Jul 3, 2013 ... that are rapidly eliminated by phagocytosis (13). ..... Qa'Dan, M., M. Ramsey, J.
Daniel, L. M. Spyres, B. Safiejko-Mroczka, W. Ortiz-Leduc, and ... Brito, G. A., J.
Fujji, B. A. Carneiro-Filho, A. A. Lima, T. Obrig, and R. L. Guerrant.
JCM Accepts, published online ahead of print on 3 July 2013 J. Clin. Microbiol. doi:10.1128/JCM.01161-13 Copyright © 2013, American Society for Microbiology. All Rights Reserved.
1
Caspase Activation as a Versatile Assay Platform for Detection of Cytotoxic
2
Bacterial Toxins
3
Angela M. Payne*, Julie Zorman, Melanie Horton, Sheri Dubey, Jan terMeulen,
5
Kalpit A Vora*
6
Vaccines Basic Research, Merck Research Laboratories
7
West Point, PA 19486
8
Running Title: Caspase induction for detection of bacterial toxins
9
10
*Corresponding Authors
11
Address: Merck and Co., Inc.
12
770 Sumneytown Pike
13
P.O. Box 4
14
West Point, PA 19486
15
Phone: 215-652-8892
16
FAX: 215-652-2142
17
e-mail:
[email protected]
18
e-mail:
[email protected]
19 20
1
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
4
Abstract
22
Pathogenic bacteria produce several virulence factors that help them establish infection in permissive
23
hosts. Bacterial toxins are a major class of virulence factors and hence are attractive therapeutic targets
24
for vaccine development. Here we describe the development of a rapid sensitive and high-throughput
25
assay that could be used as a versatile platform to measure the activity of bacterial toxins. We have
26
exploited the ability of toxins to cause cell death via apoptosis of sensitive cultured cell lines as a read
27
out to measure toxin activity. Caspases are induced early in the apoptotic pathway and hence we used
28
their induction to measure the activity of Clostridium difficile toxins A (TcdA), B (TcdB), binary (CDTa-
29
CDTb), Corynebacterium diphtheria (DT) and Pseudomonas aeruginosa exotoxin A (PEA). The caspase
30
induction in cell lines, upon exposure to toxins, was optimized for toxin concentration and intoxication
31
time, and the specificity of caspase activity was established using genetically mutated toxin and use of a
32
pan-caspase inhibitor. In addition, we demonstrate the utility of the assay to measure toxin potency as
33
well as neutralizing antibody (NAb) activity against C. difficile toxins. Furthermore, the caspase assay
34
showed excellent correlation with the F-actin polymerization assay to measure TcdA and TcdB
35
neutralization titers upon vaccination of hamsters. These results demonstrate that the detection of
36
caspase induction due to toxin exposure using a chemiluminescence readout would be able to support
37
potency and clinical immunogenicity testing for bacterial toxin vaccine candidates in development.
38
2
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
21
Introduction
40
Microorganisms cause pathogenesis by means of a wide variety of molecules called virulence factors. A
41
large number of divergent microbial pathogens synthesize toxins recognized as primary virulence factors
42
which affect the metabolism and cause damage to eukaryotic cells, many times with lethal effects to the
43
host (1, 2). Major symptoms associated with diseases such as diphtheria, whooping cough, cholera,
44
anthrax and dysentery, are all related to activities of toxins produced by bacteria. In recognizing the
45
central role of toxins in these and other diseases, bacterial toxins have become attractive targets for the
46
development of vaccines (1, 3). Bacterial toxins affect susceptible host cells by a variety of modes of
47
action: damage of cell membranes, inhibition of protein synthesis, activation of immune response
48
leading to cellular damage, resulting in direct cell lysis, and facilitating bacterial spread through tissues
49
(4). Organisms such as C. difficile, C. diphtheriae and P. aeruginosa, secrete toxins involved in different
50
ways in the pathogenesis of disease. C. difficile toxins, for example, cause cellular toxicity through
51
glucosylation of Rho G-protein, and ADP-ribosylation of actin, while C. diphtheriae and P. aeruginosa
52
toxins catalyze the transfer of ADP-ribose to elongation factor 2 to block host cell protein synthesis,
53
leading to target cell death(5-8). The clostridial toxin TcdB of C. difficile inactivates the small GTPases
54
Rho, Rac and Cdc42, which has been shown to trigger cell death via apoptosis (2, 9-11).
55
Apoptosis is a fundamental feature of all animal cells, and is essential for normal development and
56
tissue homeostasis, whereas unregulated apoptosis can create an imbalance in the normal cell
57
proliferation processes (4, 7). Apoptosis is characterized by the presence of distinct morphological and
58
biochemical features (12). Morphologically, it can be characterized by DNA fragmentation, membrane
59
blebbing, cell rounding, cytoskeletal collapse and the formation of membrane-bound apoptotic vesicles
60
that are rapidly eliminated by phagocytosis (13). Biochemical features of apoptotic cell death include
61
the activation of a family of intracellular cysteine endopeptidases known as caspases, that specifically
3
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
39
cleave target proteins at a cysteine amino acid that follows an aspartic acid residue (14, 15). Caspases
63
are synthesized as inactive pro-enzymes, which are converted into active heterodimers by proteolytic
64
cleavage, and are responsible for the deliberate disassembly of the cells into apoptotic bodies(16). Their
65
activation indicates progression of the pathway of cellular apoptosis. The initiator caspases 8 and 9, and
66
the executioner caspase 3, are positioned at crucial junctions in the apoptosis pathways. The activation
67
of the initiator caspases, in response to extracellular cytotoxic agents, activates the executioner caspase
68
3, resulting in a series of events leading eventually to cell lysis and disruption of the normal cell
69
processes (8, 12, 16-18).
70
Bacterial toxins can activate the apoptotic pathways and hence, caspases are molecules of particular
71
interest in assay development as potential indicators of apoptosis due to cell exposure to toxins. A
72
number of cultured cell lines undergo apoptosis when exposed to various cytotoxic signals from
73
pathogens or other sources. Caspase activation occurs early in the programed cell death pathway, and
74
thus allows for early detection from exposure to these toxins. Measurements of caspase activation due
75
to bacterial toxin exposure, or its inhibition, may be used as potency or release tests in vaccine
76
development. The ability to inhibit toxin-induced caspase activation in vitro, by animal and human
77
serum neutralizing antibodies is valuable in evaluation of vaccine efficacy. Conventionally, cell
78
susceptibility to bacterial toxins and neutralizing antibody responses in vitro rely on radioactive
79
cytotoxicity measurements for protein synthesis, or are evaluated by microscopic observation of
80
intoxicated cell monolayers. These methods may be subjective, time consuming and inherently low
81
throughput, due to its requirement for manual observation and counting of cells (5, 7, 9, 19-22).
82
Because of these limitations, we sought to develop an alternative assay which may be used as a versatile
83
platform for measuring bacterial toxin activity, and to evaluate the immunogenicity of toxin-based
84
vaccines. Here we describe that cytotoxic activity of several unrelated bacterial toxins can be easily and
85
reliably quantified by measuring in-cell caspase activation. The assay is sensitive and makes use of
4
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
62
86
multi-well plates and automated reagent handling systems, allowing high throughput quantification of
87
cellular apoptosis due to toxin.
88
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
5
MATERIALS AND METHODS
90
Bacterial Toxins: Native C. difficle toxins, VPI ribotype 087 (TcdA from VPI10463, product number 01A01
91
and TcdB from VPI10463, product number 01A02) were purchased from tgcBIOMICS GmbH (Mainz,
92
Germany) and stored at 4oC lyophilized and -70oC after reconstitution in pyrogen-free sterile water, with
93
limited (≤3 times) freeze-thawing. The TcdA and TcdB toxins were mutated as described in the
94
literature. These mutations included W101A, D287A and W519A for TcdA and W102A, D288A and
95
W520Afor TcdB (23-25). These point mutations are demonstrated to destroy the enzyme activity
96
(glucosylase) and substrate binding of the toxin. The muted toxins were expressed in Baculovirus and
97
purified from cell lysates using Ceramic Hydroxyapatite Type II (Biorad, Hercules, CA) column
98
chromatography, and stored at -70oC. Recombinant, his-tagged C. difficile binary CDTa-CDTb toxins
99
were expressed in E.coli and purified from E.coli lysates by affinity chromatography, then buffer
100
exchanged into 50mM Hepes, pH 7.5, 150mM NaCL, and stored at -70oC. Pseudomonas aeruginosa
101
Exotoxin A (PEA) and Corynebacterium diphtheria toxin (DT) were purchased from EMD Millipore
102
Corporation (Billerica, MA), re-constituted in sterile water to 1mg/ml, and stored at 4oC according to
103
vendor’s recommendation. CRM197, the mutated form of DT, produced from a single missense
104
mutation (Gly52 to Glu) within the fragment A region(26-28), was purchased from MBL International
105
Corporation (Woburn, MA), in solution of 1mg/ml, and stored at -70oC. All toxins had ≥90% purity as
106
indicated by accompanying literature of purchased toxins, and SDS analysis of internally produced
107
material (data not shown).
108
Cell lines and cell culture: Vero cells (African green monkey kidney) were obtained from the American
109
Type Culture Collection (ATCC, Manassas, VA) and cultured in Dulbecco’s minimum essential medium
110
(DMEM) supplemented with 10% heat-inactivated fetal bovine sera (FBS) per ATCC procedures. HeLa
6
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
89
cells (human carcinoma epithelial) were obtained from ATCC and cultured in Eagle’s minimum essential
112
medium (EMEM) supplemented with 10% heat inactivated FBS per ATCC instructions.
113
Caspase inhibition reagents: Z-VAD-FMK, a cell-permeant pan caspase inhibitor that irreversibly binds to
114
the catalytic site of caspase proteases and inhibits induction of apoptosis (Promega, Madison, WI), was
115
re-constituted in DMSO at 20mM, stored at -20oC, and used in the assay at a final concentration of
116
20µM for inhibition of caspase, per vendor’s recommendation. Sera from hamsters immunized with
117
inactivated TcdA and TcdB toxins, as previously described (29) were used for C. difficile toxin
118
neutralization assays.
119
Caspase 3/7 Assay Reagent: The Caspase-Glo 3/7 Assay reagent (Promega, Madison, WI) was used for
120
caspase detection in treated cells in vitro. The reagent provides a proluminescent caspase-3/7
121
substrate, which contains the tetrapeptide sequence DEVD, in combination with luciferase and a cell
122
lysing agent. The addition of the Caspase-Glo® 3/7 reagent, directly, to the assay well results in cell lysis,
123
followed by caspase cleavage of the DEVD substrate, and generation of luminescence. The resulting
124
luminescence read out is proportional to the amount of caspase activity in the sample.
125
Caspase Assay optimization and Toxin evaluation: The caspase assay was optimized for several
126
parameters which included toxin concentrations, cell seeding density, and sera-toxin pre-incubation
127
time for C. difficile toxins TcdA, TcdB, binary CDTa-CDTb, and DT and PA. Vero cells were used to assess
128
caspase activation by TcdA, TcdB, and binary CDTa-CDTb. HeLa cells were used as target cells for DT and
129
PA. Varying cell input experiments (data not shown) lead to the selection of a seeding density of
130
2.5X104cells/well in a total volume of 100µl/well of black, glass bottom (Thermo Scientific, Rochester,
131
NY) 96-well tissue culture plates, or 50µl cell suspension at 3X104 cells/ml in 384-well plate. Cells were
132
incubated at 37oC, 5% CO2, and allowed to grow to ~90% confluence. The following day, cell
133
supernatants were replaced by toxins diluted in tissue culture medium at a different range of
7
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
111
concentrations for each toxin (as shown in Figure 1), and incubated overnight at 37oC, 5% CO2. Caspase-
135
Glo reagent was added directly to individual wells at 30μl/well in 96-well plates (15 μl/well in 384-well
136
plates), mixed gently on an orbital shaker for about 30 seconds, and further incubated at 37oC, 5% CO2
137
for 30-60 minutes prior to reading. Luminescence measurements were obtained using Perkin Elmer’s
138
2030 Multilabel Reader Victor 4X plate reader. The toxin concentrations used for the time course
139
studies were EC90 and were 20 ng/ml (TcdA), 80pg/ml (TcdB), 1µg/ml (DT) and 10µg/ml PAE. In these
140
time course experiments, the Caspase –Glo reagent was added as described above at various times after
141
intoxication, depending on toxin and cells being evaluated, ranging from 2-48 hours, to determine
142
optimum time for toxin exposure and caspase activation. To demonstrate the specificity of the toxin-
143
induced caspase activation, the genetically modified toxins were tested alongside the corresponding
144
active toxins at the same concentrations, or the pan-caspase inhibitor (Z-Val-ALa-Asp- fmk [Z-VAD-fmk])
145
was added to the active toxin wells at 20µM.
146
F-actin polymerization assay: Detection of F-actin polymerization was employed as a means to
147
correlate results to the Caspase activation assay data. The F-actin assay was previously described in
148
detail by Xie et al (29). Briefly, Vero cell suspensions were added to 384-well plates, and incubated
149
overnight at 37oC, 5% CO2. Following incubation, supernatants were replaced by toxin pre-incubated
150
with or without neutralizing serum and plates returned to the incubator for an additional 48 hours (see
151
below for toxin concentrations used in our assay comparisons). Wells were stained following a series of
152
centrifugations, washes, and incubations, as described in Xie et al (29) to enable detection and imaging
153
of F-actin polymerization in the treated cells using a scanning cytometer (Molecular Device, Sunnyvale,
154
CA).
155
Toxin neutralization assays: EC90 concentrations for TcdA and TcdB, (4ng/ml TcdA and 40pg/ml TcdB)
156
were used for demonstrating neutralization of toxins by pre-incubation of toxins with sera from animals
8
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
134
previously hyper-immunized with toxins as described earlier (29). For both caspase and F-actin
158
polymerization assays, Vero cells were seeded at 3X104cells/ml in 50μl/well of a 384-well plate, and
159
incubated at 37oC, 5% CO2 for 24hr. The following day, supernatants were replaced by pre-incubated
160
toxins with 1:2 serially diluted immune sera, and cells were further incubated for 24hr for the caspase
161
assay, or 48hr for the F-actin assay. The cytometric F-actin neutralization assay was conducted over a
162
period of four consecutive days, and the data acquisition for the assay involves an image of the
163
monolayer acquired using the scanning cytometer. The caspase assay was conducted over a period of
164
three consecutive days, and data acquisition involves the luminescence measurements obtained from
165
the addition of caspase substrate, using a luminescence–capable plate reader.
166
The EC50 value was calculated by four parameter regression fitting of the titration curve using GraphPad
167
Prism 5 for Windows computer software, version 5.04, for both the Caspase-Glo and the cytometric
168
assays.
169 170 171
9
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
157
RESULTS
173
Toxin-induced caspase activation over a range of toxin concentrations was demonstrated in Vero cells
174
for C. difficile toxins (TcdA, TcdB, and binary CDTa and CDTb), and in HeLa cells for C. diphtheria and P.
175
aeruginosa toxins as shown in Figure 1. The average ED50 of 5 replicate wells across 2 assays for TcdA
176
and TcdB were calculated to be approximately 1.26ng/mL and 12.5pg/mL respectively. The average ED50
177
for binary CDTa-CDTb was 3.0ng/ml. The average ED50 of 5 replicate wells across 2 assays for DT and PA
178
were found to be approximately, 0.0026µg/ml and 0.42µg/ml, respectively (Figure 1).
179
To further optimize the assay we determined the time course of caspase activation upon toxin exposure.
180
Appropriate cell monolayers (seeded at 2.5X104cells/well, 24 hours prior) were treated with individual
181
toxin at EC90 concentrations shown to produce high caspase activation signal (80pg/ml of TcdB, 20ng/ml
182
of TcdA, 1µg/ml of DT and 10µg/ml PA). Caspase levels were assessed at various time intervals ranging
183
from 2-48 hours post intoxication.
184
incubation times (2, 4, 5, 7h for DT and PA, and 4, 8, 16 h for TcdA and TcdB, data not shown). Results
185
showed that levels of caspase activation increased in a time-dependent manner and the kinetics varied
186
for each toxin (Figure 2).
187
decreasing at 28hr, and considerable loss of activity by 48hr. DT-induced caspase activation was seen as
188
early as 4hr, with peak levels at 6hr, and decreasing thereafter with a significant drop of activity by 16hr.
189
PEA-induced caspase activation peaked around 10hr, noticeably decreased by 14hr, followed by almost
190
a complete loss of activity at 24hr.
191
To test the specificity of the toxins’ ability to induce caspase, we assayed genetically inactivated toxins
192
(Figure 3). Toxin mutagenesis has been described in peer-reviewed literature as a means to decrease or
193
inactivate bacterial toxins, and thus enable their use in development of vaccines. Previous studies have
194
shown that, introduction of specific mutations in bacterial toxin genes lead to inactivation or decrease of
Preliminary time courses were conducted at fewer, shorter
TcdA and TcdB showed significant increase at 16hr, peaking at 24hr,
10
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
172
195
toxicity (23, 27, 30, 31). To test the specificity of the toxins ability to induce caspase, we assayed
196
genetically altered toxins.
197
corresponding active toxins, and none were assayed for nuclease activity. Genetically inactivated TcdA
198
and TcdB were unable to induce caspase activity at concentrations where the wild type toxins clearly
199
demonstrated caspase induction (Figure 3A, 3B).
200
mutant DT (Figure 3C). Furthermore, we were able to inhibit the caspase signal induced by the four wild
201
type toxins in the assay by the addition of a caspase inhibitor at all concentrations of toxin tested (Figure
202
3A-D). These data demonstrate the specificity of caspase induction by individual toxins.
203
We next compared the caspase assay performance with an established cytometric method (29) to
204
evaluate cellular toxicity caused by the toxins. Both assays were conducted to assess the ability of
205
serum antibodies induced by immunization with TcdA and TcdB toxoids to neutralize the pro-apoptotic
206
activity of the wild type toxins.
207
hyperimmune hamster sera (as described in Materials and Methods section) protected the cells from
208
toxin-induced cellular toxicity, as measured by inhibition of caspase activation (Figure 4A), and inhibition
209
of cell rounding via F-actin depolymerization (data not shown). The assays were conducted head to
210
head using identical conditions for both in terms of cell numbers, toxin concentrations, serum dilutions
211
and pre-incubation of toxins with immune-sera. Following incubation of sera and toxins on the cells,
212
development of the plates and data acquisition were conducted using parameters optimized for each
213
individual assay. The study evaluated toxin neutralization from sera collected from 20 individual animals
214
previously immunized with inactivated toxins.
215
concentration previously determined (29) to cause 90% of cytotoxicity in the well. Comparable
216
neutralization titers were obtained by both methods with R2 value of 0.807 for anti-TcdA titers and
217
0.904 for anti-TcdB titers (Figure 4B).
All mutant-toxins were evaluated using the same titrations as their
Similar data were also obtained with CRM197, a
Both methods used 384 well plates, and toxin
11
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
Pre-incubation of TcdA or TcdB toxins with serial dilutions of
DISCUSSION:
219
Bacterial toxins are attractive targets for drug and vaccine development. Hence, several
220
methods have been reported to assess their functional activity on eukaryotic cells in vitro.
221
Moreover, functional assays are increasingly used in determining the type of immune response
222
induced by toxins, quality control of toxins as vaccine antigens, and as potency release tests in
223
toxin-based vaccine development. This has led to an increased demand for high throughput cell
224
based assays with decreased cycle times. Several assays reported in the literature rely on visual
225
observations of cytopathic effects of toxins e.g. rounding of cells or loss of viability (5, 7, 9, 19-
226
22). Recently, an elegant high throughput assay has been reported that can quantitate actin
227
polymerization to measure the effects of C. difficile toxins TcdA and TcdB on Vero cells(29).
228
Other alternative methods available for quantifying toxins biological (enzymatic) activity within
229
a cell are often cumbersome, low throughput, and require costly instruments for acquiring
230
measurements related to the effects of toxin on eukaryotic cells. There are several reports in the
231
literature linking C. difficile toxins to apoptosis and activation of caspases. Guerrant et al have
232
shown caspase 3 and 9 induction by western blotting in human intestinal epithelial cell line T84
233
with TcdA exposure (32). Similar data has been reported for TcdB in HeLa cells by Qa’dan et al
234
(10). Moreover, Hippensteil et al have shown that TcdA and TcdB mediated rho-inactivation
235
leads to caspase induction and apoptosis(33). Since other bacterial toxins are known to cause
236
death of the target cells via apoptosis, we postulated that measuring caspase activation could
237
not only be used to detect the cytotoxic effects of a majority of bacterial toxins, but also
238
developed into a simple luminescent assay allowing for an earlier readout and higher
239
throughput compared to standard cell-based assays which rely on quantifying morphological
240
changes.
12
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
218
The cell lines were chosen for their sensitivity to the respective toxin and for historical reasons.
242
Vero cells have been shown to be very sensitive to C. difficile toxins and have been used in the
243
published F-actin polymerization assay (19, 29, 33, 34). Since we were comparing the caspase
244
assay to the F-actin polymerization assay, Vero cells were selected for our caspase studies with
245
TcdA and TcdB. Similarly, diphtheria toxin and Pseudomonas enterotoxins were historically
246
tested on HeLa cells in our institution, and the use of HeLa cells for studying caspase activation
247
and thus, apoptosis, have been reported in the published literature (12, 27, 35, 36).
248
Additionally, these cell lines provided the sensitivity to carry out the assay rapidly and reliably,
249
leading to their selection as preferred cell substrates. Ultimately, the choice of cell lines will
250
depend on the individual toxin source and user needs. We were able to observe caspase
251
induction by TcdA, TcdB, and CDTa-CDTb in Vero cells, and by DT and PEA in HeLa cells.
252
Interestingly, the kinetics of the caspase induction varied with the type of toxins used. DT was
253
able to demonstrate peak caspase induction upon intoxication within 6h, followed by PEA at
254
10h. TcdA and TcdB were found to have slower kinetics (peak around 20h) of caspase induction.
255
This differential kinetics could reflect on their modes of action. That is, DT and PEA act on
256
elongation factor thus, shutting down protein synthesis, producing early induction of caspases.
257
In contrast, C. difficile toxins are known to interfere with actin polymerization, which results in
258
detachment and rounding, and may require more time to induce cell death. Hence, it was
259
notably important to determine optimal time for caspase detection after exposure to individual
260
toxins.
261
We further used a pan-caspase inhibitor Z-VAD-fmk to ascertain the specificity of the signals
262
induced upon toxin exposure. The pan-caspase inhibitor was able to inhibit the caspase activity
263
at all concentrations of toxins used. Furthermore, the cells appeared healthy upon visualization
264
suggesting that the inhibitor was able to prevent cell death induced by TcdA, TcdB, PEA and DT
13
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
241
toxins (data not shown). We also determined the mutant toxin’s ability to induce caspase.
266
Mutations that abolished the ADP-ribosylation activity of DT resulted in toxin that was incapable
267
of inducing caspase(28). Similarly, mutations that lead to loss of glucosytransferase activity of
268
TcdA and TcdB resulted in toxins that failed to induce caspase activity (24, 34, 37-39). The data
269
strongly suggests biological/enzymatic activity of the toxins is primarily responsible for caspase
270
induction which results in loss of viability and cell death. As demonstrated by our studies and
271
results, the assay allows for differentiation of specific toxins at given concentrations, and at
272
different levels of biological activity.
273
Natural infection with Clostridium difficile or immunization with toxins TcdA and TcdB results in
274
antibody responses capable of neutralizing the toxins in vitro, which have shown to be
275
protective against disease in vivo (40). Therefore, we evaluated the ability of the caspase assay
276
to quantify the neutralization titers generated in hamsters upon immunization with toxin. We
277
observed that caspase activation was inhibited by hyperimmune sera raised against TcdA and
278
TcdB, demonstrating specificity of the assay to reliably detect caspase signal due to bacterial
279
toxins and to detect neutralization of toxin by specific antibodies. Furthermore, we compared
280
the neutralizing antibody titers generated in the animal study as measured by caspase assay
281
with those measured in the established neutralizing antibody assay (actin polymerization
282
cytometric assay described earlier), and observed excellent correlation between the two assays
283
with shorter cycle time of 3 days for caspase assay.
284
In summary, we present a versatile platform of quantifying the cytotoxic activity of several
285
common bacterial toxins via measurement of caspase activation as an early indicator for cell
286
death. The signal obtained for caspase activity and therefore, apoptosis, is directly proportional
287
to toxin dose, and can be specifically blocked by anti-toxin antibodies, at a magnitude relative to
14
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
265
288
antibody titer. The assay can be performed in high throughput format. It is a fast, efficient, and
289
reliable method for evaluating cell apoptosis in response to bacterial toxins, as well as a means
290
to measure neutralizing antibody titers. This assay provides a valuable tool in the study of toxin-
291
based vaccine candidates, their efficacy in clinical trials, and in establishing immune correlate to
292
protective antibody levels for bacterial toxins causing disease.
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
293
15
294 295
ACKNOWLEDGEMENTS
296 The authors would like to recognize the contributions of Rachel Xoconostle for the purification
298
efforts of C. difficile recombinant toxins. We are also grateful to Andy Xie, Tony Kanavage and
299
Suzanne Cole for helpful discussions and Joe Joyce and Jon Heinrichs for critical reading of the
300
manuscript. All authors are current employees of Merck and co. Inc., and may own stocks for the
301
company
302
16
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
297
Figure 1
306
Dose response of in vitro toxin-induced caspase activity:(A) C. difficile TcdA on Vero cells, (B) C. difficile
307
TcdB on Vero cells, (C) C. difficile binary toxin CDTa-CDTb on Vero cells, (D) C. diphtheria toxin on HeLa
308
cells and (E) P. aeruginosa Exotoxin A on HeLa cells. Caspase induction was measured in toxin-treated
309
cell cultures using Caspase-Glo 3/7 Assay kit (Promega) and results shown as mean luminescence (RLU)
310
plus standard deviation for 5 replicate wells across 2 assays. EC50 (50% effective toxin concentration)
311
was calculated by four-parameter logistic regression of the titration curve.
312 313
Figure 2
314
Time course of in vitro toxin-induced caspase activity: (A) C. difficile TcdA at 20 ng/ml, (B) C. difficile TcdB
315
at 80 pg/ml, (C) C. diphtheria toxin at 1 μg/ml and (D) P. aeruginosa Exotoxin A at 10 μg/ml. Caspase
316
induction was measured in toxin-treated cell cultures using Caspase-Glo 3/7 Assay kit (Promega) and
317
results shown as mean luminescence (RLU) for duplicate wells.
318 319
Figure 3
320
Specificity of toxin-induced caspase activity in vitro was demonstrated by addition of 20 μM caspase
321
inhibitor Z-VAD-FMK (open circles) or genetic inactivation of toxin (open triangles) compared to active
322
toxins (closed symbols) for (A) C. difficile TcdA, (B) C. difficile TcdB , (C) C. diphtheria toxin, and (D) P.
323
aeruginosa Exotoxin A . Genetically inactivated C. difficile toxins were produced from point mutations in
324
the enzymatic domains of TcdA and TcdB. Inactive C. diphtheria toxin CRM197 was produced by a single
325
missense mutation (Gly52 to Glu) within the fragment A region. Caspase induction was measured in
326
toxin-treated cell cultures using Caspase-Glo 3/7 Assay kit (Promega) and results shown as mean
327
luminescence (RLU) plus standard deviation for 3 replicate wells.
328 329 330
17
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
303 304 305
Figure 4
332
Neutralization of C. difficile toxin-induced caspase activity in vitro by pre-incubation of toxin with
333
hamster serum containing anti-toxin neutralizing antibodies: Titration of sera from a subset of four
334
hamsters vaccinated with inactivated TcdA (A) and TcdB (B) showed dose-dependent inhibition of
335
caspase induction by both toxins. Comparison of TcdA (C) and TcdB (D) neutralizing antibody titers in
336
sera from 20 vaccinated hamsters determined by the caspase assay (Y axis) correlates well with titers
337
determined in the F-actin cytometric assay (X-axis) for both toxins. Results are shown as antibody titers
338
determined from inverse of antibody dilution that achieved 50% inhibition of toxin activity at an EC90
339
dose, calculated using a four-parameter logistical fit of each serum titration curve. Caspase induction
340
was measured in toxin-treated cell cultures using Caspase-Glo 3/7 Assay kit (Promega) and F-actin was
341
measured by a scanning cytometer referenced in Methods. Both assays were conducted once each at
342
the same time with the same cell cultures, toxins and sera.
343 344 345 346 347
18
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
331
348 References
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
1. 2. 3. 4. 5.
6. 7. 8.
9.
10.
11. 12.
13. 14.
15. 16. 17. 18. 19.
20.
Rappuoli, R., M. Pizza, G. Douce, and G. Dougan. 1996. New vaccines against bacterial toxins. Adv Exp Med Biol 397:55. Schmitt, C. K., K. C. Meysick, and A. D. O'Brien. 1999. Bacterial toxins: friends or foes? Emerg Infect Dis 5:224. Dorner, F., and J. L. McDonel. 1985. Bacterial toxin vaccines. Vaccine 3:94. Weinrauch, Y., and A. Zychlinsky. 1999. The induction of apoptosis by bacterial pathogens. Annu Rev Microbiol 53:155. Morimoto, H., and B. Bonavida. 1992. Diphtheria toxin- and Pseudomonas A toxin-mediated apoptosis. ADP ribosylation of elongation factor-2 is required for DNA fragmentation and cell lysis and synergy with tumor necrosis factor-alpha. J Immunol 149:2089. Burnette, W. N. 1997. Bacterial ADP-ribosylating toxins: form, function, and recombinant vaccine development. Behring Inst Mitt:434. Keppler-Hafkemeyer, A., R. J. Kreitman, and I. Pastan. 2000. Apoptosis induced by immunotoxins used in the treatment of hematologic malignancies. Int J Cancer 87:86. Jenkins, C. E., A. Swiatoniowski, A. C. Issekutz, and T. J. Lin. 2004. Pseudomonas aeruginosa exotoxin A induces human mast cell apoptosis by a caspase-8 and -3-dependent mechanism. J Biol Chem 279:37201. Castex, F., G. Corthier, S. Jouvert, G. W. Elmer, F. Lucas, and M. Bastide. 1990. Prevention of Clostridium difficile-induced experimental pseudomembranous colitis by Saccharomyces boulardii: a scanning electron microscopic and microbiological study. J Gen Microbiol 136:1085. Qa'Dan, M., M. Ramsey, J. Daniel, L. M. Spyres, B. Safiejko-Mroczka, W. Ortiz-Leduc, and J. D. Ballard. 2002. Clostridium difficile toxin B activates dual caspase-dependent and caspaseindependent apoptosis in intoxicated cells. Cell Microbiol 4:425. Voth, D. E., and J. D. Ballard. 2005. Clostridium difficile toxins: mechanism of action and role in disease. Clin Microbiol Rev 18:247. Imre, G., J. Heering, A. N. Takeda, M. Husmann, B. Thiede, D. M. zu Heringdorf, D. R. Green, F. G. van der Goot, B. Sinha, V. Dotsch, and K. Rajalingam. 2012. Caspase-2 is an initiator caspase responsible for pore-forming toxin-mediated apoptosis. Embo J 31:2615. Kochi, S. K., and R. J. Collier. 1993. DNA fragmentation and cytolysis in U937 cells treated with diphtheria toxin or other inhibitors of protein synthesis. Exp Cell Res 208:296. Bossy-Wetzel, E., D. D. Newmeyer, and D. R. Green. 1998. Mitochondrial cytochrome c release in apoptosis occurs upstream of DEVD-specific caspase activation and independently of mitochondrial transmembrane depolarization. Embo J 17:37. Wyllie, A. H., J. F. Kerr, and A. R. Currie. 1980. Cell death: the significance of apoptosis. Int Rev Cytol 68:251. Li, J., and J. Yuan. 2008. Caspases in apoptosis and beyond. Oncogene 27:6194. Alnemri, E. S., D. J. Livingston, D. W. Nicholson, G. Salvesen, N. A. Thornberry, W. W. Wong, and J. Yuan. 1996. Human ICE/CED-3 protease nomenclature. Cell 87:171. Salvesen, G. S., and S. J. Riedl. 2008. Caspase mechanisms. Adv Exp Med Biol 615:13. Sundriyal, A., A. K. Roberts, R. Ling, J. McGlashan, C. C. Shone, and K. R. Acharya. 2010. Expression, purification and cell cytotoxicity of actin-modifying binary toxin from Clostridium difficile. Protein Expr Purif 74:42. Brito, G. A., J. Fujji, B. A. Carneiro-Filho, A. A. Lima, T. Obrig, and R. L. Guerrant. 2002. Mechanism of Clostridium difficile toxin A-induced apoptosis in T84 cells. J Infect Dis 186:1438.
19
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
349
21. 22.
23.
24.
25. 26.
27.
28. 29.
30.
31. 32.
33.
34.
35. 36. 37.
Sullivan, N. M., S. Pellett, and T. D. Wilkins. 1982. Purification and characterization of toxins A and B of Clostridium difficile. Infect Immun 35:1032. Brito, G. A., B. Carneiro-Filho, R. B. Oria, R. V. Destura, A. A. Lima, and R. L. Guerrant. 2005. Clostridium difficile toxin A induces intestinal epithelial cell apoptosis and damage: role of Gln and Ala-Gln in toxin A effects. Dig Dis Sci 50:1271. Wang, H., X. Sun, Y. Zhang, S. Li, K. Chen, L. Shi, W. Nie, R. Kumar, S. Tzipori, J. Wang, T. Savidge, and H. Feng. 2012. A chimeric toxin vaccine protects against primary and recurrent Clostridium difficile infection. Infect Immun 80:2678. Teichert, M., H. Tatge, J. Schoentaube, I. Just, and R. Gerhard. 2006. Application of mutated Clostridium difficile toxin A for determination of glucosyltransferase-dependent effects. Infect Immun 74:6006. Jank, T., T. Giesemann, and K. Aktories. 2007. Clostridium difficile glucosyltransferase toxin Bessential amino acids for substrate binding. J Biol Chem 282:35222. Mitamura, T., T. Umata, F. Nakano, Y. Shishido, T. Toyoda, A. Itai, H. Kimura, and E. Mekada. 1997. Structure-function analysis of the diphtheria toxin receptor toxin binding site by sitedirected mutagenesis. J Biol Chem 272:27084. Kageyama, T., M. Ohishi, S. Miyamoto, H. Mizushima, R. Iwamoto, and E. Mekada. 2007. Diphtheria toxin mutant CRM197 possesses weak EF2-ADP-ribosyl activity that potentiates its anti-tumorigenic activity. J Biochem 142:95. Giannini, G., R. Rappuoli, and G. Ratti. 1984. The amino-acid sequence of two non-toxic mutants of diphtheria toxin: CRM45 and CRM197. Nucleic Acids Res 12:4063. Xie, J., J. Zorman, L. Indrawati, M. Horton, K. Soring, J. M. Antonello, Y. Zhang, S. Secore, M. Miezeiewski, S. Wang, A. D. Kanavage, J. M. Skinner, I. Rogers, J. L. Bodmer, and J. H. Heinrichs. 2013. Development and optimization of a novel assay to measure neutralizing antibodies against Clostridium difficile toxins. Clin Vaccine Immunol 20:517. Sun, X., X. He, S. Tzipori, R. Gerhard, and H. Feng. 2009. Essential role of the glucosyltransferase activity in Clostridium difficile toxin-induced secretion of TNF-alpha by macrophages. Microb Pathog 46:298. Rappuoli, R. 1994. Toxin inactivation and antigen stabilization: two different uses of formaldehyde. Vaccine 12:579. Carneiro, B. A., J. Fujii, G. A. Brito, C. Alcantara, R. B. Oria, A. A. Lima, T. Obrig, and R. L. Guerrant. 2006. Caspase and bid involvement in Clostridium difficile toxin A-induced apoptosis and modulation of toxin A effects by glutamine and alanyl-glutamine in vivo and in vitro. Infect Immun 74:81. Hippenstiel, S., B. Schmeck, P. D. N'Guessan, J. Seybold, M. Krull, K. Preissner, C. V. EichelStreiber, and N. Suttorp. 2002. Rho protein inactivation induced apoptosis of cultured human endothelial cells. Am J Physiol Lung Cell Mol Physiol 283:L830. Tian, J. H., S. R. Fuhrmann, S. Kluepfel-Stahl, R. J. Carman, L. Ellingsworth, and D. C. Flyer. 2012. A novel fusion protein containing the receptor binding domains of C. difficile toxin A and toxin B elicits protective immunity against lethal toxin and spore challenge in preclinical efficacy models. Vaccine 30:4249. Ogura, K., K. Yahiro, H. Tsutsuki, S. Nagasawa, S. Yamasaki, J. Moss, and M. Noda. 2011. Characterization of Cholix toxin-induced apoptosis in HeLa cells. J Biol Chem 286:37207. Venter, B. R., and N. O. Kaplan. 1976. Diphtheria toxin effects on human cells in tissue culture. Cancer Res 36:4590. Spyres, L. M., J. Daniel, A. Hensley, M. Qa'Dan, W. Ortiz-Leduc, and J. D. Ballard. 2003. Mutational analysis of the enzymatic domain of Clostridium difficile toxin B reveals novel inhibitors of the wild-type toxin. Infect Immun 71:3294.
20
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
442 443 444 445 446 447 448 449
38.
39. 40.
Busch, C., F. Hofmann, J. Selzer, S. Munro, D. Jeckel, and K. Aktories. 1998. A common motif of eukaryotic glycosyltransferases is essential for the enzyme activity of large clostridial cytotoxins. J Biol Chem 273:19566. Lanis, J. M., S. Barua, and J. D. Ballard. 2010. Variations in TcdB activity and the hypervirulence of emerging strains of Clostridium difficile. PLoS Pathog 6:e1001061. Giannasca, P. J., and M. Warny. 2004. Active and passive immunization against Clostridium difficile diarrhea and colitis. Vaccine 22:848.
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
21
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest
Downloaded from http://jcm.asm.org/ on October 5, 2017 by guest