This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Molecular & Biochemical Parasitology 175 (2011) 1–9

Contents lists available at ScienceDirect

Molecular & Biochemical Parasitology

Eimeria bovis-induced modulation of the host cell proteomeat the meront I stage

Kathleen Lutza, Sigrid Schmittb, Monica Linderb, Carlos Hermosillac,∗,Horst Zahnera, Anja Tauberta

a Institute of Parasitology, Justus Liebig University Giessen, 35392 Giessen, Germanyb Institute of Biochemistry, Justus Liebig University Giessen, 35392 Giessen, Germanyc Department of Pathology and Infectious Diseases, Royal Veterinary College, Hawkshead Lane, North Mymms, Hatfield, Herts AL9 7TA, UK

a r t i c l e i n f o

Article history:Received 12 December 2009Received in revised form 25 July 2010Accepted 20 August 2010Available online 27 August 2010

Keywords:Eimeria bovisApicomplexaProteomeFirst merogonyHost cell

a b s t r a c t

The proteome of Eimeria bovis meront I-carrying host cells was analyzed by two-dimensional gel elec-trophoresis (2DE) at 14 days p.i. and compared to non-infected control cells. A total of 221 proteinspots were modulated in their abundance in E. bovis-infected host cells and were subsequently ana-lyzed by matrix-assisted laser desorption ionization time-of-flight mass spectometry (MALDI-TOF-MS).These analyses identified 104 proteins in total with 25 host cell proteins being up-regulated and 79proteins being down-regulated in E. bovis-infected host cells. Moreover, 20 newly expressed proteinswere identified exclusively in E. bovis-infected host cells and were most likely of parasite origin. Parasite-induced differences in protein abundance concerned distinct functional categories, with most proteinsbeing involved in host cell metabolism, cell structure, protein fate and gene transcription. Some of themodulated molecules also indicated regulatory processes on the level of host cell stress response (HSP70,HSP90), host cell apoptosis (caspase 8) and actin elongation/depolymerization (�-actinin-1, gelsonin,tropomodulin-3, transgelin). Since merozoites I were already released shortly after cell sampling, thecurrent data reflect the situation at the end of first merogony. This is the first proteomic approach on E.bovis-infected host cells that was undertaken to gain a rather broad insight into Eimeria-induced hostcell modulation. The data processed in this investigation should provide a useful basis for more detailedanalyses concerning Eimeria–host cell interactions.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Eimeria bovis is an obligate intracellular apicomplexan par-asite which represents one of the most pathogenic species incattle coccidiosis, causing severe haemorrhagic typhlocolitis incalves and high economic losses worldwide [1]. Within the gut,freely released E. bovis sporozoites traverse the epithelium inorder to invade endothelial cells of the central lymph capillar-ies of the ileal villi, where they undergo the first merogony [2].Enclosed by a parasitophorous vacuole (PV), the sporozoites of E.bovis develop into ≥300 �m sized macromeronts within 14–18days [3,4]. It appears likely that once the parasite begins growthand proliferation it must acquire nutrients from the host cellas reported for other intracellular apicomplexans [5,6]. Further-more, the massive replication and the final enlargement of thehost cell far beyond the physiological size may cause consider-able stress to the host cell [7,8], and cell stress, in turn, is well

∗ Corresponding author. Tel.: +44 1707666040; fax: +44 1707667052.E-mail address: chermosilla@rvc.ac.uk (C. Hermosilla).

known to trigger host cell defense mechanisms and apoptosis [9].In consequence, the obligate intracellular parasite E. bovis mustrely on several regulatory processes to actively modulate the hostcell proteome to guarantee survival and first merogony develop-ment.

The general capacity of apicomplexan parasites to manipulateinfected host cells for their advantage is well-documented in caseof Toxoplasma gondii. This particular parasite acquires auxotrophnutrients from infected host cells, interacts with the host cellmitochrondria, endoplasmic reticulum and cytoskeleton [10–12]and actively interferes with the apoptotic capacity of the host cell[10,13–15].

In the case of E. bovis only few reports deal with its capacityfor host cell modulation, although its merogony takes much longerand results in much larger intracellular meronts than T. gondii.Some evidence is reported on the level of distinct modulation ofhost cell cytoskeleton [16] and inhibition of host cell apoptosis byenhanced expression of anti-apoptotic factors [17], the latter ofwhich is in agreement with other Eimeria spp. [18]. Activation ofNF�B in sporozoite-infected, non-permissive epithelial host cellswas shown for E. bovis as well [19].

0166-6851/$ – see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.molbiopara.2010.08.003

Author's personal copy

2 K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9

E. bovis also appears to modulate host endothelial cell-mediatedimmune reactions [20–23]. At least generally transcript levels ofgenes encoding for pro-inflammatory molecules in E. bovis-infectedendothelial cells were significantly lower when compared withT. gondii- or N. caninum- infected cells [22].

The aim of this study was to gain a broad insight into E. bovis-induced host cell proteome modulation at the meront I stage.Therefore, whole cell proteomes of E. bovis-infected host cells andnon-infected controls were analyzed by two-dimensional gel elec-trophoresis (2DE) followed by the identification of individual spotsby matrix-assisted laser desorption ionization time-of-flight massspectrometry (MALDI-TOF-MS). Overall, 27 host cell proteins werefound up-regulated and 82 proteins down-regulated in host cellscarrying mature E. bovis meronts I, the majority of them beinginvolved in cell metabolism, cell structure, protein synthesis andgene transcription. In addition, we identified 20 proteins in infectedcells which were most probably of parasite origin. With these datawe create a basis for more detailed analyses targeting functional E.bovis–host cell interactions.

2. Materials and methods

2.1. Parasite

The E. bovis strain H used in the present study was isolated inthe field in northern Germany [24] and maintained by passagesin parasite-free Holstein Friesian calves. For oocysts production,calves were infected orally at the age of 10 weeks with 5 × 105

sporulated E. bovis oocysts. Excreted oocysts were isolated from thefaeces beginning 18 days p.i. according to the method of Jackson[25]. Sporulation of oocysts was achieved by incubation in a 2%(w/v) potassium dichromate (Merck) solution at room temperature(RT). Sporulated oocysts were stored in this solution at 4 ◦C untilfurther use.

Sporozoites were excysted from sporulated oocysts as previ-ously described [26]. Free sporozoites were washed three timesin phosphate buffered-solution (PBS), resuspended in completeIscove’s modified Dulbecco medium (IMDM; Gibco) and countedin a Neubauer haemocytometre as described elsewhere [27].

2.2. Host cells and parasite infection

Bovine foetal gastrointestinal cells (BFGC), an immortalized pri-mary cell line originally isolated from in utero foetuses 4–6 monthsafter conception [26], were cultured in complete IMDM mediumsupplemented with 500 U/ml penicillin, 50 �g/ml streptomycin(Sigma), 1% l-glutamine (w/v) and 10% heat-inactivated foetal calfserum [(v/v); FCS; Gibco]. Cells were cultured in 75 cm2 tissue cul-ture flasks (Nunc) and incubated (37 ◦C, 5% CO2 atmosphere) untilconfluency. BFGC monolayers were infected with 5 × 105 freshlyexcysted sporozoites. Sporozoite viability was determined by try-pan blue exclusion test according to Lang et al. [17]. Cell culturemedium was changed 1 day after infection and thereafter everythird day. One-day p.i. infection rates of the BFGC cultures weredetermined microscopically.

2.3. Sample preparation for the two-dimensional gelelectrophoresis (2DE)

Previous transcriptome studies on E. bovis-infected host cellsrevealed most molecules being modulated at 14 days p.i. whencompared to earlier time points of infection. In accordance to theseresults we restricted the current analysis to this time point. At14 days p.i. whole cell preparations of E. bovis-infected and non-infected BFGC were subjected to 2DE. BFGC layers were washedthree times with PBS. Cells were scraped off the tissue culture flasks

using a sterile rubber policeman (Nunc) and resuspended in PBS.The cell suspension was pelleted (600 × g, 15 min) and washed twotimes with sorbitol–buffer [25 mM sorbitol, 10 mM Tris-base (v/v,pH 7); all Sigma]. Cells were lyzed in a lysis buffer [8 M urea, 2 Mthiourea, 4% CHAPS, 30 mM DTT, 20 mM Tris, 2% pharmalyte buffer(v/v, pH 3–10); all Sigma] and stored at −80 ◦C.

For 2DE 375 U of benzoase (Calbiochem) and 8 �l proteaseinhibitor (Sigma) for mammalian cells were added to each sampleand samples were incubated (30 min, RT), sonicated twice (200 W,working time 10 s, interval time 15 s, 50 cycles, ice bath) andcentrifuged (21,000 × g, 30 min, 20 ◦C). The supernatant was pre-cipitated with acetone [1:7; (Merck)] and centrifuged (8510 × g,30 min, 20 ◦C). The resulting pellet was washed in acetone (Sigma)and resuspended in the above-described lysis buffer. Solubilizedproteins were then quantified using the Plus OneTM 2D Quant Kit®

technique (GE Healthcare Life Sciences) and subjected immediatelyto 2DE.

2.4. Iso-electro focusing (IEF) and 2DE

Proteins were solubilized by incubating the washed pellet atleast 1 h at RT in iso-electro focusing rehydration buffer onto immo-bilized pH gradient (IPG) strips [pH 3–11 non-linear; length 11 cm;DryStrip® (GE Healthcare Life Science)]. Each strip was rehydratedovernight with 300 �g of protein sample in 220 �l lysis buffer.The IEF was carried out in a Multiphor chamber® (AmershamBiosciences) at 20 ◦C applying the following conditions: Phase 1:gradient from 0 V to 50 V for 0.25 kVh; Phase 2: gradient from100 V to 3500 V for 11 kVh; Phase 3: 3500 V for 21 kVh. Currentwas limited to 0.25 mA per strip. For the second dimension theprotein-loaded IPG strips were equilibrated for 15 min by rock-ing in DTT-equilibration buffer [10 mg/ml 1,4-dithio-dl-threitol,20 mg/ml sodium dodecyl sulfate (SDS) in 6 M urea (all Fluka), 30%(v/v) glycerin (Sigma), 50 mM Tris–HCl (Merck), pH 8.8, 4% (w/v)SDS 0.01% bromophenol blue (Fluka)] and then washed for 15 minin jodacetamid-solution [40 mg/ml jodacetamid (Fluka), 20 mg/mlSDS in 6 M urea, 30% (v/v) glycerin, 50 mM Tris–HCl, pH 8.8, 4%(w/v) SDS 0.01% bromophenol blue (all Fluka)]. The strips werethen embedded on a precast separating gel [12.5%, v/v acrylamide(Fluka)] and sealed into place with agarose-solution (Fluka). Gelelectrophoresis was carried out at 25 ◦C applying a constant voltageof 500 V for 4.5 h in a 14.5 cm × 14.5 cm vertical gel electrophoresischamber (Hoefer 600®; Amersham Biosciences).

The gels were then stained by colloidal Coomassie (Fluka).Thereafter, the 2DE gels were agitated for 60 min in staining solu-tion [0.2% (w/v) Coomassie R 250, 50% (v/v) MeOH, 50% (v/v)ddH2O] and then overnight in stripping solution [50% (v/v) MeOH,6% (v/v) acetic acid, 44% (v/v) ddH2O]. Analysis of 2DE gels was per-formed using a flatbed scanner (Powerlook 2100 XL®; Umax) witha resolution of 600 dpi. For each sample, four gels were run andanalyzed in parallel.

2.5. Image analysis

Differences in protein abundance between non-infected andE. bovis-infected BFGC samples were detected by image analy-sis using the software Proteomweaver® (Version 4.0; BioRad).To prevent that proteins originating from E. bovis-macromerontsmay falsify the total amount of host cell protein, a consistentlyexpressed “house-keeping” protein was needed for calibration.Since recent studies showed that vimentin intermediate filamentsare not altered in their abundance in E. bovis-infected host cellsduring the first merogony [16] we chose the bovine vimentinspot as internal reference to normalize the 2DE gels of infectedand non-infected BFGC. Therefore five vimentin spots were iden-tified in 2DE gels by immunoblotting (anti-vimentin, clone Vim

Author's personal copy

K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9 3

Fig. 1. E. bovis-infected and non-infected BFGC in vitro. Confluent BFGC layers were either left non-infected (A) or infected with freshly isolated E. bovis sporozoites(5 × 105/75 cm2 culture flask, B1/2) and cultured for 14 days. Note huge macromeronts containing mature merozoites I in infected cell layers (B1/2).

3B4; 1:10,000, Sigma) and subsequent MALDI-TOF-MS sequenceanalysis. For normalization we always referred to one vimentinspot [characteristics: 60 kDa; pI: 4.8; Mowse score: 309; massvalue matched: 35; seq. cov. (%): 63] consistently expressed in allprobes.

The factor “infected/non-infected” was calculated from the spotaverage intensity of four gels and normalized to the vimentin spotaverage intensity of four gels by division.

The criteria for proteins being judged modulated in their abun-dance were the following: ≥2.0-fold up/down-regulation and eachprotein spot had to be detected in at least three out of four 2DEgels made from identical samples. The cut-off of >2-fold changetypically applied in proteome analyses was chosen to reduce exper-imental variation and thereby to emphasize significant effects.Furthermore, spots required an intensity of ≥0.1 and a Student t-

test value of ≥90% to be considered for subsequent MALDI-TOF-MSanalyses.

The apparent molecular mass of the proteins was calculatedusing a standard 10 kDa ladder run in the gels. Additionally, thepI (iso-electro point) of the proteins was determined according tothe pH value in the strips.

Indications of phosphorylation were predicted by databasescreening of peptide spectra of single proteins via MASCOT analysis.

2.6. In-gel digestion and peptide mass fingerprinting

Protein spots of interest were manually excised from the gels,washed in distilled water and twice in 50 mM ammonium hydro-gen carbonate:acetonitrile (1:1) and acetonitrile (Fluka). Gel pieceswere then incubated in 10 ng/�l trypsin solution [sequencing grade

Fig. 2. Two-dimensional gel electrophoresis (2DE) of non-infected (A) and E. bovis-infected (B) BFGC. Confluent BFGC monolayers were infected with freshly isolated E. bovissporozoites (5 × 105/75 cm2 tissue culture flask) and sampled 14 days after infection. Non-infected BFGC were used as controls. For 2DE PAGE, 300 �g of protein were loadedper gel. After electrophoresis the gels were stained with Coomassie and scanned. Differences in protein abundance (represented as an average of the total number of gels)were detected using Proteomweaver software. Only proteins that differed by at least 2-fold in abundance when compared with non-infected controls were subjected tofurther analyses (peptide mass fingerprinting and MALDI-TOF-MS).

Author's personal copy

4 K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9

(Roche Diagnostics) in 25 mM ammonium hydrogen carbonate,16 h, 37 ◦C]. Peptides were extracted with 5 �l of 1% (v/v) triflu-oroacetic acid (Bruker Daltonics) containing 5 mM octylglycoside(Bruker Daltonics). Aliquots (2 �l) of the solution were applied to athin layer of �-cyano-4-hydroxycinnamic acid on an AnchorChip®

target (Bruker Daltonics). After 10 min of incubation the super-natant was removed, and the spot was washed twice with 2 �l0.1% (v/v) trifluoroacetic acid solution (Bruker Daltonics). Massfingerprints of tryptic digests were obtained by MALDI-TOF-MSanalysis using an UltraflexTM TOF/TOF mass spectrometer (BrukerDaltonics). Peptide mass standards (Bruker Daltonics) were usedfor external calibration of the mass spectra. Proteins were iden-tified by database searching (NCBI, SWISS-PROT, TrEMBL) withpeptide masses using the program ‘Mascot in house® (Version 2.1,http://www.matrixscience.com). Protein identification was com-pleted by significance of p < 0.05 probability based values on MowseScores (≥61).

3. Results and discussion

3.1. Comparative proteome analysis by 2D-electrophoresis andMALDI-TOF-MS

The study compares the proteomes of E. bovis-infected and non-infected host cells 14 days after in vitro infection. As expected, invitro culture had resulted in large macromeronts containing eitheralmost mature or fully mature merozoites I (Fig. 1). Shortly aftercell sampling, freely released merozoites I were observed in super-natants of corresponding control samples. The initial infection rateas measured on day one p.i. was 10%.

Overall, comparative 2DE of E. bovis-infected (Fig. 2A) and non-infected (Fig. 2B) BFGC revealed that, in total, 221 protein spotswere modified in their abundance by this protozoan infection with137 protein spots being down-regulated and 84 spots being up-regulated. Amongst the latter group 20 proteins were annotated toparasite origin (data not shown) and for 23 proteins no referencewas found.

Overall, only a proportion of the modulated proteins fulfilled thecriteria for subsequent MALDI-TOF-MS analyses. Consequently, atotal of 104 proteins were further characterized by this method. Themajority of these proteins was down-regulated (n = 79, Table 1 ),whilst only 25 proteins (Table 2) were enhanced in their abundance.

3.2. Functional distribution of differentially expressed proteins

Functional annotation analyses revealed that parasite-induceddifferences in host cell protein abundance concerned differentfunctional categories. In general, it appears noteworthy to men-tion that grouping the different proteins in just one category maybe very simplistic with respect to the multifunctional characteris-tics of many proteins, but had to be performed in order to achieveoverview in general.

The majority of up-regulated proteins was involved inmetabolism (35%), protein fate (19%) and cell structure (15%), whilstdown-regulated proteins were mainly grouped into the functionalcategories of metabolism (20%), cell structure (23%), core proteinsand transcription (17%) (see overview, Fig. 3). Moreover, somedown-regulated proteins were involved in other biological func-tions such as signal transduction (6%), protein fate and translation(both 5%), cellular transport (4%), apoptosis (1%), immune (1%) andstress response (1%).

3.2.1. MetabolismHost cell metabolism was by far the mostly affected functional

category in E. bovis-infected host cells, indicating that the para-site’s development is fundamentally linked to the acquisition of

essential host cell molecules. It appears likely that with the onsetof parasite growth and proliferation E. bovis needs access to host cellnutrients as described for other apicomplexan parasites [11,12,28].Unfortunately, in contrast to T. gondii exhibiting well-documentedauxotrophies for purines [29], hypoxanthine, xanthine and othermolecules [30], no data are available for nutritional requirementsof E. bovis or other Eimeria species.

Overall, modulated molecules were involved in differentmetabolic pathways, such as glycolysis, citric acid cycle or alcoholand lipid degradation. In total, 16 proteins related to metabolismwere down-regulated and only 9 molecules were found to beup-regulated. Interestingly, the total number of modulated pro-teins linked to metabolism corresponded well to those reported inT. gondii-infected host cells [31]. In E. bovis-infected BFGC manydown-regulated proteins play a role in glycolysis and the citricacid cycle, suggesting that this major pathway of energy genera-tion is not or no longer exploited by the parasite. This may relateto the stage of development of the parasite at the time of sam-pling, 14 days p.i. At this point of time most infected cells carriedalready mature macromeronts filled with merozoites I, i.e. para-site replication was completed and it was therefore rather a matterof conservation of a status quo than of high energy expenditure.This is partially in accordance with proteome analyses on T. gondiitachyzoite-infected host cells showing that most of the host cellproteins grouped in metabolic processes were also down-regulated[31]. Admittedly, in contrast to our findings in E. bovis-infectedcells, molecules involved in glycolysis were rather up- than down-regulated in case of T. gondii. It has to be considered that in the lattercase analyses were performed 24 h after infection [31], i.e., duringthe parasite’s proliferative phase – considering the much shortermerogony – and may consequently reflect a different developmen-tal situation.

3.2.2. Cell structureStructural molecules were the second largest group of pro-

teins to be influenced in their abundance by E. bovis infection.Previous studies on E. bovis-triggered modulation of the hostcell cytoskeleton in primary endothelial cells demonstrated thatboth actin filaments and microtubules were increased in theirabundance and reorganized within the host cell with ongoinginfection [16]. Interestingly, and in accordance with findings in T.gondii-infected host cells [31], several proteins known to interactwith actin and involved in the elongation and depolymerizationof actin filaments (alpha-actinin-1, gelsolin, actin-like protein-2,gelsolin-like capping protein, tropomodulin-3, transgelin) werealso down-regulated. However, in contrast to T. gondii-infectedhost cells, myosin-related proteins (myosin-10, �-tropomyosin,myosin regulatory light chain 2) were down-regulated as well.Since tubulin-associated proteins (tubulin beta 5 and 6) were alsodecreased in their abundance, the modulation of these two maincytoskeleton systems suggests a general mechanism taking placeat the final of merogony when macromeronts need to rupture formerozoite I release.

Lamin A/C, an intermediate filament protein of the nuclearmembrane, was also down-regulated in E. bovis-infected BFGC.Lamin filaments represent one of the major constituents of thenuclear membrane and exhibit supportive characteristics for thenuclear double lipid layer forming a fiber network [32]. Lamins arefunctionally regulated by phosphorylation leading to disintegra-tion of the nuclear membrane during mitosis [32,33]. Since several,particularly phosphorylated isoforms of lamin A/C were down-regulated in E. bovis-infected host cells this phenomenon could beassociated with reduced host cell mitotic activity. Actually we havenever observed division of E. bovis-infected cells in vitro and a sim-ilar situation seems to exist in T. gondii-infected cells [31]. In thiscontext, the fact that E. bovis infection down-regulates the abun-

Author's personal copy

K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9 5

Table 1Down-regulated host proteins in E. bovis-infected BFGC (14 days p.i.).

Proteins and functional categories Accession number(SwissProt/Trembl or NCBI)

Factor inf./non-inf. P* Theoret. MW/pI Seq. cov. (%)

MetabolismProcollagen-lysine, 2-oxoglutarate 5-dioxygenase 2 gi|119885261 0.452 88.4/6.42 40Aconitate hydratase P20004 0.342 P 86.0/7.87 48Transketolase Q6B855 0.438 P 68.5/7.56 36Adenylyl cyclase-associated protein 1 Q3SYV4 0.526 P 51.6/7.16 57Alpha-enolase Q9XSJ4 0.377 P 47.6/6.37 53Protein-arginine deiminase type-2 Q9Y2J8 0.537 P 76.3/5.4 23Acyl-CoA thioesterase 9 gi|77736067 0.0 50.2/8.87 53LOC505323 (Ornithine aminotransferase) gi|77735431 0.449 48.4/6.1 35Fructose-1,6-bisphosphate aldolase A gi|119916947 0.0 52.0/9.08 44Isocitrate dehydrogenase 3 (NAD+) alpha gi|109939980 0.420 40.1/6.76 56Glyceraldehyde-3-phosphate dehydrogenase P10096 0.397 36.1/8.5 63l-Lactate dehydrogenase A chain P19858 0.420 P 36.9/8.12 57l-Xylulose reductase Q91X52 0.360 25.9/6.82 27Phosphoglycerate mutase 1 Q3SZ62 0.466 28.9/6.67 83Triosephosphate isomerase Q5E956 0.271 26.9/6.45 77Protein disulfide-isomerase A3 P38657 0.391 57.3/6.23 65

Structure proteinsAlpha-actinin-1 Q3B7N2 0.439 103.5/5.25 63Alpha-actinin-1 Q3B7N2 0.226 P 103.5/5.25 54Myosin-10 Q27991 0.271 229.9/5.44 25Actinin, alpha 4 gi|119910351 0.439 105.3/5.27 54Gelsolin gi|74356373 0.287 P 81/5.54 41Moesin Q2HJ49 0.387 P 68.1/5.9 52Lamin A/C** gi|77404182 0.168–0.456 ±P 65.1/6.54 54–66Lymphocyte cytosolic protein 1 (l-plastin) gi|77736385 0.530 70.7/5.21 44Beta-actin P60712 0.481 42.1/5.29 32Gelsolin-like capping protein gi|30466254 0.272 P 39.1/6.17 58Tropomodulin 3 gi|115496236 0.222 P 38.3/4.92 55Beta-tropomyosin gi|83778524 0.448 33.3/4.62 42PDZ and LIM domain protein 1 Q5E9E1 0.239 P 36.3/6.76 44F-actin capping protein subunit alpha-2 Q5E997 0.541 33.1/5.57 65Transgelin Q9TS87 0.0 P 22.6/8.87 84Myosin regulatory light chain 2 Q5E9E2 0.0 P 19.9/4.67 80Actin-like protein 2 P61160 0.337 45.0/6.3 42Tubulin beta 5 gi|74204140 0.441 50.1/4.75 71Tubulin beta 6 gi|114052148 0.441 50.3/4.76 61

Core proteins/transcriptionMatrin-3 gi|119895555 0.263 100.5/5.71 35Heterogeneous nuclear ribonucleoprotein M P52272 0.000 77.7/8.84 30Heterogeneous nuclear ribonucleoprotein K Q3T0D0 0.348 51.3/5.14 48U4/U6 small nuclear ribonucleoprotein Prp4 Q3MHE2 0.000 59.05/7.06 41LOC513868 (nuclear matrix protein 200) gi|115497226 0.395 P 55.6/6.14 55Histone-binding protein RBBP4 Q3MHL3 0.296 47.9/4.74 31Heterogeneous nuclear ribonucleoprotein D0B isoform 6 gi|76620242 0.432 32.99/8.23 33Heterogeneous nuclear ribonucleoprotein A/B Q99729 0.166 36.7/9.04 34Heterogeneous nuclear ribonucleoprotein C gi|73977325 0.419 27.8/4.55 58Heterogeneous nuclear ribonucleoprotein H3 gi|14141157 0.238 36.96/6.37 64Poly(rC)-binding protein 1 Q5E9A3 0.367 37.98/6.66 74SET (myeloid leukemia-associated) gi|3953617 0.255 P 24.3/4.97 49Heterogeneous nuclear ribonucleoproteins A2/B1 Q2HJ60 0.0 P 36.04/8.67 58Heterogeneous nuclear ribonucleoproteins A2/B1 P22626 0.482 37.5/8.97 45High mobility group protein B1 P10103 0.212 25.1/5.62 46High mobility group protein B1 P10103 0.456 P 25.1/5.62 31GTP-binding nuclear protein Ran Q3T054 0.190 24.6/7.01 55Histone H2B type 1 P62808 0.0 P 13.9/10.31 48Nonhistone chromosomal protein HMG-14 P02316 0.0 P 10.8/9.49 46

TranslationElongation factor 2 Q3SYU2 0.4 P 96.3/6.41 44Eukaryotic translation elongation factor 1 gamma gi|95769122 0.341 50.6/6.33 68Probable ATP-dependent RNA helicase DDX48 Q2NL22 0.318 47.1/6.3 7840S ribosomal protein S17 P63273 0.155 P 15.6/9.85 70

Protein fateHeat shock protein HSP 90-alpha Q76LV2 0.284 P 85.1/4.93 46Heat shock protein HSP 90-beta Q76LV1 0.0 P 83.5/4.97 52Heat shock 70 kDa protein 5 gi|115495027 0.209 P 72.5/5.07 5510 kDa heat shock protein P61603 0.129 10.9/8.89 66

Signal transductionIntegrin alpha-V P80746 0.199 P 117.1/5.55 23Integrin alpha-V P80746 0.389 117.1/5.55 30Rab GDP dissociation inhibitor alpha P21856 0.308 51.1/5.0 44Ras suppressor protein 1 Q5E9C0 0.0 P 31.5/8.56 52

Author's personal copy

6 K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9

Table 1 (Continued )

Proteins and functional categories Accession number(SwissProt/Trembl or NCBI)

Factor inf./non-inf. P* Theoret. MW/pI Seq. cov. (%)

14-3-3 protein theta Q3SZI4 0.423 P 28.0/4.68 65Annexin A2 P04272 0.407 38.9/6.92 72

Cellular transportVoltage-dependent anion-selective channel protein 2 P68002 0.310 32.1/7.48 72Voltage-dependent anion-selective channel protein 3 Q9MZ13 0.383 31.1/8.95 51Chloride intracellular channel protein 4 Q9XSA7 0.259 P 28.9/5.6 78

ApoptoseCaspase 8 gi|114051073 0.0 P 57.4/5.51 36

ProteaseCatalase P00432 0.0 60.1/6.78 3226S protease regulatory subunit S10B Q2KIW6 0.216 44.3/6.74 58

Immune responseAnnexin A1 P46193 0.416 39.2/6.37 64

Stress responsePeroxiredoxin-4 Q9BGI2 0.383 P 30.9/6.01 57

Interaction with environmentAnnexin A5 P81287 0.324 36.1/4.86 62

UnclassifiedLOC504957 (alpha-2-macroglobulin receptor-associated protein) gi|122692301 0.360 41.9/7.36 40Growth hormone-like protein 1 gi|58339172 0.0 P 25.2/5.33 56Glypican 4 gi|119919671 0.5 P 63.5/5.85 30Protein SIX6OS1 Q9CTN5 0.409 67.3/5.45 24Cysteine-rich protein 2 Q0VFX8 0.126 P 23.4/9.01 62

Seq. cov. (%)—sequence coverage.* Indication of phosphorylation (Biotool).

** Several detected down-regulated protein spots.

Table 2Up-regulated host proteins in E. bovis-infected BFGC (14 days p.i.).

Proteins and functional categories Accession number(SwissProt/Trembl or NCBI)

Factor (inf./non-inf) P* Theoret. MW/pI Seq. cov. (%)

MetabolismGlutamate dehydrogenase 1 P00366 N.A. 61.8/7.25 36Protein disulfide-isomerase A3 P38657 4.331 57.3/6.23 31d-3-Phosphoglycerate dehydrogenase Q5EAD2 3.476 57.3/6.47 49Aldehyde dehydrogenase P20000 4.051 57.1/7.55 46Isocitrate dehydrogenase Q9XSG3 3.372 47.1/6.13 48ADP/ATP translocase 1 P48962 3.610 P 33.1/9.73 35Enoyl-CoA hydratase Q58DM8 2.684 31.6/8.82 51Malate dehydrogenase Q32LG3 N.A. 36.1/8.82 67GTP:AMP phosphotransferase P08760 5.230 P 25.7/9.02 62

Structure proteinsTransgelin-2 Q5E9F5 3.908 22.6/8.39 88Transgelin-2 Q5E9F5 4.965 P 22.6/8.39 85Desmin O62654 7.109 53.6/5.21 64Caldesmon 1** gi|27806279 2.789–3.326 62.1/6.24 51Alpha-actinin-1 Q3B7N2 2.671 103.5/5.25 41

Protein fateEndoplasmin (heat shock protein 90 kDa) Q95M18 2.693 92.7/4.76 48Heat shock cognate 71 kDa P19120 4.010 71.4/5.49 38Heat shock 70 kDa protein 9B gi|77735995 N.A. P 73.9/5.97 42Calreticulin P52193 2.323 48.2/4.31 26Calumenin Q3T0K1 2.250 37.2/4.47 49

Signal transductionRab GDP dissociation inhibitor beta P50397 2.691 50.96/5.94 50

Cellular transportVacuolar ATP synthase subunit B P31408 3.446 56.9/5.66 49ATP synthase subunit alpha P19483 3.317 59.8/9.21 53

UnclassifiedGRIP and coiled-coil domain-containing protein 2 Q8IWJ2 N.A. 185.5/5.08 14Hypothetical protein Chro.70256 gi|67614925 6.090 62.6/9.17 19Hypothetical protein XP 532413 gi|73975430 2.658 46.9/6.34 31Homocysteine-inducible, endoplasmic reticulum

stress-inducible, ubiquitin-like domain member 1gi|73949814 N.A. 18.9/7.03 44

N.A.: a factor could not be calculated, as the respective protein spot was not detectable in non-infected samples.* Indication of phosphorylation (Biotool).

** Several detected up-regulated protein spots.

Author's personal copy

K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9 7

Fig. 3. Functional distribution of host cell proteins up-regulated (A) and down-regulated (B) by E. bovis infection (14 days p.i.). Whole cell preparations of E. bovis-infected BFGCwere subjected to 2DE, peptide mass fingerprinting and MALDI-TOF-MS analyses and compared to non-infected controls. Significantly up-regulated (A) and down-regulated(B) host cell proteins were grouped according to annotation analyses into different functional categories and expressed relatively to the total numbers of up-regulatedmolecules.

dance of myosin-10, a protein known to be involved in cell division[34,35], may be of further interest.

3.2.3. Core proteins, transcription and translation factorsE. bovis infection clearly influenced the abundance of molecules

involved in gene transcription and translation. Besides histone-related proteins (histone H2B type 1, histone-binding proteinRBBP4) and the key molecule in nuclear ex- and import Ran [36,37],E. bovis also modulated several molecules of the heterogeneous

nuclear ribonuclear protein (hnRNP) family, a finding that agreeswith data on T. gondii [31]. hnRNPs play a key role in pre-mRNA pro-cessing within the nucleus [38] and in the transport of processedmRNA [39]. Overall, seven hnRNPs (hnRNP A/B, A2/B1, C, D0B iso-form 6, H3, M, K; see Table 1) were decreased in their abundancein infected cells, suggesting a general role of this protein family inthe regulation of Eimeria-induced changes in gene transcription.Moreover, molecules involved in mRNA splicing, such as the U4/U6small ribonuclear protein Prp4 and a protein homologue to the

Author's personal copy

8 K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9

nuclear matrix protein 200 of Saccharomyces cervesiae and othermammalian species [40] and of matrin-3, which is involved in genetranscription by binding to other matrix proteins in order to forma fibrogranular nuclear network [41] suggest an impact of E. bovisinfection on host cell gene transcription. The modified abundanceof translation-related proteins (eukaryotic translation elongationfactor 1 gamma, elongation factor 2, ATP-dependent RNA helicaseDDX48, 40S ribosomal protein S17) in infected cells additionallyindicates the parasite-triggered modulation of this cellular process.

3.2.4. ApoptosisIt is well known that intracellular stages of other apicom-

plexan parasites, such as Cryptosporidium parvum [42], Theileriaparva [43–45], T. gondii [46–49,14], Neospora caninum [50–52]and Eimeria [18] modulate host cell apoptosis to guarantee suc-cessful intracellular development (for review see Ref. [53]). Todate, only few details are known about the molecular mechanismsallowing for long-term survival of E. bovis within adequate hostcells. In detailed studies Lang et al. [17] have recently shown thatcells infected with viable sporozoites of E. bovis are protected byenhanced expression of c-FLIP and c-IAP1, anti-apoptotic moleculesinvolved in different apoptosis signalling pathways, i. e., the inner[54–58] and the receptor-related apoptotic pathway [59–61]. Alsoenhanced levels of NF�B in E. bovis-infected cells were consideredan indicator of apoptosis inhibition [18]. In addition, the currentstudy revealed a strongly reduced abundance of caspase 8 in E.bovis-infected host cells, an effect which is also seen in T. gondii-infected cells [13]. So far, it is not known whether this is a resultof E. bovis-triggered c-IAP1 suppression, or if caspase 8 representsan additional target of parasite-induced modulation of the Fas/Fasligand-mediated pathway of apoptosis.

Also heat shock proteins (HSP) play a role in host cell apoptosis.Certain members of the HSP27-, HSP70- and HSP90-families exhibitanti-apoptotic effects via both the receptor-mediated pathway andthe inner, mitochondrial pathway [62,63]. In the current study dif-ferent HSPs belonging to the HSP70 (heat shock 70 kDa protein5, heat shock 70 kDa protein 9B) and HSP-90 (heat shock protein90 kDa, HSP90-alpha, HSP90-beta) families were modulated in E.bovis-infected host cells. This might indicate an additional mech-anism used by E. bovis to interfere with host cell apoptosis for itsbenefit.

Another interesting candidate related to apoptosis – besidesother functions – is the voltage-dependent anion channel protein 3(VDAC-3) which was also found down-regulated in E. bovis-infectedhost cells. VDAC-3 forms part of a mitochondrial pore which trans-ports adenine nucleotides through the mitochondrial membrane[64]. Especially during cell stress these pores can become perme-able for cytochrome c, the release of which is the key factor of themitochondrial-mediated pathway of apoptosis [65]. Further stud-ies may show whether a direct link exists between E. bovis-inducedlowered abundance of VDAC-3 and inhibition of host cell apoptosis.

4. Conclusion

This is the first report on the modulation of the host cell pro-teome by an Eimeria species and it should be interpreted as a firststep to a better understanding of global parasite–host cell inter-actions. The data of this investigation reflect the situation at thevery end of first merogony with down-modulatory processes beingmost prominent. For more detailed insights into host cell regula-tion induced by different developmental stages of E. bovis furtherproteomic studies comprising kinetic analyses are needed. To date,data of microarray-based kinetic studies are under evaluation andcomparative analyses of data resulting from both techniques willthen provide useful tools to elucidate active modulation of the host

cell on the level of both the Eimeria-specific and, in comparison toother pathogens, the protozoan- or pathogen-specific one.

Acknowledgments

We are deeply indebted to Brigitte Hofmann and Birgit Rein-hardt for their outstanding technical assistance in cell culture andto Michael Dreisbach, Hans Günther Welker and Christina Schmidtat the Institute of Biochemistry for their excellent support in pro-tein purification. This work was supported by the German ResearchFoundation through the Graduated College 455 “Molecular Veteri-nary Medicine” at the Justus Liebig University Giessen (K. Lutz).

References

[1] Daugschies A, Najdrowski M. Eimeriosis in cattle: current understanding. J VetMed B Infect Dis Vet Public Health 2005;52:417–27.

[2] Nyberg PA, Hammond DM. Description of the sporulated oozysts and sporo-zoites of four species of bovine coccidia. J Parasitol 1965;51:669–73.

[3] Hammond DM, Ernst JV, Miner ML. The development of first generation sch-izonts of Eimeria bovis. J Protozool 1966;13:559–64.

[4] Hammond DM, Fayer R. Cultivation of Eimeria bovis in three established celllines and in bovine tracheal cell line cultures. J Parasitol 1968;54:559–68.

[5] Coppens I, Sinai AP, Joiner KA. Toxoplasma gondii exploits host low-densitylipoprotein receptor-mediated endocytosis for cholesterol acquisition. J CellBiol 2000;149:167–80.

[6] Sinai AP, Joiner KA. Safe haven: the cell biology of nonfusogenic pathogenvacuoles. Annu Rev Microbiol 1997;51:415–62.

[7] Frias MA, Rebsamen MC, Gerber-Wicht C, et al. Prostaglandin E2 activates Stat3in neonatal rat ventricular cardiomyocytes: a role in cardiac hypertrophy. Car-diovasc Res 2007;73:57–65.

[8] Fisch S, Gray S, Heymans S, et al. Kruppel-like factor 15 is a regulator of car-diomyocyte hypertrophy. Proc Natl Acad Sci USA 2007;104:7074–9.

[9] Green DR. Apoptotic pathways: paper wraps stone blunts scissors. Cell2000;102:1–4.

[10] Laliberte J, Carruthers VB. Host cell manipulation by the human pathogen Tox-oplasma gondii. Cell Mol Life Sci 2008.

[11] Martin AM, Liu T, Lynn BC, et al. The Toxoplasma gondii parasitophorous vacuolemembrane: transactions across the border. J Eukaryot Microbiol 2007;54:25–8.

[12] Sinai AP, Joiner KA. The Toxoplasma gondii protein ROP2 mediates hostorganelle association with the parasitophorous vacuole membrane. J Cell Biol2001;154:95–108.

[13] Vutova P, Wirth M, Hippe D, et al. Toxoplasma gondii inhibits Fas/CD95-triggeredcell death by inducing aberrant processing and degradation of caspase 8. CellMicrobiol 2007;9:1556–70.

[14] Keller P, Schaumburg F, Fischer SF, et al. Direct inhibition of cytochromec-induced caspase activation in vitro by Toxoplasma gondii reveals novelmechanisms of interference with host cell apoptosis. FEMS Microbiol Lett2006;258:312–9.

[15] Molestina RE, Payne TM, Coppens I, et al. Activation of NF-kappaB by Toxo-plasma gondii correlates with increased expression of antiapoptotic genes andlocalization of phosphorylated IkappaB to the parasitophorous vacuole mem-brane. J Cell Sci 2003;116:4359–71.

[16] Hermosilla C, Schröpfer E, Stowasser M, et al. Cytoskeletal changes in Eimeriabovis-infected host endothelial cells during first merogony. Vet Res Commun2008.

[17] Lang M, Kann M, Zahner H, et al. Inhibition of host cell apoptosis by Eimeriabovis sporozoites. Vet Parasitol 2009;160:25–33.

[18] del Cacho E, Gallego M, Lopez-Bernad F, et al. Expression of anti-apoptotic fac-tors in cells parasitized by second-generation schizonts of Eimeria tenella andEimeria necatrix. Vet Parasitol 2004;125:287–300.

[19] Alcala-Canto Y, Ibarra-Velarde F. Cytokine gene expression and NF-kappaB acti-vation following infection of intestinal epithelial cells with Eimeria bovis orEimeria alabamensis in vitro. Parasite Immunol 2008;30:175–9.

[20] Hermosilla C, Zahner H, Taubert A. Eimeria bovis modulates adhesion moleculegene transcription in and PMN adhesion to infected bovine endothelial cells.Int J Parasitol 2006;36:423–31.

[21] Taubert A, Zahner H, Hermosilla C. Eimeria bovis infection enhances adhe-sion of peripheral blood mononuclear cells to and their transmigrationthrough an infected bovine endothelial cell monolayer in vitro. Parasitol Res2007;101:591–8.

[22] Taubert A, Hermosilla C, Behrendt J, et al. Responses of bovine endothelial cellsin vitro to coccidia (Eimeria bovis, Toxoplasma gondii, Neospora caninum) infec-tions possibly involved in innate immunity. Berl Münch Tierarztl Wochenschr2006;119:274–81.

[23] Taubert A, Zahner H, Hermosilla C. Dynamics of transcription of immunomod-ulatory genes in endothelial cells infected with different coccidian parasites.Vet Parasitol 2006;142:214–22.

[24] Fiege N, Klatte D, Kollmann D, et al. Eimeria bovis in cattle: colostral transferof antibodies and immune response to experimental infections. Parasitol Res1992;78:32–8.

Author's personal copy

K. Lutz et al. / Molecular & Biochemical Parasitology 175 (2011) 1–9 9

[25] Jackson AR. The isolation of viable coccidial sporozoites. Parasitology1964;54:87–93.

[26] Hermosilla C, Barbisch B, Heise A, et al. Development of Eimeria bovis in vitro:suitability of several bovine, human and porcine endothelial cell lines, bovinefetal gastrointestinal, Madin–Darby bovine kidney (MDBK) and African greenmonkey kidney (VERO) cells. Parasitol Res 2002;88:301–7.

[27] Hermosilla C, Stamm I, Taubert A, et al. Fluorescent Eimeria bovis sporozoitesand meront stages in vitro: a helpful tool to study parasite–host cell interac-tions. Parasitol Res 2008;102:777–86.

[28] Forst CV. Host–pathogen systems biology. Drug Discov Today 2006;11:220–7.

[29] Krug EC, Marr JJ, Berens RL. Purine metabolism in Toxoplasma gondii. J Biol Chem1989;264:10601–7.

[30] Sibley LD. No more free lunch. Nature 2002;415:843–4.[31] Nelson MM, Jones AR, Carmen JC, et al. Modulation of the host cell proteome

by the intracellular apicomplexan parasite Toxoplasma gondii. Infect Immun2008;76:828–44.

[32] Gruenbaum Y, Wilson KL, Harel A, et al. Review: nuclear lamins—structuralproteins with fundamental functions. J Struct Biol 2000;129:313–23.

[33] Ottaviano Y, Gerace L. Phosphorylation of the nuclear lamins during interphaseand mitosis. J Biol Chem 1985;260:624–32.

[34] Titus MA. Cell biology: myosins meet microtubules. Nature 2004;431:252–3.[35] Woolner S, O’Brien LL, Wiese C, et al. Myosin-10 and actin filaments are essen-

tial for mitotic spindle function. J Cell Biol 2008;182:77–88.[36] Moore MS, Blobel G. The GTP-binding protein Ran/TC4 is required for protein

import into the nucleus. Nature 1993;365:661–3.[37] Pemberton LF, Paschal BM. Mechanisms of receptor-mediated nuclear import

and nuclear export. Traffic 2005;6:187–98.[38] Dreyfuss G, Matunis MJ, Pinol-Roma S, et al. hnRNP proteins and the biogenesis

of mRNA. Annu Rev Biochem 1993;62:289–321.[39] Pinol-Roma S, Dreyfuss G. hnRNP proteins: localization and transport between

the nucleus and the cytoplasm. Trends Cell Biol 1993;3:151–5.[40] Lu X, Legerski RJ. The Prp19/Pso4 core complex undergoes ubiquitylation and

structural alterations in response to DNA damage. Biochem Biophys Res Com-mun 2007;354:968–74.

[41] Valencia CA, Ju W, Liu R. Matrin 3 is a Ca2+/calmodulin-binding protein cleavedby caspases. Biochem Biophys Res Commun 2007;361:281–6.

[42] Chen XM, Levine SA, Splinter PL, et al. Cryptosporidium parvum activates nuclearfactor kappaB in biliary epithelia preventing epithelial cell apoptosis. Gastroen-terology 2001;120:1774–83.

[43] Heussler VT, Kuenzi P, Rottenberg S. Inhibition of apoptosis by intracellularprotozoan parasites. Int J Parasitol 2001;31:1166–76.

[44] Heussler VT, Machado Jr J, Fernandez PC, et al. The intracellular parasite Thei-leria parva protects infected T cells from apoptosis. Proc Natl Acad Sci USA1999;96:7312–7.

[45] Kuenzi P, Schneider P, Dobbelaere DA. Theileria parva-transformed T cellsshow enhanced resistance to Fas/Fas ligand-induced apoptosis. J Immunol2003;171:1224–31.

[46] Dobbelaere DA, Kuenzi P. The strategies of the Theileria parasite: a newtwist in host–pathogen interactions. Curr Opin Immunol 2004;16:524–30.

[47] Caamano J, Tato C, Cai G, et al. Identification of a role for NF-kappa B2 in theregulation of apoptosis and in maintenance of T cell-mediated immunity toToxoplasma gondii. J Immunol 2000;165:5720–8.

[48] Lüder CG, Gross U. Apoptosis and its modulation during infection with Tox-oplasma gondii: molecular mechanisms and role in pathogenesis. Curr TopMicrobiol Immunol 2005;289:219–37.

[49] Carmen JC, Hardi L, Sinai AP. Toxoplasma gondii inhibits ultraviolet light-induced apoptosis through multiple interactions with the mitochondrion-dependent programmed cell death pathway. Cell Microbiol 2006;8:301–15.

[50] Nishikawa Y, Mishima M, Nagasawa H, et al. Interferon-gamma-inducedapoptosis in host cells infected with Neospora caninum. Parasitology2001;123:25–31.

[51] Nishikawa Y, Makala L, Otsuka H, et al. Mechanisms of apoptosis in murinefibroblasts by two intracellular protozoan parasites, Toxoplasma gondii andNeospora caninum. Parasite Immunol 2002;24:347–54.

[52] Herman RK, Molestina RE, Sinai AP, et al. The apicomplexan pathogen Neosporacaninum inhibits host cell apoptosis in the absence of discernible NF-kappa Bactivation. Infect Immun 2007;75:4255–62.

[53] Graumann K, Hippe D, Gross U, et al. Mammalian apoptotic signalling path-ways: multiple targets of protozoan parasites to activate or deactivate host celldeath. Microbes Infect 2009;11:1079–87.

[54] Duckett CS, Nava VE, Gedrich RW, et al. A conserved family of cellular genesrelated to the baculovirus iap gene and encoding apoptosis inhibitors. EMBO J1996;15:2685–94.

[55] Liston P, Roy N, Tamai K, et al. Suppression of apoptosis in mammalian cells byNAIP and a related family of IAP genes. Nature 1996;379:349–53.

[56] Deveraux QL, Reed JC. IAP family proteins—suppressors of apoptosis. Genes Dev1999;13:239–52.

[57] Irmler M, Thome M, Hahne M, et al. Inhibition of death receptor signals bycellular FLIP. Nature 1997;388:190–5.

[58] Bratton SB, Walker G, Srinivasula SM, et al. Recruitment, activation andretention of caspases-9 and -3 by Apaf-1 apoptosome and associated XIAPcomplexes. EMBO J 2001;20:998–1009.

[59] Micheau O, Thome M, Schneider P, et al. The long form of FLIP is an acti-vator of caspase-8 at the Fas death-inducing signaling complex. J Biol Chem2002;277:45162–71.

[60] Scarabelli TM, Knight R, Stephanou A, et al. Clinical implications of apoptosisin ischemic myocardium. Curr Probl Cardiol 2006;31:181–264.

[61] Golks A, Brenner D, Krammer PH, et al. The c-FLIP-NH2 terminus (p22-FLIP)induces NF-kappaB activation. J Exp Med 2006;203:1295–305.

[62] Beere HM, Wolf BB, Cain K, et al. Heat-shock protein 70 inhibits apoptosis bypreventing recruitment of procaspase-9 to the Apaf-1 apoptosome. Nat CellBiol 2000;2:469–75.

[63] Beere HM. Death versus survival: functional interaction between the apop-totic and stress-inducible heat shock protein pathways. J Clin Invest2005;115:2633–9.

[64] Sampson MJ, Lovell RS, Craigen WJ. Isolation, characterization, and map-ping of two mouse mitochondrial voltage-dependent anion channel isoforms.Genomics 1996;33:283–8.

[65] Crompton M. The mitochondrial permeability transition pore and its role in celldeath. Biochem J 1999;341(Pt 2):233–49.