8
Click here to load reader
Click here to load reader
Aec
CPSa
b
c
h
III
a
ARRAA
KDEPHB
Bmm
GT
O
0h
Toxicology Letters 215 (2012) 193– 200
Contents lists available at SciVerse ScienceDirect
Toxicology Letters
jou rn al h om epage: www.elsev ier .com/ locate / tox le t
chronic oral exposure of pigs with deoxynivalenol partially prevents the acuteffects of lipopolysaccharides on hepatic histopathology and blood clinicalhemistry
assandra Staneka, Nicole Reinhardta, Anne-Kathrin Diesinga, Constanze Nossola, Stefan Kahlerta,atricia Panthera,1 , Jeannette Kluessa,∗ , Hermann-Josef Rothköttera , Doerthe Kuesterb , Bianca Brosigc ,usanne Kerstenc, Sven Dänickec
Institute of Anatomy, Medical Faculty, Otto-von-Guericke University, 39120 Magdeburg, GermanyInstitute of Pathology, Medical Faculty, Otto-von-Guericke University, 39120 Magdeburg, GermanyInstitute of Animal Nutrition, Federal Research Institute for Animal Health, 38116 Braunschweig, Germany
i g h l i g h t s
Systemic deoxynivalenol alone does not alter hepatic morphology and function.Escherichia coli LPS induces hepatic inflammation and haemorrhage and impairs liver function.Deoxynivalenol feeding alleviates detrimental effects of a subsequent LPS challenge.
r t i c l e i n f o
rticle history:eceived 3 September 2012eceived in revised form 11 October 2012ccepted 13 October 2012vailable online 30 October 2012
eywords:eoxynivalenolscherichia coli lipopolysaccharide
a b s t r a c t
Lipopolysaccharides (LPS), a cell wall component of gram-negative bacteria, and deoxynivalenol (DON),a prevalent Fusarium-derived contaminant of cereal grains, are each reported to have detrimental effectson the liver. A potentiating toxic effect of the combined exposure was reported previously in a mousemodel and hepatocytes in vitro, but not in swine as the most DON-susceptible species. Thus, pigs werefed either a control diet (CON) or a Fusarium contaminated diet (DON, 3.1 mg DON/kg diet) for 37 days. Atday 37 control pigs were infused for 1 h either with physiological saline (CON CON), 100 �g/kg BW DON(CON DON), 7.5 �g/kg BW LPS (CON LPS), or both toxins (CON DON/LPS) and Fusarium-pigs with saline(DON CON) or 7.5 �g/kg BW LPS (DON LPS). Blood samples were taken before and after infusion (−30,
orcine liveristopathologylood chemistry
+30, +60, +120, and +180 min) for clinical blood chemistry. Pigs were sacrificed at +195 min and liverhistopathology was performed. LPS resulted in higher relative liver weight (p < 0.05), portal, periportaland acinar inflammation (p < 0.05), haemorrhage (p < 0.01) and pathological bilirubin levels (CON CON1.0 �mol/L vs. CON LPS 5.4 �mol/L, CON DON/LPS 8.3 �mol/L; p < 0.001). DON feeding alleviated effectsof LPS infusion on histopathology and blood chemistry to control levels, whereas DON infusion alone hadno impact.
Abbreviations: ANOVA, analysis of variance; AST, aspartate-aminotransferase;W, body weight; DON, deoxynivalenol; GGT, �-glutamyltransferase; GLDH, gluta-ate dehydrogenase; HAI, histology activity index; LPS, lipopolysaccharide; MAPK,itogen-activated protein kinase; PSEM, pooled standard error of means.∗ Corresponding author at: Institute of Anatomy, Medical Faculty, Otto-von-uericke University Magdeburg, Leipziger Straße 44, 39120 Magdeburg, Germany.el.: +49 391 67 13602; fax: +49 391 67 13630.
E-mail address: [email protected] (J. Kluess).1 Present address: Department of Stereotactic Neurosurgery, Medical Faculty,tto-von-Guericke University, 39120 Magdeburg, Germany.
378-4274/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.toxlet.2012.10.009
© 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
The trichothecene deoxynivalenol (DON) is a secondary metabo-lite mainly produced by the plant pathogens Fusarium graminearumand Fusarium culmorum, to which men and livestock can be exposedvia food and feed. DON prevalence on corn, wheat, barley, oatsand other grains is dependent on certain climatic conditions suchas frequent rainfalls and moderate temperatures. Therefore, con-taminated cereals are often detected in North America and Europe.
Particularly pigs are highly susceptible to DON, resulting in consid-erable economic losses in pig farming due to reduced feed intakeand live weight gain, vomiting and a compromised immune reac-tion (Bondy and Pestka, 2000; Rotter et al., 1996; Young et al., 1983).
1 y Lett
A6ss2tcT
pg(i(tsc(cpttMaeMcPemLlabMcs
lp
2
2
Rpa(dwtmdtaeT
2
3oaaifo0
94 C. Stanek et al. / Toxicolog
ccording to Ehrlich and Daigle (1987), trichothecenes bind to the0S subunit of ribosomes and interfere with eukaryotic proteinynthesis by translation inhibition. Based on inhibition of proteinynthesis and proliferation (Rotter et al., 1996; Wollenhaupt et al.,006) it is often suggested that tissues and organs with a high pro-ein turnover rate or quickly proliferating cells, such as immuneells and epithelial cells, are major targets of DON (Eriksen, 2003).he liver qualifies as such an organ.
Dose-dependent changes in liver function and morphology werereviously reported, among other things increase in liver enzymes,lycogen depletion and haemosiderosis as features of liver damageTiemann et al., 2006). Albumin synthesis, taking place exclusivelyn the liver, was also decreased in DON-treated fattening pigsGoyarts et al., 2006), further substantiating the impact of DON onhe liver. Besides the mentioned interference with protein synthe-is, DON was also reported to trigger a ribotoxic stress responseharacterised by activation of mitogen-activated protein kinasesMAPK), subsequent release of cytokines and apoptosis of immuneells (Pestka, 2010a). Lipopolysaccharides (LPS), as major com-onent of the outer cell membrane of gram negative bacteria, ishe principal mediator in the pathophysiology of local and sys-emic inflammation and is also involved in this ribotoxic stress via
APK pathway. Gram-negative bacteria are part of the intestinalnd environmental microbiota, thus always present to a certainxtent in vivo, besides their major role in systemic inflammation.oreover, it was demonstrated that LPS enhanced the effect of tri-
hothecenes on pro-inflammatory cytokines in mice (Islam andestka, 2006) and in porcine primary hepatocyte cultures (Döllt al., 2009a,b). Combining these data with the liver’s major tasks ofetabolisation and detoxification of absorbed substances such as
PS and DON puts emphasis on the importance of investigating theiver. However, at present there are scarce data on the impact of LPSnd DON and their putative interaction on pigs as most suscepti-le species, in particularly the impact on the porcine liver function.oreover, the duration and route of exposure of the toxins are cru-
ial factors to their potential toxicity on epithelial cells as previouslyhown in vitro (Diesing et al., 2011).
Therefore, the aim of the present investigation was to ana-yse the potential additive or synergistic effect of DON and LPS onorcine liver morphology and function in vivo.
. Materials and methods
.1. Animals and diets
The experiment was carried out at the Institute of Animal Nutrition, Federalesearch Institute for Animal Health, in Brunswick, Germany according to the Euro-ean Community regulations concerning the protection of experimental animalsnd the guidelines of the regional council of Brunswick, Lower Saxony, Germanyfile number 33.14-42502-04-037/08). Thirty-nine crossbred barrows (German Lan-race × Piétrain) were used with an initial body weight (BW) of 26.4 ± 4.0 kg. Pigsere fed restrictively, 500 g/pig twice daily, and had ad libitum access to water for
he entire experimental period. Diets were formulated to meet or exceed the recom-endations by the German Society of Nutrition Physiology (GfE, 2006). The control
iet (CON) contained 569 g/kg barley, 190 g/kg wheat, 180 g/kg soybean extrac-ion meal, 30 g/kg soybean oil, 3.5 g/kg dicalcium-phosphate, 2.5 g/kg l-lysine–HClnd 25 g/kg premix containing minerals, trace elements and vitamins. Wheat wasxchanged for a DON-contaminated batch (3.1 mg/kg DON) to obtain a DON-diet.he latter contained three times the guidance value for DON (EFSA, 2004).
.2. Experimental design
The experimental design is depicted in Fig. 1. The entire trial was conducted for7 days and comprised a dietary and an infusion regime. BW of pigs was measuredn days 1 and 37. Animals were kept in floor pens without bedding for 27 daysnd thereafter re-allocated to individual metabolism cages as described by Farries
nd Oslage (1961). On day 35 all pigs received two permanent indwelling cathetersnto the left and right external jugular vein, one for blood sampling and the otheror the infusion regimen, according to Goyarts and Dänicke (2006). After two daysf recovery (day 37) pigs fed the control diet received an infusion for 1 h of either.9% NaCl, 100 �g DON/kg BW (D0 156; Sigma–Aldrich, Germany), 7.5 �g LPS/kg BWers 215 (2012) 193– 200
(Escherichia coli O111:B4, Sigma–Aldrich, Germany) or a combination of DON andLPS at the same concentration. Pigs fed the DON-diet received an infusion for 1 hof either 0.9% NaCl or 7.5 �g LPS/kg BW. This resulted in six experimental groups:CON CON, CON DON, CON LPS, CON DON/LPS, DON CON and DON LPS, where thefirst abbreviation denotes the dietary and the second the infusion regimen.
Blood samples were taken at various time points before and after start of infu-sion (−30, +30, +60, +120 and +180 min) into heparinised tubes (S-Monovette® ,Sarstedt, Germany). Animals were sacrificed 195 min after start of infusion and liverswere excised, macroscopically examined and samples taken for histopathologicalanalysis.
2.3. Blood sample preparation and analysis
Blood samples were centrifuged at 2016 × g for 10 min and the separated serumwas analysed for aspartate-aminotransferase (AST), �-glutamyltransferase (GGT),glutamate dehydrogenase (GLDH), albumin, total protein and bilirubin as describedby Dänicke et al. (2012).
2.4. Liver preparation
Directly after removal from the abdominal cavity, livers were weighed anda macroscopic photographic documentation of each organ was performed beforesampling for histopathological examination (Fig. 2). Samples were always takenfrom Margo inferior of the right hepatic lobe. All samples were immediately fixed in4% paraformaldehyde, incubated in distilled water and thereafter in an ascendingalcohol series and finally embedded in paraffin. Samples were cut into 5 �m sectionson a HM 355S rotation microtom (Microm International GmbH, Germany), mountedon glass slides (2–4 sections/slide) and afterwards stained for haematoxylin andeosin (Sigma Aldrich, Germany) to enable histopathological analysis.
2.5. Liver histopathology
The histopathological examination was carried out on an Axioplan 2 micro-scope (Zeiss, Germany) at 200× magnification (unless stated otherwise) equippedwith a colour camera (Diagnostic Instruments Inc.). Micrographs of each samplewere examined using the modified histology activity index (HAI) according to Ishaket al. (1995). Inflammatory, necrotic and haemorrhage parameters were added upfor a cumulative HAI with a maximum possible score of 40, representing the highestdegree of damage (Supplementary Table S1). Inflammation was analysed in por-tal (A), periportal (B) and acinar (C) region and the distribution of inflammationseparately examined for neutrophil (A1, B1 and C1) and eosinophil granulocytesoccurrences (A2, B2 and C2) based on the presence of neutrophil or eosinophil granu-locytes. We also evaluated focal lytic necrosis (D) and confluent necrosis (E). Becauseof high occurrence in the liver sections we added the category haemorrhage (F).During scoring all micrographs were blinded for treatments and allocated to theirrespective treatments after evaluation.
Supplementary data associated with this article can be found, in the onlineversion, at http://dx.doi.org/10.1016/j.toxlet.2012.10.009.
2.6. Statistical analyses
Technical parameters (BW, total and relative liver weight) of all 39 pigs wereanalysed by one-way-ANOVA with a Tukey post hoc test and HAI (n = 39) byKruskal–Wallis test and Dunn’s post hoc test using SAS (Software package, version9.1, SAS Institute, USA). Blood chemistry parameters (n = 37, due to blocked cathetersin two pigs) were analysed using the procedure “Mixed” of SAS with “treatmentgroup”, “time” and their interactions as fixed factors. The repeated measurementon the same pig in the course of time was additionally considered by implementingthe “Repeated” statement into the statistical model. An adjusted Tukey–Kramer posthoc test was further applied to evaluate the least square mean differences.
3. Results
3.1. Body weight, total and relative liver weight
No significant differences were detected in final BW or totalliver weight at the end of the study (Table 1). However, relativeliver weight [g/kg BW] revealed a significant treatment effect andwas highest (approx. by 20%) in all three LPS treatment groups ascompared to CON CON, CON DON and DON CON.
3.2. Macroscopic liver inspection
The evaluation of porcine livers regarding visible pathologi-cal lesions by photographic documentation is shown exemplaryin Fig. 2. Patchy dark red coloured liver surfaces and, after liver
C. Stanek et al. / Toxicology Letters 215 (2012) 193– 200 195
rimen
ippoam
FaD
Fig. 1. Expe
ncisions, parenchyma were the most striking alterations onlyresent in LPS treated groups. These changes were attributed to
etechiae, ecchymoses and sugillations and thus specific formsf haemorrhage. Cholestasis was not observed. Neither DONdministrated animals nor the control group showed any notableacroscopic changes.ig. 2. Representative liver macroscopy in pigs treated with DON and LPS. One photo is gnimals in the specific group. Haemorrhage is only apparent in LPS treated groups (CON LON (CON DON and DON CON) nor the control group (CON CON).
tal design.
3.3. Microscopic liver scoring (HAI)
Fig. 3 displays exemplary micrographs for each of the six exam-ined histopathological criteria based on a modified HAI. Eachtissue alteration such as inflammation, necrosis and haemorr-hage was scored individually (Table 2) and summarised for each
iven exemplary for each treatment, reflecting the common liver macroscopy of allPS, CON DON/LPS, and DON LPS), whereas none are macroscopically visible in the
196 C. Stanek et al. / Toxicology Letters 215 (2012) 193– 200
Table 1Effect of DON and LPS on zoo-technical parameters.
CON CON CON DON CON LPS CON DON/LPS DON CON DON LPS PSEMa p-Valueb
BW (kg)c 42.2 42.3 40.5 42.1 44.3 43.1 0.9 p = 0.935LW (g)d 878.2 849.0 1007.0 1074.0 920.8 1062.0 31.6 p = 0.183rel. LWe (g/kg BW) 21.0 20.1 25.4 25.6 20.9 24.6 0.7 p < 0.05
a PSEM, pooled standard error of means.b One-way-ANOVA with group as fixed factor for each parameter.c BW, body weight.d LW, liver weight.e rel. LW, relative liver weight.
F ramel and reC rrhage
ecFcHm
TE
a
ig. 3. Representative microscopic photograph for each scored histopathological paysed at 200× magnification. Following parameters were histopathologically scored
– acinar inflammation; D – focal lytic necrosis; E – confluent necrosis; F – haemo
xperimental group as a cumulative HAI. The HAI score and theontribution of each criterion to the cumulative score are shown in
ig. 4. The highest scores and thus the most severe histopathologi-al damage were detected for CON LPS and CON DON/LPS. The totalAI score resulted primarily from an increase in eosinophil inflam-ation and haemorrhage in the liver whereas necrotic effectsable 2ffects of DON and LPS on histopathological score in porcine liver.
CON CON CON DON CON LPS
Total score 16.3abc 12.8a 22.5bc
A1 portal neutrophil inflammation 2.4 2.5 3.4
A2 portal eosinophil inflammation 2.0ab 2.5ab 2.8ab
B1 periportal neutrophil inflammation 2.0 2.8 3.3
B2 periportal eosinophil inflammation 1.3ab 0.8a 1.9ab
C1 acinar neutrophil inflammation 1.1a 1.4ab 2.3b
C2 acinar eosinophil inflammation 1.6a 1.3a 3.3b
D focal necrosis 2.3 0.8 1.0
E confluent necrosis 3.4 0.8 1.7
F haemorrhage 0.1ab 0.0a 2.8ab
bc unlike superscripts within a row differ (p < 0.05; Dunn’s post hoc test).A PSEM, pooled standard error of means.B Kruskal, Wallis non-parametric test with group as fixed factor.
ter. Paraffin-embedded liver sections were stained for HE and microscopically ana-presentative pictures given: A – portal inflammation; B – periportal inflammation;; G – healthy liver tissue.
were negligible. However, between those two groups no differencewas observed. Interestingly, neither CON CON nor group DON LPS
showed any statistical significance to the other groups.Assessing the individual scoring parameters showed no signif-icant effect of LPS or DON for neutrophil infiltration and necroses.In contrast, eosinophil granulocytes were always prominent in all
CON DON/LPS DON CON DON LPS PSEMA p-ValueB
24.6b 14.1ac 18.9abc 0.9 p < 0.0013.4 2.0 3.1 0.2 p = 0.083.2b 1.8a 2.6ab 0.1 p < 0.052.8 1.3 2.4 0.2 p = 0.082.1b 1.0ab 1.4ab 0.1 p < 0.052.6b 1.7ab 1.9ab 0.1 p < 0.013.5b 2.0ab 2.4ab 0.2 p < 0.0011.5 1.5 1.5 0.2 p = 0.271.6 2.4 1.3 0.3 p = 0.153.8b 0.3ab 2.3ab 0.3 p < 0.01
C. Stanek et al. / Toxicology Lett
Fig. 4. Histopathological score of porcine liver sections based on the modified his-tological activity index (HAI). Several histopathological parameters were scored inHE-stained porcine liver sections and a cumulative total HAI score was subsequentlycalculated for each experimental group, represented by a column per group. Addi-tionally, contribution of each individually scored parameter to the total score isdetailed in different colours and provided in the legend below the graph. Data wereanalysed by a non-parametric Kruskal–Wallis test with group as fixed factor andlu
sd
3
sdcb
3
(gA(dfa
3
ii
3
a(s(ipt
same time. We aimed to investigate these interactions on porcine
east square mean differences calculated by Dunn’s post hoc test. abc columns withnlike superscripts differ (p < 0.05; Dunn’s post hoc test).
cored regions for CON DON/LPS, albeit not always significantlyifferent to all other experimental groups.
.4. Blood chemistry
All blood chemistry parameters (Table 3) at −30 min before infu-ion were within the physiological range and no difference wasetectable between the experimental groups. Thus, feeding theontrol or DON diet for 37 days elicited no effect on the analysedlood chemistry parameters.
.4.1. Aspartate-aminotransferase (AST)A significant time effect (p < 0.001) and group × time interaction
p < 0.001) were detected, but no group effect (p = 0.90). Assessingroups individually in the time course revealed an increase inST level at +180 min compared to −30 min for CON DON/LPS
p < 0.001) and DON LPS (p < 0.05). Interestingly, no significantifferences were detected for CON LPS. However, all statistical dif-erences were still in the physiological range (8.0–35.0 U/L; Kraftnd Dürr, 2005).
.4.2. �-Glutamyltransferase (GGT)Analysing GGT showed neither a group and time effect nor an
nteraction between both factors and values were within the phys-ological range (<10.0–40.0 U/L; Kraft and Dürr, 2005).
.4.3. Glutamate dehydrogenase (GLDH)A tendency for a time effect (p = 0.09) and a significant inter-
ction between group and time (p < 0.001), but no group effectp = 0.24) was revealed for GLDH. At +180 min GLDH showed aignificant increase for CON DON/LPS compared with CON DONp < 0.01) and DON CON (p < 0.05). Furthermore, a time dependent
ncrease for GLDH was found in CON DON/LPS at +180 min com-ared to the other time points analysed. All changes were still inhe physiological range (0.0–5.0 U/L; Kraft and Dürr, 2005).ers 215 (2012) 193– 200 197
3.4.4. AlbuminA significant interaction between group and time (p < 0.05) and a
significant time effect (p < 0.001) was determined. Analysing albu-min over time showed a significant reduction at +120 (p < 0.05) and+180 min (p < 0.001) compared with the previous time points forCON DON/LPS, although still in physiological range (18.0–31.0 g/L;Kraft and Dürr, 2005).
3.4.5. Total proteinThere was no significant effect of group, but of time (p < 0.001)
and the interaction between both factors (p < 0.001). Analysisof total protein values over time demonstrated a significantdecrease at +180 min for all LPS groups, even marginally belowthe physiological value (55.0–86.0 g/L; Kraft and Dürr, 2005).Furthermore, CON DON/LPS and DON LPS revealed already a sig-nificantly reduced protein content at +120 min. Additionally thegroup DON CON also showed a protein decrease already at +30 min(p < 0.05) and +120 min (p < 0.01) compared with −30 min.
3.4.6. Total bilirubinHighly significant effects were observed for group (p < 0.001),
time (p < 0.001), and the interaction between both factors(p < 0.001). At +180 min all three LPS groups increased bilirubinlevels pathologically (physiological range: 0.1–4.3 �mol/L; Kraftand Dürr, 2005) in time and compared with CON CON (p < 0.001),CON DON (p < 0.05), and DON CON (p < 0.001). Bilirubin levels ofCON DON/LPS were again significantly higher than that of CON LPSand DON LPS (both p < 0.001). Considering that bilirubin levels forCON DON/LPS increased by 83% compared with CON DON and by35% compared with CON LPS, an additive effect for intravenousapplication of both toxins compared with their individual appli-cation can be assumed.
4. Discussion
DON is a common contaminant of crops like wheat, barley,corn and oats and of high importance in food and feed industry.Understanding of variable DON effects also requires the investi-gation of interactions with other agents such as LPS and coherentresponses on animal and human health. Trichothecenes are sug-gested to be both immunostimulatory and immunosuppressivedepending on dose, frequency and duration of exposure as well astype of immune function assay (Rotter et al., 1996; Pestka, 2008,2010b). It was already shown that the toxicity of trichothecenemycotoxins is potentiated by LPS, with the immune system beinga primary target (Islam and Pestka, 2006). Low-dose LPS exposurestimulates immune response that resulted in removal of invadingbacteria, while moderate LPS exposure can evoke tissue injury viaactivation of neutrophils and intravascular coagulation. Further-more, high-dose LPS exposure initiates a chain of inflammatoryevents that results in cell death, tissue injury, and organ fail-ure (Roth et al., 1998). We used a pig model for the purpose ofjoining different routes of toxin exposure. The design aimed tosuperimpose maximal stress combining DON and LPS adminis-tered via different routes. The focus of interest was laid on theliver being the first metabolic station possibly affected both byDON and LPS. The hepatic gland is of major importance in hostdefence due to its clearance and detoxification potential and theproduction of proteins such as acute phase proteins and inflam-matory mediators (Van Amersfoort et al., 2003). Moreover, due tothe pro-inflammatory effects both of LPS and DON this organ mighteven more seriously affected in the presence of both toxins at the
liver histopathology and blood chemistry to test the hypothesis thatLPS-induced inflammation is amplified by the parallel presence ofDON.
198 C. Stanek et al. / Toxicology Letters 215 (2012) 193– 200
Table 3Effect of DON and LPS on functional liver markers in peripheral blood.
Time (min) CON CON CON DON CON LPS CON DON/LPS DON CON DON LPS
AST GGT GLDH AST GGT GLDH AST GGT GLDH AST GGT GLDH AST GGT GLDH AST GGT GLDH
−30 10.7 15.0 2.4 12.5 16.8 2.4 11.8 14.8 2.4 11.7A 17.7 1.2A 14.2 20.7 2.1 12.3A 16.5 1.630 12.2 13.7 2.4 13.2 16.2 1.4 12.8 14.3 1.6 13.7AB 16.7 2.2A 13.0 20.2 2.4 13.2AB 15.0 2.060 10.3 13.3 1.8 14.3 16.3 1.7 12.8 14.8 1.3 13.1A 17.0 2.4A 13.3 16.8 1.7 12.3A 16.7 2.9
120 9.7 13.5 1.6 12.7 16.7 1.5 11.6 15.8 2.4 13.4AB 18.7 3.1AB 12.3 21.5 1.7 12.7AB 17.0 3.2180 10.6 13.7 1.8ab 12.8 14.2 1.4a 14.3 16.7 2.2ab 15.7B 19.1 4.5bB 12.3 15.8 1.6a 15.2B 17.5 3.8ab
ALB PROT BILI ALB PROT BILI ALB PROT BILI ALB PROT BILI ALB PROT BILI ALB PROT BILI
−30 30.1 59.5 0.9 29.9 60.8 1.1 28.7 59.5A 1.2A 31.6A 61.3A 0.9A 30.3 60.4A 1.0 30.3 60.3A 0.9A
30 27.5 57.8 1.1 29.6 58.6 1.2 27.7 58.7A 1.2A 30.9A 61.2A 1.2A 26.4 54.2B 1.6 26.2 62.5A 1.1A
60 27.4 57.7 1.0 30.5 59.6 1.0 25.5 54.6AB 1.4A 29.1AB 58.4AB 0.9A 27.4 56.7AB 1.6 27.9 57.8AB 1.4A
120 26.2 55.0 1.1 28.4 56.7 1.1 24.1 54.7AB 2.8A 25.9B 53.6B 2.5A 26.1 54.8B 1.0 26.4 54.1B 1.8A
180 28.5 58.8 1.0a 28.7 56.2 1.4a 24.7 51.5B 5.4bB 24.6B 51.4C 8.3cB 26.7 56.1AB 1.1a 26.3 53.6B 4.3bB
AST, aspartate-aminotransferase (U/L); GGT, �-glutamyltransferase (U/L); GLDH, glutamate dehydrogenase (U/L); ALB, albumin (g/L); PROT, total protein (g/L); BILI, totalbilirubin (�mol/L).a meterT
tigd7nttdmplt2tBs
afserpg(rbaifiaehf
tFh2cte
bc unlike lowercase superscripts within a row (group effect for corresponding paraukey–Kramer post hoc test).
In general, feeding or infusing DON singularly, with no other fac-or combined, elicited no impact on liver morphology and functionn our study. This is in agreement with previous results shown inilts (Goyarts et al., 2007; Tiemann et al., 2008). These experimentsemonstrated, that feeding DON at 4.4 mg/kg feed from days 35 to0 of gestation, amounting to a DON-intake of 49 �g/kg BW d, didot alter liver morphology and blood chemistry, which was alsorue for other organs such as spleen and kidneys. In contrast tohese studies and our own trial, Mikami et al. (2010) could show aetrimental effect of acute DON injection (1000 �g/kg BW) on liverorphology. Apoptotic bodies and DNA strand-breaks were most
rominent at 6 h after DON exposure and returned to moderateevels at 24 h post exposure as evidenced by analysis of immunohis-ochemistry. Furthermore, periacinar necrosis was evident at 6 and4 h in some animals. However, Mikami and co-workers used a tenimes higher dosage of DON compared with our study (100 �g/kgW DON infused), which might explain the difference to othertudies and our own one.
In contrast to the absent DON effects we could demonstrate most prominent LPS impact both on liver morphology andunction. An increase in relative liver weight was already demon-trated for broilers and dogs (Mireles et al., 2005; MacLeant al., 1956). We assumed that this LPS-associated increase inelative liver weight is a result of haemorrhage which is sup-orted by our macroscopic and also histological findings. Inross macroscopic examination the livers of the three LPS groupsCON LPS, CON DON/LPS and DON LPS) impressed by patchy darked coloured liver surfaces and parenchyma and are supposed toe shock like (Utili et al., 1977). Such haemorrhagic liver alter-tions have been reported by Korish and Arafa (2011) in LPSnjected rats. The observed macroscopic intrahepatic hyperper-usion might be causative for the already mentioned increasen relative liver weight in LPS groups. The macroscopic findingsre further substantiated by microscopic examinations, whereosinophil and neutrophil granulocyte infiltration and haemorr-age was found to be exclusively caused by LPS infusion in control
ed pigs.Analysing the score more closely revealed a higher eosinophilic
han neutrophilic contribution to the inflammatory liver response.urthermore, bacterial toxins can enhance vasodilatation,aemostasis and intravasal coagulation (Dauphinee and Karsan,
006). Increased membrane permeability by apoptotic endothelialells and resulting haemorrhage per diapedesis are other charac-eristics of LPS-induced changes (Frey and Finlay, 1998; Cybulskyt al., 1988).s) and ABC unlike uppercase superscripts within a column differ (p < 0.05; adjusted
In literature various studies are reported investigating theimpact of sepsis, a systemic inflammatory response caused eitherby pure LPS or bacteria, on liver morphology and function.Martens et al. (2007) induced a septic state in pigs of about25 kg BW by infusing Aeromonas hydrophila, a gram-negative �-proteobacterium, over 4 h continuously and investigating livermorphology in the histopathological sections. Hepatocytes wereswollen and sinusoids congested with blood, thus intrahepatic hae-morrhage evident. The detrimental effect of sepsis on liver was alsoreported by Leifsson et al. (2010), using Staphylococcus aureus injec-tion in the ear vein to induce endotoxaemia in 25 kg pigs. Theyreported a sequence of events in the liver, with a peak in leuco-cytes after 24 h post infectionem (p.i.), a pathological increase inserum bilirubin and AST levels after 36 h p.i. and interstitial fibrinaccumulation and occurrence of microabscesses and necrotic fociafter 48 h. An even more detailed sequence of events was demon-strated in the study of Saetre et al. (2001) applying 1.7 �g/kg BWE. coli LPS over 6 h to growing pigs and taking liver biopsies before,during and at the end of infusion. Already after 1 h leucocyte infil-tration, oedema and sinusoidal dilatation were visible, whereasafter 3 h endothelial damage, lipid accumulation in and damage ofhepatocytes were obvious besides phagocyting Kuppfer cells andthrombi. At 6 h the mentioned damages were more widespread andnecrotic areas also present. Nakajima et al. (2000) reported a sim-ilar sequence of events after injection of 100 �g/kg BW E. coli LPS,but with necrotic changes only at 24 h p.i. rather than already at6 h. However, it needs to be mentioned that the number of animalswas low, ranging from one to three pigs, thus risking a biased result.Most liver changes were also demonstrated in our own study withthe exception of the necrotic areas. One reason for the absence ofnecrotic changes in the liver in our experiment might be the timeof sampling in relation to LPS infusion. We terminated the trial at195 min after start of LPS infusion, whereas necrotic foci seem toappear later in the chronological sequence of sepsis.
Besides the liver morphology, various groups (Myers et al., 1997;Eriksson et al., 1998; Saetre et al., 2001) also demonstrated changesin liver function in response to E. coli LPS infusion as evidenced in anelevation of bilirubin and AST over time. This was also observed inour pigs fed control diets and receiving LPS infusion solely. Bilirubinwas already above physiological levels at 180 min after start of LPSinfusion, whereas AST increased in time but still in the physiologi-
cal range. In the above-mentioned publications, the shift of serumbilirubin response into pathological levels varied from 30, 180 to300 min, whereas AST levels increased in a time-dependent mannerin most studies with earliest pathological concentrations at 180 h.
y Lett
HmltAhepsnNtl
taticeddrtrDf
fbboLrpeasaohhlvmld
wottaroiipahcoenim
C. Stanek et al. / Toxicolog
yperbilirubinaemia indicates a problem in haem-metabolism,ost commonly resulting from hepatic or bile duct disorders. The
atter can be excluded for our study because GGT, a marker forhe biliary tract, was unaltered. The time-dependent elevation ofST and GLDH, albeit not yet pathological at 180 min, confirms aepatic dysfunction already evidenced in liver histopathology. Bothnzymes are largely located in mitochondrial membranes and cyto-lasm of hepatocytes, although AST is also present in other tissuesuch as cardiac muscle and kidneys. Thus, cells have to undergoecrosis in order to release the enzymes into the blood stream.ecrotic areas were – not yet – increased in the livers of the LPS-
reated pigs, which supports the still physiological AST and GLDHevels.
Total protein and albumin levels in blood were depressed overime by LPS infusion in our study, but not to such an extent that
statistical difference between treatments could be proofed athe endpoint. This effect was previously reported in pigs receiv-ng 3 �g/kg BW E. coli LPS over 24 h and showing decreased proteinontent in blood still 24 h after cessation of LPS exposure (Bruinst al., 2002). Albumin is the most abundant plasma protein and pro-uced in the liver and thus its decrease can be connected to liverysfunction caused for example by inflammation or acute-phaseeaction in response to an intravenous endotoxin treatment. Addi-ionally, hepatic protein synthesis rate as well as albumin synthesisate was significantly diminished by LPS infusion irrespective ofON co-exposure in our experimental pigs (Kullik et al., submitted
or publication).One reason for the observed changes in liver morphology and
unction during septic conditions could be an alteration in liverlood circulation. A marked decrease in porcine liver perfusionoth on venous (portal vein) and arterial branch (hepatic artery)f circulation was reported 3 h after start of infusion of E. coliPS (Saetre et al., 2001). This decrease in liver perfusion and theesulting decrease in oxygen saturation in response to sepsis wasreviously reported by other groups (Dahm et al., 1999; Anderssont al., 2010) using low levels of E. coli LPS (0.5 �g/kg BW for 180 minnd 2.5 �g/kg BW for 300 min). Lactate levels in peripheral blooduccessively increased, most likely due to a shift from aerobic tonaerobic glycolysis in the liver resulting from lack of sufficientxygen for these processes. This in turn can lead to damage ofepatocytes and endothelial lining of sinusoids with subsequentaemorrhage per diapedesis visible in the histopathology of septic
ivers. Although we did not determine flow conditions in the portalein and hepatic artery nor hepatic microcirculation in our experi-ent we can assume that in our LPS-animals oxygen supply in the
iver was insufficient and thus contributed to the morphologicalamage.
Interestingly, pigs fed a DON-contaminated diet and infusedith LPS did not differ from control animals in the majority
f parameters investigated in our experiment, with the excep-ion of bilirubin and total protein levels. This might imply thathe body’s response to an immune challenge such as LPS wasttenuated by chronic feeding of DON. Similar observations wereeported by Grenier et al. (2011) who challenged pigs withvalbumin injections fed either a control diet, a DON diet, a fumon-sin B-contaminated diet, or a diet with both mycotoxins. Themmune response (immunoglobulins, lymphocyte proliferation,ro-inflammatory cytokine mRNA expression) of DON-fed pigs wasttenuated, whereas liver lesion score and GGT were significantlyigher and serum albumin lower in DON-fed pigs compared withontrol-fed animals. The difference in liver lesion compared withur study can be explained by the nature of score criteria: Grenier
t al. (2011) analysed at a cellular level, i.e., cytoplasmatic anduclear vacuolisation and apoptotic bodies of hepatocytes, whereasn our study the livers were examined on tissue level. The latter isost likely less sensitive in picking up early changes in hepatocytes.
ers 215 (2012) 193– 200 199
Beyond that, one explanation given for the attenuated immuneresponse was that mycotoxin-feeding prior to immune challengedecreased chemotaxis, as evidenced by low IL-8 expression, andthus impaired recruitment and migration of antigen presentingcells to peripheral lymphoid tissues. This in turn leads to insuf-ficient activation of lymphocytes and could explain the lower IgGresponse. Rotter et al. (1994) also reported an attenuated secondaryantibody response in DON-fed pigs when challenged with sheep redblood cells as compared with control-fed pigs. It is noteworthy, thatpigs fed a DON-contaminated diet showed an attenuated immuneresponse to a subsequent immune stimulus. In contrast to thesestudies the contrary effect was demonstrated by Islam and Pestka(2006) in mice receiving a reversed sequence of treatment. Micewere challenged with Salmonella-LPS in a dose-dependent man-ner and received 24 h later a DON-gavage. Animals showed a ratherdramatic proinflammatory cytokine response to DON when sensi-tised before with LPS compared with mice obtaining DON withoutprevious LPS challenge. Translated into practice, these findingswould imply that pigs exposed to chronic dietary DON exhibitan attenuated response to an immune challenge, whereas alreadyimmuno-challenged animals can react overtly strong to DON expo-sure as compared with their non-challenged counterparts. Whileinterpreting data on mycotoxin and immune-challenge in pigs thisimportant interplay need to be taken into account.
In conclusion, we could demonstrate that DON alone, eitheradministered orally or systemically, did not have an impact onliver morphology and function, whereas LPS challenge in control-fed pigs evoked a strong inflammatory response with concomitantimpaired liver function. Combined infusion of DON and LPS incontrol-fed pigs neither exhibited an additive nor a synergisticeffect, with the exception of bilirubin levels showing an additiveeffect for co-exposure of DON and LPS compared with each toxinalone. Interestingly, DON feeding prior to LPS challenge profoundlyalleviated effects of LPS on the porcine liver morphology and func-tion. Further studies are necessary to explain this modulating DONeffect and to elucidate the mechanisms behind the impact of DONon LPS toxicity.
Conflict of interest statement
The authors declare no conflict of interest.
Acknowledgements
We gratefully acknowledge the Deutsche Forschungsgemein-schaft (DFG, DA 558/1-3 and RO 743/3-2) and the European Union(FP 7 “INTERPLAY”, project-number 227549) for financial support.
The authors also thank Nicola Mickenautsch (Institute of Ani-mal Nutrition, Federal Institute for Animal Health, Braunschweig,Germany), Brigitte Ketzler and Christine Gerlach (Institute ofAnatomy, Medical Faculty, Otto-von-Guericke University, Magde-burg, Germany) for their technical assistance.
References
Andersson, A., Fenhammar, J., Weitzberg, E., Sollevi, A., Hjelmqvist, H., Frithiof, R.,2010. Endothelin-mediated gut microcirculatory dysfunction during porcineendotoxaemia. British Journal of Anaesthesia 105, 640–647.
Bondy, G.S., Pestka, J.J., 2000. Immunomodulation by fungal toxins. Journal of Toxi-cology and Environmental Health Part B: Critical Reviews 3, 109–143.
Bruins, M.J., Soeters, P.B., Lamers, W.H., Deutz, N.E., 2002. l-Arginine supplemen-tation in pigs decreases liver protein turnover and increases hindquarter proteinturnover both during and after endotoxemia. American Journal of Clinical Nutri-tion 75, 1031–1044.
Cybulsky, M.I., Chan, M.K., Movat, H.Z., 1988. Acute inflammation and microthrom-bosis induced by endotoxin, interleukin-1, and tumor necrosis factor and theirimplication in gram-negative infection. Laboratory Investigation 58, 365–378.
Dahm, P.L., Thorne, J., Myhre, E., Grins, E., Martensson, L., Blomquist, S., 1999. Intesti-nal and hepatic perfusion and metabolism in hypodynamic endotoxic shock.
2 y Lett
D
D
D
D
D
E
E
E
E
F
F
G
G
G
G
G
I
I
K
K
K
00 C. Stanek et al. / Toxicolog
Effects of nitric oxide synthase inhibition. Acta Anaesthesiologica Scandinavica43, 56–63.
änicke, S., Brosig, B., Klunker, L.R., Kahlert, S., Kluess, J., Döll, S., Valenta, H., Rothköt-ter, H.J., 2012. Systemic and local effects of the Fusarium toxin deoxynivalenol(DON) are not alleviated by dietary supplementation of humic substances (HS).Food and Chemical Toxicology 50, 979–988.
auphinee, S.M., Karsan, A., 2006. Lipopolysaccharide signaling in endothelial cells.Laboratory Investigation 86, 9–22.
iesing, A.K., Nossol, C., Dänicke, S., Walk, N., Post, A., Kahlert, S., Rothkötter, H.J.,Kluess, J., 2011. Vulnerability of polarised intestinal porcine epithelial cells tomycotoxin deoxynivalenol depends on the route of application. PLoS ONE 6,e17472.
öll, S., Schrickx, J.A., Valenta, H., Dänicke, S., Fink-Gremmels, J., 2009a. Interactionsof deoxynivalenol and lipopolysaccharides on cytotoxicity protein synthesis andmetabolism of DON in porcine hepatocytes and Kupffer cell enriched hepatocytecultures. Toxicology Letters 189, 121–129.
öll, S., Schrickx, J.A., Dänicke, S., Fink-Gremmels, J., 2009b. Interactions of deoxyni-valenol and lipopolysaccharides on cytokine excretion and mRNA expression inporcine hepatocytes and Kupffer cell enriched hepatocyte cultures. ToxicologyLetters 190, 96–105.
FSA, 2004. Opinion of the scientific panel on contaminants in the food chain on arequest from the commission related to deoxynivalenol (DON) as undesirablesubstance in animal feed. The EFSA Journal 73, 1–42.
hrlich, K.C., Daigle, K.W., 1987. Protein synthesis inhibition by 8-oxo-12,13-epoxytrichothecenes. Biochimica et Biophysica Acta 923, 206–213.
riksen, G.S., 2003. Metabolism and toxicity of trichothecenes. Doctoral Thesis, Upp-sala.
riksson, M., Larsson, A., Lundkvist, K., Loof, L., 1998. Endotoxaemic liver injury ina porcine model–relation to tumour necrosis factor-alpha release and survival.Scandinavian Journal of Infectious Diseases 30, 169–172.
arries, F.E., Oslage, H.J., 1961. Zur Technik langfristiger Stoffwechselversuche anwachsenden Schweinen. Zeitschrift fur Tierphysiologie, Tierernährung und Fut-termittelkunde 16, 11–18.
rey, E.A., Finlay, B.B., 1998. Lipopolysaccharide induces apoptosis in a bovineendothelial cell line via a soluble CD14 dependent pathway. Microbial Patho-genesis 24, 101–109.
fE, 2006. Empfehlungen zur Energie- und Nährstoffversorgung von Schweinen.Ausschuss für Bedarfsnormen der Gesellschaft für Ernährungsphysiologie,Frankfurt/Main.
oyarts, T., Dänicke, S., 2006. Bioavailability of the Fusarium toxin deoxynivalenol(DON) from naturally contaminated wheat for the pig. Toxicology Letters 163,171–182.
oyarts, T., Grove, N., Dänicke, S., 2006. Effects of the Fusarium toxin deoxyni-valenol from naturally contaminated wheat given subchronically or as onesingle dose on the in vivo protein synthesis of peripheral blood lympho-cytes and plasma proteins in the pig. Food and Chemical Toxicology 44,1953–1965.
oyarts, T., Dänicke, S., Brüssow, K.P., Valenta, H., Ueberschar, K.H., Tiemann, U.,2007. On the transfer of the Fusarium toxins deoxynivalenol (DON) and zear-alenone (ZON) from sows to their fetuses during days 35–70 of gestation.Toxicology Letters 171, 38–49.
renier, B., Loureiro-Bracarense, A.P., Lucioli, J., Pacheco, G.D., Cossalter, A.M., Moll,W.D., Schatzmayr, G., Oswald, I.P., 2011. Individual and combined effects of sub-clinical doses of deoxynivalenol and fumonisins in piglets. Molecular Nutritionand Food Research 55, 761–771.
shak, K., Baptista, A., Bianchi, L., Callea, F., De, G.J., Gudat, F., Denk, H., Desmet,V., Korb, G., MacSween, R.N., 1995. Histological grading and staging of chronichepatitis. Journal of Hepatology 22, 696–699.
slam, Z., Pestka, J.J., 2006. LPS priming potentiates and prolongs proinflammatorycytokine response to the trichothecene deoxynivalenol in the mouse. Toxicologyand Applied Pharmacology 211, 53–63.
orish, A.A., Arafa, M.M., 2011. Propolis derivatives inhibit the systemic inflam-matory response and protect hepatic and neuronal cells in acute septic shock.
Brazilian Journal of Infectious Diseases 15, 332–338.raft, W., Dürr, U.M., 2005. Klinische Labordiagnostik in der Tiermedizin, sixth ed.Schattauer, Stuttgart/New York.
ullik, K., Brosig, B., Kersten, S., Valenta, H., Diesing, A.-K., Panther, P., Reinhardt, N.,Kluess, J., Rothkötter, H.-J., Breves, G., Dänicke, S., 2012. Interactions between
ers 215 (2012) 193– 200
the Fusarium toxin deoxynivalenol and lipopolysaccharides on tissue proteinsynthesis in pigs. submitted for publication.
Leifsson, P.S., Iburg, T., Jensen, H.E., Agerholm, J.S., Kjelgaard-Hansen, M., Wiinberg,B., Heegaard, P.M., Astrup, L.B., Olsson, A.E., Skov, M.G., Aalbaek, B., Nielsen, O.L.,2010. Intravenous inoculation of Staphylococcus aureus in pigs induces severesepsis as indicated by increased hypercoagulability and hepatic dysfunction.FEMS Microbiology Letters 309, 208–216.
MacLean, L.D., Weil, M.H., Spink, W.W., Visscher, M.B., 1956. Canine intestinal andliver weight changes induced by E. coli endotoxin. Proceedings of the Society forExperimental Biology and Medicine 92, 602–605.
Martens, M., Kumar, M.M., Kumar, S., Goldenberg, M., Kawata, M., Pennycooke, O.,Strande, L., Hadeed, J., Camacho, J., Hewitt, C., Slotman, G.J., 2007. Quantita-tive analysis of organ tissue damage after septic shock. American Surgeon 73,243–248.
Mikami, O., Yamaguchi, H., Murata, H., Nakajima, Y., Miyazaki, S., 2010. Inductionof apoptotic lesions in liver and lymphoid tissues and modulation of cytokinemRNA expression by acute exposure to deoxynivalenol in piglets. Journal ofVeterinary Science 11, 107–113.
Mireles, A.J., Kim, S.M., Klasing, K.C., 2005. An acute inflammatory response altersbone homeostasis body composition, and the humoral immune response ofbroiler chickens. Poultry Science 84, 553–560.
Myers, M.J., Farrell, D.E., Evock-Clover, C.M., McDonald, M.W., Steele, N.C., 1997.Effect of growth hormone or chromium picolinate on swine metabolism andinflammatory cytokine production after endotoxin challenge exposure. Ameri-can Journal of Veterinary Research 58, 594–600.
Nakajima, Y., Mikami, O., Yoshioka, M., Arai, S., Miura, K., Koike, Y., Sato, M.,Kobayashi, M., Nakajima, E., 2000. Involvement of apoptosis in the endotox-emic lesions of the liver and kidneys of piglets. Journal of Veterinary MedicalScience 62, 621–626.
Pestka, J.J., 2008. Mechanisms of deoxynivalenol-induced gene expression and apo-ptosis. Food Additives and Contaminants 25, 1128–1140.
Pestka, J.J., 2010a. Deoxynivalenol: mechanisms of action, human exposure, andtoxicological relevance. Archives of Toxicology 84, 663–679.
Pestka, J.J., 2010b. Deoxynivalenol-induced proinflammatory gene expression:mechanisms and pathological sequelae. Toxins (Basel) 2, 1300–1317.
Roth, R.I., Su, D., Child, A.H., Wainwright, N.R., Levin, J., 1998. Limulusantilipopolysaccharide factor prevents mortality late in the course of endotox-emia. Journal of Infectious Diseases 177, 388–394.
Rotter, B.A., Thompson, B.K., Lessard, M., Trenholm, H.L., Tryphonas, H., 1994. Influ-ence of low-level exposure to Fusarium mycotoxins on selected immunologicaland hematological parameters in young swine. Fundamental and Applied Toxi-cology 23, 117–124.
Rotter, B.A., Prelusky, D.B., Pestka, J.J., 1996. Toxicology of deoxynivalenol (vomi-toxin). Journal of Toxicology and Environmental Health 48, 1–34.
Saetre, T., Hovig, T., Roger, M., Gundersen, Y., Aasen, A.O., 2001. Hepatocellular dam-age in porcine endotoxemia: beneficial effects of selective versus non-selectivenitric oxide synthase inhibition? Scandinavian Journal of Clinical and LaboratoryInvestigation 61, 503–512.
Tiemann, U., Brüssow, K.P., Jonas, L., Pöhland, R., Schneider, F., Dänicke, S., 2006.Effects of diets with cereal grains contaminated by graded levels of two Fusariumtoxins on selected immunological and histological measurements in the spleenof gilts. Journal of Animal Science 84, 236–245.
Tiemann, U., Brüssow, K.P., Dannenberger, D., Jonas, L., Pohland, R., Jager, K., Dänicke,S., Hagemann, E., 2008. The effect of feeding a diet naturally contaminatedwith deoxynivalenol (DON) and zearalenone (ZON) on the spleen and liverof sow and fetus from day 35 to 70 of gestation. Toxicology Letters 179,113–117.
Utili, R., Abernathy, C.O., Zimmerman, H.J., 1977. Endotoxin effects on the liver. LifeSciences 20, 553–568.
Van Amersfoort, E.S., Van Berkel, T.J., Kuiper, J., 2003. Receptors mediators, andmechanisms involved in bacterial sepsis and septic shock. Clinical MicrobiologyReviews 16, 379–414.
Wollenhaupt, K., Dänicke, S., Brüssow, K.P., Tiemann, U., 2006. In vitro and
in vivo effects of deoxynivalenol (DNV) on regulators of cap dependenttranslation control in porcine endometrium. Reproductive Toxicology 21,60–73.Young, L.G., McGirr, L., Valli, V.E., Lumsden, J.H., Lun, A., 1983. Vomitoxin in corn fedto young pigs. Journal of Animal Science 57, 655–664.
