. Both the toxigenic and nontoxigenic C.
42
ulcerans strains were capable of killing mice within three days after
43
inoculation at 107 colony-forming units (CFU) per mouse. In
44
experimentally infected animals, C. ulcerans was detected in the
45
respiratory tract but not in the intestinal tract. The bacterium was also
46
detected in peripheral blood and it disseminated into the lung, kidney, and
47
spleen to produce a systemic infection. This experimental infection system 2
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
31
provides a platform for analyzing C. ulcerans virulence in future studies. 48
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
3
Introduction
50
Corynebacterium ulcerans is a zoonotic pathogen. The organism can be
51
lysogenized with bacteriophages harboring the tox gene (Sekizuka, et al.,
52
2012) (Sangal & Hoskisson, 2013, Sangal, et al., 2014) and produces
53
diphtheria toxin. Human infection by toxigenic C. ulcerans causes an
54
illness clinically indistinguishable from diphtheria caused by
55
Corynebacterium diphtheriae. Therefore, C. ulcerans infection is
56
categorized as diphtheria in regions including European countries (Bonnet
57
& Begg, 1999).
58
Unlike the mechanism of infection by C. diphtheriae, the infection
59
mechanisms of C. ulcerans are not well understood. In C. diphtheriae
60
infection, many factors linked to the early stage of the infection have been
61
identified through in vitro studies. These include adhesion factors,
62
including pili (Mandlik, et al., 2007), and invasion factors (Ott, et al., 2010,
63
Sabbadini, et al., 2012). In contrast, in C. ulcerans, phospholipase D has
64
been identified as a virulence factor (McNamara, et al., 1995) and after the
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
49
4
genome sequence became available, several other candidate virulence
66
factors have been identified (Trost, et al., 2011, Sekizuka, et al., 2012).
67
However, the significance of these factors for pathogenicity remains to be
68
elucidated. Only a few reports have been published on animal models for C.
69
ulcerans infection: an intravenous mouse infection model (Dias, et al.,
70
2011) and a nematode and Galleria moth larvae infection model with
71
nontoxigenic strains (Ott, et al., 2012). As C. ulcerans is considered to be a
72
primarily respiratory pathogen, a system reflecting respiratory infection is
73
desirable in order to further elucidate its infection mechanisms.
74
One of the expected problems associated with analyzing the early stage of
75
infection by animal experiment is that diphtheria toxin might interfere the
76
analysis by its potent deteriorating effects on infection site. We thus chose
77
mice, as host animals, which are insensitive to diphtheria toxin due to the
78
lack of a functional receptor.
79
In this respect, a mouse intranasal infection model could provide a suitable
80
novel platform to analyze the pathogenicity of C. ulcerans in vivo. In fact,
5
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
65
intranasal infection models have previously been employed for in vivo
82
experimental infection models of various kinds of related pathogenic
83
bacterial species including Corynebacterium kutscheri (Brownstein, et al.,
84
1985), Rhodococcus hoagii (Mutimer & Woolcock, 1982,
85
González-Iglesias, et al., 2014), suggesting that this model could be
86
successfully applied to C. ulcerans as well.
87
In the present study, we constructed a mouse intranasal infection system in
88
order to establish a platform for analyzing the pathogenicity of C. ulcerans.
89
Experimental lethal infection with toxigenic and nontoxigenic C. ulcerans,
90
as well as analysis of distribution of the bacteria in infected animals and an
91
approach for measuring serum titers are described.
92 93 94
6
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
81
Materials and Methods
96
Animals
97
Female BALB/cCrSlc mice (6 weeks old) were used for all experiments in
98
this study. Mice were acclimatized for 4 days before being subjected to
99
experiments. Animals were sacrificed by exsanguination under sedation
100
with sodium pentobarbital (Somnopentyl, Kyoritsu Seiyaku Corporation,
101
Tokyo, Japan). All animal experiments were carried out with strict adhesion
102
to the Guidelines for Animal Welfare of the Japanese National Institute of
103
Infectious Diseases.
104 105
Culture media
106
Blood agar plates (Nissui Pharmaceutical Co. Ltd., Japan) were used for
107
the propagation of bacterial cells from frozen stocks. Brain Heart infusion
108
(BHI) broth (Becton-Dickinson, NJ, USA) was used for broth culture in
109
preparation of the inoculum and for agar plates for bacterial detection and
110
colony counting in experimentally infected animals. Furazolidone-
7
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
95
Nalidixic acid- Colimycin agar (FNC agar) plates were prepared according
112
to Amao et al. (Amao, et al., 1995) using PearlCore heart infusion agar
113
(Eiken Chemical Co. Ltd., Tokyo, Japan), furazolidone (Sigma-Aldrich CO.
114
LLC., MO, USA), nalidixic acid (Wako Pure Chemical Industries Ltd.
115
Osaka, Japan) and Colimycin® (Pola Pharma Inc., Tokyo, Japan).
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
111
116 117
Bacterial strains
118
Toxigenic Corynebacterium ulcerans 0102 was isolated from a case of
119
respiratory infection in Japan in 2001 (Hatanaka, et al., 2003).
120
Nontoxigenic C. ulcerans ATCC51799 and Corynebacterium glutamicum
121
ATCC13032 were purchased from the American Type Culture Collection
122
(VA, USA).
123 124
Experimental infection of mice
125
Inoculum preparation: Bacteria were propagated from frozen stocks on a
126
blood agar plate at 37°C for 48 h, and a single colony was picked and used
8
to seed 10 mL of BHI broth. The culture was shaken at 37°C for 72 h and
128
then vortexed to disperse bacterial cells. The optical density (OD) at 600
129
nm was measured to assure that growth did not exceed an OD of 2.0.
130
Bacterial cell pellets were washed twice and finally resuspended in a sterile
131
saline solution, and adjusted to an OD600 of 8.8 (= 1× suspension). By
132
plating the bacteria on BHI agar plates after appropriate dilution, the 1×
133
suspension was confirmed to give 2 × 109 colony-forming units (CFU)
134
mL-1 (= 108 CFU per 50 µL).
135
Intranasal challenge: Mice were anesthetized by intramuscular injection
136
of a mixture of ketamin hydrochloride (Ketalar, Daiichi Sankyo Co. Ltd.,
137
Tokyo, Japan) and xylazine hydrochloride (Bayer Yakuhin Ltd., Osaka,
138
Japan). Ketamin hydrochloride was handled by a license holder according
139
to the Japanese Narcotics and Psychotropics Control Law. To the nasal
140
cavity of anesthetized mice, 50 µL (25 µL for each side) of bacterial
141
suspension was inoculated, aided by the breathing of the mice. Sterile
142
saline was used as a negative control.
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
127
9
Observation and sampling of challenged mice: After the bacterial
144
challenge, mice were observed for 14 days. To recover the inoculated
145
bacterium, the deceased or sacrificed mice were subjected to bleeding and
146
swabs were collected from the nasal cavity, oral cavity, and trachea.
147
Additional samples were collected after autopsy: the lung, liver, kidney,
148
and spleen were aseptically removed, weighed, and homogenized in a
149
sterile saline solution (10× volume) with a "Biomasher" homogenizer
150
(Nippi Inc., Tokyo, Japan). Cecal content were collected aseptically by
151
opening the cecum at autopsy. Colorectal feces were collected by
152
evacuation by the mice into sterile test tubes.
153
Identification of viable C. ulcerans from collected samples: Viable
154
C. ulcerans was detected by swab streaking (nasal and tracheal swabs) or
155
by plating appropriately diluted organ homogenates (lung, liver, kidney,
156
spleen) or blood on BHI agar plates. For samples in which interference by
157
normal flora components is expected (oral and eye swabs, cecal content,
158
and colorectal feces), streaking and plating occurred on selective FNC agar
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
143
10
plates. For swabs, a rough population estimation was done by counting the
160
colonies appearing on the streaks. For cecal content, colorectal feces, blood,
161
and organ homogenates, all colonies appearing on the plates were counted.
162
C. ulcerans identification: Colonies appearing on the BHI or FNC agar
163
plates were confirmed to be C. ulcerans based on colony morphology or
164
Gram staining, and when needed by identification with API Coryne kit
165
(bioMérieux SA, Marcy l'Etoile, France) and PCR directed to the
166
diphtheria toxin gene (A fragment) and the phospholipase D (PLD) gene.
167
The following primers were used: Tox1 (5′-ATC CAC TTT TAG TGC
168
GAG AAC CTT CGT CA-3′) and Tox2 (5′-GAA AAC TTT TCT TCG
169
TAC CAC GGG ACT AA-3′) (Nakao & Popovic, 1997) for the diphtheria
170
toxin gene, and PLD-F (5′-ATA AGC GTA AGC AGG GAG CA-3′) and
171
PLD-R (5′-ATC AGC GGT GAT TGT CTT CC-3′) (Torres, et al., 2013) for
172
the phospholipase D gene.
173
Measurement of serum anti-diphtheria toxin and anti-C. ulcerans titers
174
Measurement of serum antitoxin titer was performed according to
11
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
159
previously described procedures (Miyamura, et al., 1974, Miyamura, et al.,
176
1974). Briefly, serially diluted sera (25 µL) were mixed with 25 µL of
177
diphtheria toxin (16 CD50 in 25 µL in Minimum Essential Medium (Eagle)
178
(Sigma-Aldrich Co. LLC.) containing 3% newborn calf serum (Life
179
Technologies Corporation, CA, USA)) and added to the wells of 96-well
180
plates containing 175 µL of medium with 1.5 × 105 Vero cells. These
181
cultures were incubated for 4 days at 37°C and the endpoint of
182
neutralization was determined by observation of the cell morphology. The
183
outcome was compared to the endpoint of standard diphtheria antitoxin
184
based on the defined International Unit (Japanese National Diphtheria
185
Antitoxin, equine, lot 10). Antitoxin titers of the tested sera were calculated
186
by parallel line assays.
187
The anti-C. ulcerans serum IgG titer was measured by ELISA. C. ulcerans
188
0102 cells were collected from overnight culture in BHI broth. After
189
washing with PBS, the bacteria were killed by heating at 60°C for 1 h and
190
then treated with 0.01% formalin at 4°C for 48 h with daily changing of the
12
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
175
formalin solution. Formalin-denatured antigen was then appropriately
192
diluted in PBS and coated on Polysorp™ plates (Thermo Scientific)
193
overnight at 4°C. Rabbit or mouse sera were serially diluted with PBS
194
containing 0.5% BSA. One hundred microliters of diluted sera were added
195
to wells as primary antibody and incubated for 1 h at 37°C. After washing
196
three times with PBS containing 0.05% Tween 20, appropriately diluted
197
HRP-anti rabbit immunoglobulins (Dako) or HRP-anti mouse IgG (EMD
198
Millipore) was added as secondary antibody and incubated for 1 h at 37°C.
199
After washing three times with PBS containing 0.05% Tween 20,
200
specifically bound secondary antibody was visualized with
201
3,3',5,5'-tetramethylbenzidine. The enzyme reaction was stopped with
202
sulfuric acid and the absorbance at 450 nm was measured. Anti-C. ulcerans
203
mouse serum antibody titer was expressed as a factor of endpoint dilution
204
and was significantly higher than that in negative control wells incubated
205
without primary antibody.
206
13
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
191
Results
208
Lethal effect of intranasal inoculation of C. ulcerans in mice
209
C. ulcerans toxigenic 0102 and nontoxigenic ATCC 51799 strains, as well
210
as the C. glutamicum ATCC 13032 strain were propagated in BHI broth,
211
appropriately diluted and intranasally inoculated into anesthetized mice.
212
Fig. 1 shows the survival curve of the inoculated mice. Fifty percent of the
213
mice inoculated with 106 CFU of the toxigenic C. ulcerans 0102 strain
214
(n = 20) died before the end of the observation period (14 days). At 10 7 and
215
108 CFU, all mice (n = 6 for both) died within 4 and 3 days, respectively. In
216
contrast, using the nontoxigenic C. ulcerans ATCC 51799, at 106 CFU all
217
(n = 9) mice survived until the end of the observation period. However, at
218
107 CFU, the nontoxigenic strain displayed a more pronounced lethal effect
219
than the 0102 strain, as all mice (n = 7) died within 2 days. The avirulent
220
strain C. glutamicum ATCC 13032, frequently used in the fermentation
221
industry for the production of amino acids, did not show any lethal effects
222
at 107 (n = 7). Sterile saline solution did not show any lethal effect either
14
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
207
(data not shown).
224
Distribution of C. ulcerans 0102 in infected mice
225
In order to investigate the distribution of the C. ulcerans 0102 strain after
226
experimental infection, mice were inoculated with 106 CFU of the
227
bacterium, sacrificed at 2-day intervals and the presence of viable C.
228
ulcerans 0102 was estimated by plating swabs from respiratory tracts and
229
contents of the intestinal tracts. Fig. 2 shows the recovery of C. ulcerans
230
0102 from these specimens. The population of the bacterium detected in
231
each specimen was assessed according to a grading system (- to +++). As
232
shown in Fig. 2, the bacterium was recovered from respiratory tracts, eyes,
233
and blood. In the respiratory tract, the estimated population was largest in
234
the trachea, and smaller in the nasal and oral cavities. The organism was
235
recovered from trachea throughout the observation period, whereas in nasal
236
and oral cavities the bacterium was undetectable within 8 and 6 days,
237
respectively. The organism was detected in circulating blood until day 4
238
after inoculation, but was undetectable afterwards. In addition, the
15
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
223
organism was detected until day 8 in eye swabs.
240
C. ulcerans also disseminated into various organs including the lung, liver,
241
kidney, and spleen. Fig. 3 shows the lesions formed in lung, kidney, and
242
spleen on days 5, 9 and 13, respectively, after inoculation. Clearly
243
distinguishable legions were formed in these organs. Fig. 4 illustrates the
244
presence of the organism in the lung, liver, kidney, and spleen, and the
245
increase/decrease in organ weights measured at 2-day intervals. As shown
246
in Fig. 4, the bacterium was detected in these organs until day 12 after
247
inoculation. Enlargement of the spleen was also observed at day 8 (Fig.
248
4D).
249
Serum anti-diphtheria toxin and anti-C. ulcerans antibody titer
250
Serum anti-diphtheria toxin neutralization titer was measured according to
251
Miyamura et al. (Miyamura, et al., 1974, Miyamura, et al., 1974). Mice
252
inoculated with 106 CFU of C. ulcerans 0102 were bled at day 8, 10, 12
253
(n = 5 for each), and day 14 (n = 4), and subjected to titer determination. In
254
all specimens, the anti-diphtheria toxin neutralization titer was below the
16
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
239
detection limit (data not shown). The serum anti-C. ulcerans titer was
256
measured by an ELISA system developed in this study. Fig. 5 illustrates the
257
profile of absorbance at 450 nm obtained with serially diluted rabbit and
258
mouse hyperimmune sera using this system. Within the linear region
259
corresponding to 2 × 103–1 × 105 times dilution for rabbit hyperimmune
260
serum and 4 × 102–1 × 104 times dilution for mouse hyperimmune serum,
261
anti-C. ulcerans titer determination was shown to be feasible. Using this
262
system, anti-C. ulcerans IgG titer in sera from inoculated mice was
263
measured. As was the case for the anti-diphtheria toxin neutralization titer,
264
all of the serum exhibited low absorbance at 450 nm, below the lower limit
265
of the linear range (data not shown).
266
17
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
255
Discussion
268
In this study, we intended to provide a platform for the analysis of
269
C. ulcerans virulence in vivo. To reflect the most common route of
270
infection for this organism, an intranasal experimental infection system was
271
constructed. We demonstrate here that the infection model is effective as it
272
enables lethal infection.
273
An in vivo experimental infection system is desirable for further analysis of
274
the infection mechanism of C. ulcerans (and also for C. diphtheriae). In
275
analyzing the mechanisms underlying the early stages of infection such as
276
adhesion and colonization, the toxic effects of diphtheria toxin may serve
277
as an interfering factor through deterioration of the cells interacting with
278
the organism. In this respect, mice are suitable candidates for the platform
279
as they are highly resistant to diphtheria toxin due to the lack of a
280
functional receptor (Mitamura, et al., 1995). Previous reports on
281
invertebrate C. ulcerans and C. diphtheriae infection models employ
282
nontoxigenic strains, also resulting in exclusion of diphtheria toxin effects
18
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
267
(Ott, et al., 2012). Additional advantages of using mice are that they are
284
easy to handle and that a large number of animals can be studied.
285
Using mice, we found that C. ulcerans was capable of killing the animals
286
when administered intranasally (Fig. 1). The toxigenic C. ulcerans 0102
287
killed 50% of all mice when inoculated at 106 CFU, but the total survival
288
was similar for nontoxigenic C. ulcerans ATCC 51799. At a ten times
289
larger number of inoculated bacterial cells, the potency of killing effects
290
reverted: 107 CFU of the 0102 strain killed all animals within 4 days and
291
ATCC 51799 strains killed all mice in 2 days. Avirulent C. glutamicum had
292
no effect (Fig. 1), indicating that the lethality was not merely a nonspecific
293
effect. Based on these results and the fact that the mice are resistant to the
294
toxin, it seemed unlikely that toxigenicity is a crucial factor in the lethality
295
of C. ulcerans in the experimental system presented here. In other words, C.
296
ulcerans can produce a lethal infection in mice irrespective of toxigenicity.
297
In infected animals, C. ulcerans 0102 was recovered from the surface of
298
the respiratory tract but not from the contents of the lower intestinal tract
19
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
283
(Fig. 2). Although the bacterium was initially detected in the oral cavity
300
(Fig. 2), it was not recovered from cecal content and colorectal feces
301
throughout the observation period. This is in contrast to Corynebacterium
302
kutscheri, a mouse pathogen that is recovered from cecal content (Amao, et
303
al., 1990). The failure to recover C. ulcerans 0102 from the intestine might
304
reflect a clearance of the organism prior to reaching the lower intestinal
305
tract, possibly by acidity in the stomach.
306
The organism was found in the lung, liver, kidney, and spleen. Lesions
307
were observed in most of these organs except in the liver (Fig. 3). C.
308
ulcerans 0102 was recovered from all of these organs from day 2 until day
309
12 after inoculation (Fig. 4). Since the organism was found in the blood at
310
days 2 and 4 (Fig. 2), it could have disseminated through the bloodstream
311
from the initial infection site to other organs. Such distribution was
312
observed not only in dead animals but also in sacrificed animals,
313
suggesting that intranasal inoculation eventually results in systemic
314
infection, at least at an inoculum size larger than 106 CFU per mouse, and
20
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
299
occasionally leads to death of the animals. The decrease of rectal
316
temperature and body weight (data not shown) during the observation
317
period also supports that the infection is systemic.
318
No significant increase in serum anti-diphtheria toxin neutralization titer or
319
anti-C. ulcerans IgG titer was observed, probably because the observation
320
period was not sufficiently long to develop detectable levels of antitoxin
321
anti-C. ulcerans IgG. A longer observation period is desirable to detect a
322
significant increase in these titers. Measurement of IgM titer might,
323
however, demonstrate a detectable immune response during the 14-day
324
observation period.
325
In this study, we have demonstrated the potential of an in vivo mouse
326
intranasal experimental infection system for C. ulcerans. The system will
327
serve as a suitable platform for further investigation of bacterial virulence,
328
not only for C. ulcerans but possibly also for C. diphtheriae. Moreover,
329
extended analysis using a diphtheria toxin-sensitive mouse strain, such as
330
diphtheria toxin-sensitive transgenic mice (Saito, et al., 2001), will provide
21
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
315
a more complete understanding of the virulence of both C. ulcerans and C.
332
diphtheriae.
333
Funding
334
This work was supported by a grant of Research on Emerging and
335
Re-emerging Infectious Diseases (H25 -Shinko-Ippan-008) from the
336
Ministry of Health, Labour and Welfare, Japan.
337 338
Conflict of interest
339
The authors declare no conflict of interest.
340 341
Acknowledgements
342
We are grateful to the members of Laboratory of Bacterial Toxins, Toxoids
343
and Antitoxins, Department of Bacteriology II, National Institute of
344
Infectious Diseases and members of Laboratory of Experimental Animal
345
Science, School of Animal Science, Nippon Veterinary and Life Science
346
University, for their help and encouragement.
22
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
331
347
348
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
23
References
350
[1] Amao H, Saito M, Takahashi KW & Nakagawa M (1990) Selective
351
medium for Corynebacterium kutscheri and localization of the organism in
352
mice and rats (in Japanese). Jikken Dobutsu 39: 519-529.
353
[2] Amao H, Komukai Y, Sugiyama M, Takahashi KW, Sawada T & Saito
354
M (1995) Natural habitats of Corynebacterium kutscheri in subclinically
355
infected ICGN and DBA/2 strains of mice. Lab. Anim. Sci. 45: 6-10.
356
[3] Bonnet JM & Begg NT (1999) Control of diphtheria: guidance for
357
consultants in communicable disease control. Commun. Dis. Public Health
358
2: 242-249.
359
[4] Brownstein DG, Barthold SW, Adams RL, Terwilliger GA & Aftosmis
360
JG (1985) Experimental Corynebacterium kutscheri infection in rats:
361
bacteriology and serology. Lab. Anim. Sci. 35: 135-138.
362
[5] Dias AASO, Silva Jr. FC, Santos LS, et al. (2011) Strain-dependent
363
arthritogenic potential of the zoonotic pathogen Corynebacterium ulcerans.
364
Vet. Microbiol. 153: 323-331.
24
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
349
[6] González-Iglesias P, Scortti M, MacArthur I, Hapeshi A, Rodriguez H,
366
Prescott JF & Vazquez-Boland JA (2014) Mouse lung infection model to
367
assess Rhodococcus equi virulence and vaccine protection. Vet. Microbiol.
368
172: 256-264.
369
[7] Hatanaka A, Tsunoda A, Okamoto M, et al. (2003) Corynebacterium
370
ulcerans diphtheria in Japan. Emerg. Infect. Dis. 9: 752-753.
371
[8] Mandlik A, Swierczynski A, Das A & Ton-That H (2007)
372
Corynebacterium diphtheriae employs specific minor pilins to target
373
human pharyngeal epithelial cells. Mol. Microbiol. 64: 111-124.
374
[9] McNamara PJ, Cuevas WA & Songer JG (1995) Toxic phospholipase D
375
of Corynebacterium pseudotuberculosis, C. ulcerans and Arcanobacterium
376
haemolyticum: cloning and sequence homology. Gene 156: 113-118.
377
[10] Mitamura T, Higashiyama S, Taniguchi N, Klagsburn M & Mekada E
378
(1995) Diphtheria toxin binds to the epidermal growth factor (EGF)-like
379
domain of human heparin-binding EGF-like growth factor/diphtheria toxin
380
receptor and inhibits specifically its mitogenic activity. J. Biol. Chem. 270:
25
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
365
1015-1019.
382
[11] Miyamura K, Nishio S, Ito A, Murata R & Kono R (1974) Micro cell
383
culture method for determination of diphtheria toxin and antitoxin titres
384
using VERO cells. I. Studies on factors affecting the toxin and antitoxin
385
titration. J. Biol. Stand. 2: 189-201.
386
[12] Miyamura K, Tajiri E, Ito A, Murata R & Kono R (1974) Micro cell
387
culture method for determination of diphtheria toxin and antitoxin titres
388
using VERO cells. II. Comparison with the rabbit skin method and
389
practical application for seroepidemiological studies. J. Biol. Stand. 2:
390
203-209.
391
[13] Mutimer MD & Woolcock JB (1982) Experimental Corynebacterium
392
equi infection in mice. J. Reprod. Fertil. Suppl. 32: 469-476.
393
[14] Nakao H & Popovic T (1997) Development of a direct PCR assay for
394
detection of the diphtheria toxin gene. J. Clin. Microbiol. 35: 1651-1655.
395
[15] Ott L, Höller M, Gerlach RG, Hensel M, Rheinlaender J, Schäffer TE
396
& Burkovski A (2010) Corynebacterium diphtheriae invasion-associated
26
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
381
protein (DIP1281) is involved in cell surface organization, adhesion and
398
internalization in epithelial cells. BMC Microbiology 10: 2.
399
[16] Ott L, McKenzie A, Baltazar MT, Britting S, Bischof A, Burkovski A
400
& Hoskisson PA (2012) Evaluation of invertebrate infection models for
401
pathogenic corynebacteria. FEMS Imm. Med. Microbiol. 65: 413-421.
402
[17] Sabbadini PS, Assis MC, Trost E, et al. (2012) Corynebacterium
403
diphtheriae 67-72p hemagglutinin, characterized as the protein DIP0733,
404
contributes to invasion and induction of apoptosis in HEp-2 cells. Microb.
405
Pathog.
406
[18] Saito M, Iwawaki T, Taya C, et al. (2001) Diphtheria toxin receptor -
407
mediated conditional and targeted cell ablation in transgenic mice. Nat.
408
Biotechnol. 19: 746-750.
409
[19] Sangal V & Hoskisson PA (2013) Corynephages: infections of the
410
infectors. Corynebacterium diphtheriae and Related Toxigenic
411
Species,(Burkovski A, eds.), p. 67. Springer, Berlin.
412
[20] Sangal V, Nieminen L, Weinhardt B, et al. (2014) Diphtheria-like
27
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
397
disease caused by toxigenic Corynebacterium ulcerans strain. Emerg Infect
414
Dis 20: 1257-1258.
415
[21] Sekizuka T, Yamamoto A, Komiya T, et al. (2012) Corynebacterium
416
ulcerans 0102 carries the gene encoding diphtheria toxin on a prophage
417
different from the C. diphtheriae NCTC 13129 prophage. BMC Microbiol.
418
12: 72.
419
[22] Torres LdFC, Ribeiro D, Hirata R, Jr., et al. (2013) Multiplex
420
polymerase chain reaction to identify and determine the toxigenicity of
421
Corynebacterium spp with zoonotic potential and an overview of human
422
and animal infections. Mem. Inst .Oswaldo Cruz 108: 272-279.
423
[23] Trost E, Al-Dilaimi A, Papavasiliou P, et al. (2011) Comparative
424
analysis of two complete Corynebacterium ulcerans genomes and detection
425
of candidate virulence factors. BMC Genomics 12: 383.
426 427
28
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
413
428
FIGURES
430
Fig.1. Lethal infection of C. ulcerans in mice
431
Mice were intranasally challenged with bacteria of the indicated inoculum
432
dose (colony-forming units or CFU) and observed daily for 14 days. Deaths
433
of animals were recorded and the survival rate was calculated daily. C.
434
ulcerans 0102 at 106 (n = 20) (○), 107 (n = 6) (∆), and 108 (n = 6) (□), or C.
435
ulcerans ATCC 51799 at 106 (n = 9) (●) and 107 (n = 7) (▲), or C.
436
glutamicum ATCC 13032 at 107 (n = 7) (◆).
437 29
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
429
439
Fig.2. Distribution of C. ulcerans in experimentally infected mice
440
Mice were intranasally challenged with 106/mouse of C. ulcerans 0102,
441
sacrificed at 2-day intervals between day 2 and day 14 after inoculation and
442
specimens were collected (n = 5, except for the samples collected at day 14,
443
where n = 4). Each specimen was graded according to the numbers of
444
bacteria recovered and indicated in bars as color gradient. +++: > 100 CFU,
445
++: > 10 CFU, +: > 1 CFU, －: not detected. CFU: colony-forming unit(s).
30
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
438
447
Fig. 3. Lesions caused by C. ulcerans intranasal infection
448
Animals were infected intranasally with 106 C. ulcerans 0102 per mouse.
449
Organs were removed from animals that died during the observation period.
450
Arrows indicate lesions.
451
31
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
446
452
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
453 454
Fig. 4. Dissemination of C. ulcerans 0102 in organs of experimentally
455
infected animals
456
Mice were infected intranasally with 106 C. ulcerans 0102 per mouse.
457
Organ weight is expressed relative to the organ weight in a nontreated
458
animal at day 14 (in percentage). Open bars represent recovered bacterial
459
count (CFU/g organ weight). Other symbols represent organ weight in each
460
panel. Error bars represent standard errors (n = 5, except for the samples
32
461
collected at day 14, where n = 4). A, lung; B, liver; C, kidney; D, spleen.
462
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
33
464
Fig. 5. Construction of an ELISA system for anti-C. ulcerans antibody
465
titer
466
Serially diluted rabbit and mouse hyperimmune sera, raised against
467
formalin-killed C. ulcerans 0102 cells, were added to a 96-well plate
468
coated with formalin-denatured C. ulcerans antigen. HRP-anti rabbit
469
Immunoglobulins or HRP-anti mouse IgG were added and the reaction was
470
visualized by pigment as described in the Materials and Methods. The
471
horizontal axes represent the dilution of the primary antibody in
472
logarithmic scale. Vertical axes represent the absorbance at 450 nm. Assays 34
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
463
were carried out at n=10. Bars indicate standard errors. 473
474
475
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
35
1
A novel experimental platform for toxigenic and nontoxigenic
2
Corynebacterium ulcerans infection in mice
3 4
Yu Mochizuki1,3‡, Honami Saeki1,3¶, Masaaki Iwaki1*, Hitrotaka Takagi2,
5
Keigo Shibayama1, Hiromi Amao3 and Akihiko Yamamoto1§
6
1
7
Research, National Institute of Infectious Diseases, Tokyo, Japan, 3 School
8
of Animal Science, Nippon Veterinary and Life Science University, Tokyo,
9
Japan * Corresponding Author: Tel: +81-42-561-0771 Fax: +81-42-561-7173 e-mail: [email protected] Running title: Experimental Corynebacterium ulcerans infection in mice ‡ Present address: Vaccine Research Department, Denka Seiken Co. Ltd. Niigata, Japan
20
¶ Present address: KAC Inc. Kyoto, Japan
21 22 23 24 25 26 27 28 29 30
§ Present address: Division of Biosafety Control and Research, National Institute of Infectious Diseases, Tokyo, Japan Keywords: zoonosis, diphtheria, respiratory infection, animal model, Corynebacterium, intranasal infection One-sentence summary: An intranasal experimental model for Coryebacterium ulcerans was constructed in mice to provide a platform for the analysis of its virulence.
1
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
10 11 12 13 14 15 16 17 18 19
Department of Bacteriology II, 2 Division of Biosafety Control and
Abstract
32
Corynebacterium ulcerans is a zoonotic pathogen that can produce
33
diphtheria toxin and causes an illness categorized as diphtheria in the
34
European Union, because its clinical appearance is similar to that of
35
diphtheria caused by Corynebacterium diphtheriae. Despite the importance
36
of the pathogen for public health, the organism’s mechanism of infection
37
has not been extensively studied, especially in experimental animal models.
38
In the present study, we thus constructed an intranasal infection system for
39
mice, which are insensitive to diphtheria toxin and advantageous in
40
excluding the cytotoxic effect of the toxin that might interfere the analysis
41
of the early stage of infection. Both the toxigenic and nontoxigenic C.
42
ulcerans strains were capable of killing mice within three days after
43
inoculation at 107 colony-forming units (CFU) per mouse. In
44
experimentally infected animals, C. ulcerans was detected in the
45
respiratory tract but not in the intestinal tract. The bacterium was also
46
detected in peripheral blood and it disseminated into the lung, kidney, and
47
spleen to produce a systemic infection. This experimental infection system 2
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
31
provides a platform for analyzing C. ulcerans virulence in future studies. 48
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
3
Introduction
50
Corynebacterium ulcerans is a zoonotic pathogen. The organism can be
51
lysogenized with bacteriophages harboring the tox gene (Sekizuka, et al.,
52
2012) (Sangal & Hoskisson, 2013, Sangal, et al., 2014) and produces
53
diphtheria toxin. Human infection by toxigenic C. ulcerans causes an
54
illness clinically indistinguishable from diphtheria caused by
55
Corynebacterium diphtheriae. Therefore, C. ulcerans infection is
56
categorized as diphtheria in regions including European countries (Bonnet
57
& Begg, 1999).
58
Unlike the mechanism of infection by C. diphtheriae, the infection
59
mechanisms of C. ulcerans are not well understood. In C. diphtheriae
60
infection, many factors linked to the early stage of the infection have been
61
identified through in vitro studies. These include adhesion factors,
62
including pili (Mandlik, et al., 2007), and invasion factors (Ott, et al., 2010,
63
Sabbadini, et al., 2012). In contrast, in C. ulcerans, phospholipase D has
64
been identified as a virulence factor (McNamara, et al., 1995) and after the
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
49
4
genome sequence became available, several other candidate virulence
66
factors have been identified (Trost, et al., 2011, Sekizuka, et al., 2012).
67
However, the significance of these factors for pathogenicity remains to be
68
elucidated. Only a few reports have been published on animal models for C.
69
ulcerans infection: an intravenous mouse infection model (Dias, et al.,
70
2011) and a nematode and Galleria moth larvae infection model with
71
nontoxigenic strains (Ott, et al., 2012). As C. ulcerans is considered to be a
72
primarily respiratory pathogen, a system reflecting respiratory infection is
73
desirable in order to further elucidate its infection mechanisms.
74
One of the expected problems associated with analyzing the early stage of
75
infection by animal experiment is that diphtheria toxin might interfere the
76
analysis by its potent deteriorating effects on infection site. We thus chose
77
mice, as host animals, which are insensitive to diphtheria toxin due to the
78
lack of a functional receptor.
79
In this respect, a mouse intranasal infection model could provide a suitable
80
novel platform to analyze the pathogenicity of C. ulcerans in vivo. In fact,
5
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
65
intranasal infection models have previously been employed for in vivo
82
experimental infection models of various kinds of related pathogenic
83
bacterial species including Corynebacterium kutscheri (Brownstein, et al.,
84
1985), Rhodococcus hoagii (Mutimer & Woolcock, 1982,
85
González-Iglesias, et al., 2014), suggesting that this model could be
86
successfully applied to C. ulcerans as well.
87
In the present study, we constructed a mouse intranasal infection system in
88
order to establish a platform for analyzing the pathogenicity of C. ulcerans.
89
Experimental lethal infection with toxigenic and nontoxigenic C. ulcerans,
90
as well as analysis of distribution of the bacteria in infected animals and an
91
approach for measuring serum titers are described.
92 93 94
6
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
81
Materials and Methods
96
Animals
97
Female BALB/cCrSlc mice (6 weeks old) were used for all experiments in
98
this study. Mice were acclimatized for 4 days before being subjected to
99
experiments. Animals were sacrificed by exsanguination under sedation
100
with sodium pentobarbital (Somnopentyl, Kyoritsu Seiyaku Corporation,
101
Tokyo, Japan). All animal experiments were carried out with strict adhesion
102
to the Guidelines for Animal Welfare of the Japanese National Institute of
103
Infectious Diseases.
104 105
Culture media
106
Blood agar plates (Nissui Pharmaceutical Co. Ltd., Japan) were used for
107
the propagation of bacterial cells from frozen stocks. Brain Heart infusion
108
(BHI) broth (Becton-Dickinson, NJ, USA) was used for broth culture in
109
preparation of the inoculum and for agar plates for bacterial detection and
110
colony counting in experimentally infected animals. Furazolidone-
7
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
95
Nalidixic acid- Colimycin agar (FNC agar) plates were prepared according
112
to Amao et al. (Amao, et al., 1995) using PearlCore heart infusion agar
113
(Eiken Chemical Co. Ltd., Tokyo, Japan), furazolidone (Sigma-Aldrich CO.
114
LLC., MO, USA), nalidixic acid (Wako Pure Chemical Industries Ltd.
115
Osaka, Japan) and Colimycin® (Pola Pharma Inc., Tokyo, Japan).
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
111
116 117
Bacterial strains
118
Toxigenic Corynebacterium ulcerans 0102 was isolated from a case of
119
respiratory infection in Japan in 2001 (Hatanaka, et al., 2003).
120
Nontoxigenic C. ulcerans ATCC51799 and Corynebacterium glutamicum
121
ATCC13032 were purchased from the American Type Culture Collection
122
(VA, USA).
123 124
Experimental infection of mice
125
Inoculum preparation: Bacteria were propagated from frozen stocks on a
126
blood agar plate at 37°C for 48 h, and a single colony was picked and used
8
to seed 10 mL of BHI broth. The culture was shaken at 37°C for 72 h and
128
then vortexed to disperse bacterial cells. The optical density (OD) at 600
129
nm was measured to assure that growth did not exceed an OD of 2.0.
130
Bacterial cell pellets were washed twice and finally resuspended in a sterile
131
saline solution, and adjusted to an OD600 of 8.8 (= 1× suspension). By
132
plating the bacteria on BHI agar plates after appropriate dilution, the 1×
133
suspension was confirmed to give 2 × 109 colony-forming units (CFU)
134
mL-1 (= 108 CFU per 50 µL).
135
Intranasal challenge: Mice were anesthetized by intramuscular injection
136
of a mixture of ketamin hydrochloride (Ketalar, Daiichi Sankyo Co. Ltd.,
137
Tokyo, Japan) and xylazine hydrochloride (Bayer Yakuhin Ltd., Osaka,
138
Japan). Ketamin hydrochloride was handled by a license holder according
139
to the Japanese Narcotics and Psychotropics Control Law. To the nasal
140
cavity of anesthetized mice, 50 µL (25 µL for each side) of bacterial
141
suspension was inoculated, aided by the breathing of the mice. Sterile
142
saline was used as a negative control.
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
127
9
Observation and sampling of challenged mice: After the bacterial
144
challenge, mice were observed for 14 days. To recover the inoculated
145
bacterium, the deceased or sacrificed mice were subjected to bleeding and
146
swabs were collected from the nasal cavity, oral cavity, and trachea.
147
Additional samples were collected after autopsy: the lung, liver, kidney,
148
and spleen were aseptically removed, weighed, and homogenized in a
149
sterile saline solution (10× volume) with a "Biomasher" homogenizer
150
(Nippi Inc., Tokyo, Japan). Cecal content were collected aseptically by
151
opening the cecum at autopsy. Colorectal feces were collected by
152
evacuation by the mice into sterile test tubes.
153
Identification of viable C. ulcerans from collected samples: Viable
154
C. ulcerans was detected by swab streaking (nasal and tracheal swabs) or
155
by plating appropriately diluted organ homogenates (lung, liver, kidney,
156
spleen) or blood on BHI agar plates. For samples in which interference by
157
normal flora components is expected (oral and eye swabs, cecal content,
158
and colorectal feces), streaking and plating occurred on selective FNC agar
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
143
10
plates. For swabs, a rough population estimation was done by counting the
160
colonies appearing on the streaks. For cecal content, colorectal feces, blood,
161
and organ homogenates, all colonies appearing on the plates were counted.
162
C. ulcerans identification: Colonies appearing on the BHI or FNC agar
163
plates were confirmed to be C. ulcerans based on colony morphology or
164
Gram staining, and when needed by identification with API Coryne kit
165
(bioMérieux SA, Marcy l'Etoile, France) and PCR directed to the
166
diphtheria toxin gene (A fragment) and the phospholipase D (PLD) gene.
167
The following primers were used: Tox1 (5′-ATC CAC TTT TAG TGC
168
GAG AAC CTT CGT CA-3′) and Tox2 (5′-GAA AAC TTT TCT TCG
169
TAC CAC GGG ACT AA-3′) (Nakao & Popovic, 1997) for the diphtheria
170
toxin gene, and PLD-F (5′-ATA AGC GTA AGC AGG GAG CA-3′) and
171
PLD-R (5′-ATC AGC GGT GAT TGT CTT CC-3′) (Torres, et al., 2013) for
172
the phospholipase D gene.
173
Measurement of serum anti-diphtheria toxin and anti-C. ulcerans titers
174
Measurement of serum antitoxin titer was performed according to
11
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
159
previously described procedures (Miyamura, et al., 1974, Miyamura, et al.,
176
1974). Briefly, serially diluted sera (25 µL) were mixed with 25 µL of
177
diphtheria toxin (16 CD50 in 25 µL in Minimum Essential Medium (Eagle)
178
(Sigma-Aldrich Co. LLC.) containing 3% newborn calf serum (Life
179
Technologies Corporation, CA, USA)) and added to the wells of 96-well
180
plates containing 175 µL of medium with 1.5 × 105 Vero cells. These
181
cultures were incubated for 4 days at 37°C and the endpoint of
182
neutralization was determined by observation of the cell morphology. The
183
outcome was compared to the endpoint of standard diphtheria antitoxin
184
based on the defined International Unit (Japanese National Diphtheria
185
Antitoxin, equine, lot 10). Antitoxin titers of the tested sera were calculated
186
by parallel line assays.
187
The anti-C. ulcerans serum IgG titer was measured by ELISA. C. ulcerans
188
0102 cells were collected from overnight culture in BHI broth. After
189
washing with PBS, the bacteria were killed by heating at 60°C for 1 h and
190
then treated with 0.01% formalin at 4°C for 48 h with daily changing of the
12
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
175
formalin solution. Formalin-denatured antigen was then appropriately
192
diluted in PBS and coated on Polysorp™ plates (Thermo Scientific)
193
overnight at 4°C. Rabbit or mouse sera were serially diluted with PBS
194
containing 0.5% BSA. One hundred microliters of diluted sera were added
195
to wells as primary antibody and incubated for 1 h at 37°C. After washing
196
three times with PBS containing 0.05% Tween 20, appropriately diluted
197
HRP-anti rabbit immunoglobulins (Dako) or HRP-anti mouse IgG (EMD
198
Millipore) was added as secondary antibody and incubated for 1 h at 37°C.
199
After washing three times with PBS containing 0.05% Tween 20,
200
specifically bound secondary antibody was visualized with
201
3,3',5,5'-tetramethylbenzidine. The enzyme reaction was stopped with
202
sulfuric acid and the absorbance at 450 nm was measured. Anti-C. ulcerans
203
mouse serum antibody titer was expressed as a factor of endpoint dilution
204
and was significantly higher than that in negative control wells incubated
205
without primary antibody.
206
13
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
191
Results
208
Lethal effect of intranasal inoculation of C. ulcerans in mice
209
C. ulcerans toxigenic 0102 and nontoxigenic ATCC 51799 strains, as well
210
as the C. glutamicum ATCC 13032 strain were propagated in BHI broth,
211
appropriately diluted and intranasally inoculated into anesthetized mice.
212
Fig. 1 shows the survival curve of the inoculated mice. Fifty percent of the
213
mice inoculated with 106 CFU of the toxigenic C. ulcerans 0102 strain
214
(n = 20) died before the end of the observation period (14 days). At 10 7 and
215
108 CFU, all mice (n = 6 for both) died within 4 and 3 days, respectively. In
216
contrast, using the nontoxigenic C. ulcerans ATCC 51799, at 106 CFU all
217
(n = 9) mice survived until the end of the observation period. However, at
218
107 CFU, the nontoxigenic strain displayed a more pronounced lethal effect
219
than the 0102 strain, as all mice (n = 7) died within 2 days. The avirulent
220
strain C. glutamicum ATCC 13032, frequently used in the fermentation
221
industry for the production of amino acids, did not show any lethal effects
222
at 107 (n = 7). Sterile saline solution did not show any lethal effect either
14
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
207
(data not shown).
224
Distribution of C. ulcerans 0102 in infected mice
225
In order to investigate the distribution of the C. ulcerans 0102 strain after
226
experimental infection, mice were inoculated with 106 CFU of the
227
bacterium, sacrificed at 2-day intervals and the presence of viable C.
228
ulcerans 0102 was estimated by plating swabs from respiratory tracts and
229
contents of the intestinal tracts. Fig. 2 shows the recovery of C. ulcerans
230
0102 from these specimens. The population of the bacterium detected in
231
each specimen was assessed according to a grading system (- to +++). As
232
shown in Fig. 2, the bacterium was recovered from respiratory tracts, eyes,
233
and blood. In the respiratory tract, the estimated population was largest in
234
the trachea, and smaller in the nasal and oral cavities. The organism was
235
recovered from trachea throughout the observation period, whereas in nasal
236
and oral cavities the bacterium was undetectable within 8 and 6 days,
237
respectively. The organism was detected in circulating blood until day 4
238
after inoculation, but was undetectable afterwards. In addition, the
15
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
223
organism was detected until day 8 in eye swabs.
240
C. ulcerans also disseminated into various organs including the lung, liver,
241
kidney, and spleen. Fig. 3 shows the lesions formed in lung, kidney, and
242
spleen on days 5, 9 and 13, respectively, after inoculation. Clearly
243
distinguishable legions were formed in these organs. Fig. 4 illustrates the
244
presence of the organism in the lung, liver, kidney, and spleen, and the
245
increase/decrease in organ weights measured at 2-day intervals. As shown
246
in Fig. 4, the bacterium was detected in these organs until day 12 after
247
inoculation. Enlargement of the spleen was also observed at day 8 (Fig.
248
4D).
249
Serum anti-diphtheria toxin and anti-C. ulcerans antibody titer
250
Serum anti-diphtheria toxin neutralization titer was measured according to
251
Miyamura et al. (Miyamura, et al., 1974, Miyamura, et al., 1974). Mice
252
inoculated with 106 CFU of C. ulcerans 0102 were bled at day 8, 10, 12
253
(n = 5 for each), and day 14 (n = 4), and subjected to titer determination. In
254
all specimens, the anti-diphtheria toxin neutralization titer was below the
16
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
239
detection limit (data not shown). The serum anti-C. ulcerans titer was
256
measured by an ELISA system developed in this study. Fig. 5 illustrates the
257
profile of absorbance at 450 nm obtained with serially diluted rabbit and
258
mouse hyperimmune sera using this system. Within the linear region
259
corresponding to 2 × 103–1 × 105 times dilution for rabbit hyperimmune
260
serum and 4 × 102–1 × 104 times dilution for mouse hyperimmune serum,
261
anti-C. ulcerans titer determination was shown to be feasible. Using this
262
system, anti-C. ulcerans IgG titer in sera from inoculated mice was
263
measured. As was the case for the anti-diphtheria toxin neutralization titer,
264
all of the serum exhibited low absorbance at 450 nm, below the lower limit
265
of the linear range (data not shown).
266
17
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
255
Discussion
268
In this study, we intended to provide a platform for the analysis of
269
C. ulcerans virulence in vivo. To reflect the most common route of
270
infection for this organism, an intranasal experimental infection system was
271
constructed. We demonstrate here that the infection model is effective as it
272
enables lethal infection.
273
An in vivo experimental infection system is desirable for further analysis of
274
the infection mechanism of C. ulcerans (and also for C. diphtheriae). In
275
analyzing the mechanisms underlying the early stages of infection such as
276
adhesion and colonization, the toxic effects of diphtheria toxin may serve
277
as an interfering factor through deterioration of the cells interacting with
278
the organism. In this respect, mice are suitable candidates for the platform
279
as they are highly resistant to diphtheria toxin due to the lack of a
280
functional receptor (Mitamura, et al., 1995). Previous reports on
281
invertebrate C. ulcerans and C. diphtheriae infection models employ
282
nontoxigenic strains, also resulting in exclusion of diphtheria toxin effects
18
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
267
(Ott, et al., 2012). Additional advantages of using mice are that they are
284
easy to handle and that a large number of animals can be studied.
285
Using mice, we found that C. ulcerans was capable of killing the animals
286
when administered intranasally (Fig. 1). The toxigenic C. ulcerans 0102
287
killed 50% of all mice when inoculated at 106 CFU, but the total survival
288
was similar for nontoxigenic C. ulcerans ATCC 51799. At a ten times
289
larger number of inoculated bacterial cells, the potency of killing effects
290
reverted: 107 CFU of the 0102 strain killed all animals within 4 days and
291
ATCC 51799 strains killed all mice in 2 days. Avirulent C. glutamicum had
292
no effect (Fig. 1), indicating that the lethality was not merely a nonspecific
293
effect. Based on these results and the fact that the mice are resistant to the
294
toxin, it seemed unlikely that toxigenicity is a crucial factor in the lethality
295
of C. ulcerans in the experimental system presented here. In other words, C.
296
ulcerans can produce a lethal infection in mice irrespective of toxigenicity.
297
In infected animals, C. ulcerans 0102 was recovered from the surface of
298
the respiratory tract but not from the contents of the lower intestinal tract
19
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
283
(Fig. 2). Although the bacterium was initially detected in the oral cavity
300
(Fig. 2), it was not recovered from cecal content and colorectal feces
301
throughout the observation period. This is in contrast to Corynebacterium
302
kutscheri, a mouse pathogen that is recovered from cecal content (Amao, et
303
al., 1990). The failure to recover C. ulcerans 0102 from the intestine might
304
reflect a clearance of the organism prior to reaching the lower intestinal
305
tract, possibly by acidity in the stomach.
306
The organism was found in the lung, liver, kidney, and spleen. Lesions
307
were observed in most of these organs except in the liver (Fig. 3). C.
308
ulcerans 0102 was recovered from all of these organs from day 2 until day
309
12 after inoculation (Fig. 4). Since the organism was found in the blood at
310
days 2 and 4 (Fig. 2), it could have disseminated through the bloodstream
311
from the initial infection site to other organs. Such distribution was
312
observed not only in dead animals but also in sacrificed animals,
313
suggesting that intranasal inoculation eventually results in systemic
314
infection, at least at an inoculum size larger than 106 CFU per mouse, and
20
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
299
occasionally leads to death of the animals. The decrease of rectal
316
temperature and body weight (data not shown) during the observation
317
period also supports that the infection is systemic.
318
No significant increase in serum anti-diphtheria toxin neutralization titer or
319
anti-C. ulcerans IgG titer was observed, probably because the observation
320
period was not sufficiently long to develop detectable levels of antitoxin
321
anti-C. ulcerans IgG. A longer observation period is desirable to detect a
322
significant increase in these titers. Measurement of IgM titer might,
323
however, demonstrate a detectable immune response during the 14-day
324
observation period.
325
In this study, we have demonstrated the potential of an in vivo mouse
326
intranasal experimental infection system for C. ulcerans. The system will
327
serve as a suitable platform for further investigation of bacterial virulence,
328
not only for C. ulcerans but possibly also for C. diphtheriae. Moreover,
329
extended analysis using a diphtheria toxin-sensitive mouse strain, such as
330
diphtheria toxin-sensitive transgenic mice (Saito, et al., 2001), will provide
21
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
315
a more complete understanding of the virulence of both C. ulcerans and C.
332
diphtheriae.
333
Funding
334
This work was supported by a grant of Research on Emerging and
335
Re-emerging Infectious Diseases (H25 -Shinko-Ippan-008) from the
336
Ministry of Health, Labour and Welfare, Japan.
337 338
Conflict of interest
339
The authors declare no conflict of interest.
340 341
Acknowledgements
342
We are grateful to the members of Laboratory of Bacterial Toxins, Toxoids
343
and Antitoxins, Department of Bacteriology II, National Institute of
344
Infectious Diseases and members of Laboratory of Experimental Animal
345
Science, School of Animal Science, Nippon Veterinary and Life Science
346
University, for their help and encouragement.
22
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
331
347
348
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
23
References
350
[1] Amao H, Saito M, Takahashi KW & Nakagawa M (1990) Selective
351
medium for Corynebacterium kutscheri and localization of the organism in
352
mice and rats (in Japanese). Jikken Dobutsu 39: 519-529.
353
[2] Amao H, Komukai Y, Sugiyama M, Takahashi KW, Sawada T & Saito
354
M (1995) Natural habitats of Corynebacterium kutscheri in subclinically
355
infected ICGN and DBA/2 strains of mice. Lab. Anim. Sci. 45: 6-10.
356
[3] Bonnet JM & Begg NT (1999) Control of diphtheria: guidance for
357
consultants in communicable disease control. Commun. Dis. Public Health
358
2: 242-249.
359
[4] Brownstein DG, Barthold SW, Adams RL, Terwilliger GA & Aftosmis
360
JG (1985) Experimental Corynebacterium kutscheri infection in rats:
361
bacteriology and serology. Lab. Anim. Sci. 35: 135-138.
362
[5] Dias AASO, Silva Jr. FC, Santos LS, et al. (2011) Strain-dependent
363
arthritogenic potential of the zoonotic pathogen Corynebacterium ulcerans.
364
Vet. Microbiol. 153: 323-331.
24
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
349
[6] González-Iglesias P, Scortti M, MacArthur I, Hapeshi A, Rodriguez H,
366
Prescott JF & Vazquez-Boland JA (2014) Mouse lung infection model to
367
assess Rhodococcus equi virulence and vaccine protection. Vet. Microbiol.
368
172: 256-264.
369
[7] Hatanaka A, Tsunoda A, Okamoto M, et al. (2003) Corynebacterium
370
ulcerans diphtheria in Japan. Emerg. Infect. Dis. 9: 752-753.
371
[8] Mandlik A, Swierczynski A, Das A & Ton-That H (2007)
372
Corynebacterium diphtheriae employs specific minor pilins to target
373
human pharyngeal epithelial cells. Mol. Microbiol. 64: 111-124.
374
[9] McNamara PJ, Cuevas WA & Songer JG (1995) Toxic phospholipase D
375
of Corynebacterium pseudotuberculosis, C. ulcerans and Arcanobacterium
376
haemolyticum: cloning and sequence homology. Gene 156: 113-118.
377
[10] Mitamura T, Higashiyama S, Taniguchi N, Klagsburn M & Mekada E
378
(1995) Diphtheria toxin binds to the epidermal growth factor (EGF)-like
379
domain of human heparin-binding EGF-like growth factor/diphtheria toxin
380
receptor and inhibits specifically its mitogenic activity. J. Biol. Chem. 270:
25
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
365
1015-1019.
382
[11] Miyamura K, Nishio S, Ito A, Murata R & Kono R (1974) Micro cell
383
culture method for determination of diphtheria toxin and antitoxin titres
384
using VERO cells. I. Studies on factors affecting the toxin and antitoxin
385
titration. J. Biol. Stand. 2: 189-201.
386
[12] Miyamura K, Tajiri E, Ito A, Murata R & Kono R (1974) Micro cell
387
culture method for determination of diphtheria toxin and antitoxin titres
388
using VERO cells. II. Comparison with the rabbit skin method and
389
practical application for seroepidemiological studies. J. Biol. Stand. 2:
390
203-209.
391
[13] Mutimer MD & Woolcock JB (1982) Experimental Corynebacterium
392
equi infection in mice. J. Reprod. Fertil. Suppl. 32: 469-476.
393
[14] Nakao H & Popovic T (1997) Development of a direct PCR assay for
394
detection of the diphtheria toxin gene. J. Clin. Microbiol. 35: 1651-1655.
395
[15] Ott L, Höller M, Gerlach RG, Hensel M, Rheinlaender J, Schäffer TE
396
& Burkovski A (2010) Corynebacterium diphtheriae invasion-associated
26
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
381
protein (DIP1281) is involved in cell surface organization, adhesion and
398
internalization in epithelial cells. BMC Microbiology 10: 2.
399
[16] Ott L, McKenzie A, Baltazar MT, Britting S, Bischof A, Burkovski A
400
& Hoskisson PA (2012) Evaluation of invertebrate infection models for
401
pathogenic corynebacteria. FEMS Imm. Med. Microbiol. 65: 413-421.
402
[17] Sabbadini PS, Assis MC, Trost E, et al. (2012) Corynebacterium
403
diphtheriae 67-72p hemagglutinin, characterized as the protein DIP0733,
404
contributes to invasion and induction of apoptosis in HEp-2 cells. Microb.
405
Pathog.
406
[18] Saito M, Iwawaki T, Taya C, et al. (2001) Diphtheria toxin receptor -
407
mediated conditional and targeted cell ablation in transgenic mice. Nat.
408
Biotechnol. 19: 746-750.
409
[19] Sangal V & Hoskisson PA (2013) Corynephages: infections of the
410
infectors. Corynebacterium diphtheriae and Related Toxigenic
411
Species,(Burkovski A, eds.), p. 67. Springer, Berlin.
412
[20] Sangal V, Nieminen L, Weinhardt B, et al. (2014) Diphtheria-like
27
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
397
disease caused by toxigenic Corynebacterium ulcerans strain. Emerg Infect
414
Dis 20: 1257-1258.
415
[21] Sekizuka T, Yamamoto A, Komiya T, et al. (2012) Corynebacterium
416
ulcerans 0102 carries the gene encoding diphtheria toxin on a prophage
417
different from the C. diphtheriae NCTC 13129 prophage. BMC Microbiol.
418
12: 72.
419
[22] Torres LdFC, Ribeiro D, Hirata R, Jr., et al. (2013) Multiplex
420
polymerase chain reaction to identify and determine the toxigenicity of
421
Corynebacterium spp with zoonotic potential and an overview of human
422
and animal infections. Mem. Inst .Oswaldo Cruz 108: 272-279.
423
[23] Trost E, Al-Dilaimi A, Papavasiliou P, et al. (2011) Comparative
424
analysis of two complete Corynebacterium ulcerans genomes and detection
425
of candidate virulence factors. BMC Genomics 12: 383.
426 427
28
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
413
428
FIGURES
430
Fig.1. Lethal infection of C. ulcerans in mice
431
Mice were intranasally challenged with bacteria of the indicated inoculum
432
dose (colony-forming units or CFU) and observed daily for 14 days. Deaths
433
of animals were recorded and the survival rate was calculated daily. C.
434
ulcerans 0102 at 106 (n = 20) (○), 107 (n = 6) (∆), and 108 (n = 6) (□), or C.
435
ulcerans ATCC 51799 at 106 (n = 9) (●) and 107 (n = 7) (▲), or C.
436
glutamicum ATCC 13032 at 107 (n = 7) (◆).
437 29
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
429
439
Fig.2. Distribution of C. ulcerans in experimentally infected mice
440
Mice were intranasally challenged with 106/mouse of C. ulcerans 0102,
441
sacrificed at 2-day intervals between day 2 and day 14 after inoculation and
442
specimens were collected (n = 5, except for the samples collected at day 14,
443
where n = 4). Each specimen was graded according to the numbers of
444
bacteria recovered and indicated in bars as color gradient. +++: > 100 CFU,
445
++: > 10 CFU, +: > 1 CFU, －: not detected. CFU: colony-forming unit(s).
30
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
438
447
Fig. 3. Lesions caused by C. ulcerans intranasal infection
448
Animals were infected intranasally with 106 C. ulcerans 0102 per mouse.
449
Organs were removed from animals that died during the observation period.
450
Arrows indicate lesions.
451
31
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
446
452
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
453 454
Fig. 4. Dissemination of C. ulcerans 0102 in organs of experimentally
455
infected animals
456
Mice were infected intranasally with 106 C. ulcerans 0102 per mouse.
457
Organ weight is expressed relative to the organ weight in a nontreated
458
animal at day 14 (in percentage). Open bars represent recovered bacterial
459
count (CFU/g organ weight). Other symbols represent organ weight in each
460
panel. Error bars represent standard errors (n = 5, except for the samples
32
461
collected at day 14, where n = 4). A, lung; B, liver; C, kidney; D, spleen.
462
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
33
464
Fig. 5. Construction of an ELISA system for anti-C. ulcerans antibody
465
titer
466
Serially diluted rabbit and mouse hyperimmune sera, raised against
467
formalin-killed C. ulcerans 0102 cells, were added to a 96-well plate
468
coated with formalin-denatured C. ulcerans antigen. HRP-anti rabbit
469
Immunoglobulins or HRP-anti mouse IgG were added and the reaction was
470
visualized by pigment as described in the Materials and Methods. The
471
horizontal axes represent the dilution of the primary antibody in
472
logarithmic scale. Vertical axes represent the absorbance at 450 nm. Assays 34
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
463
were carried out at n=10. Bars indicate standard errors. 473
474
475
Downloaded from http://femspd.oxfordjournals.org/ by guest on December 21, 2015
35
