Caspase Activation as a Versatile Assay Platform for Detection of ...

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


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


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


62

86

multi-well plates and automated reagent handling systems, allowing high throughput quantification of

87

cellular apoptosis due to toxin.

88


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


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


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


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


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


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


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


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


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


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.


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


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


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


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


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


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.


21



