Transcriptomic profile of host response in mouse brain after exposure to plant toxin abrin

Toxicology 299 (2012) 33–43

Contents lists available at SciVerse ScienceDirect

Toxicology journal homepage: www.elsevier.com/locate/toxicol

Transcriptomic profile of host response in mouse brain after exposure to plant toxin abrin A.S. Bala Bhaskar a , Nimesh Gupta b,1 , P.V. Lakshmana Rao a,∗ a b

Division of Pharmacology and Toxicology, Defence Research and Development Establishment, Jhansi Road, Gwalior 474002, India Division of Virology, Defence Research and Development Establishment, Jhansi Road, Gwalior 474002, India

a r t i c l e

i n f o

Article history: Received 15 March 2012 Received in revised form 2 May 2012 Accepted 4 May 2012 Available online 14 May 2012 Keywords: Abrin Microarray qRT-PCR Brain Differential gene expression Ribosome-inactivating proteins

a b s t r a c t Abrin toxin is a plant glycoprotein, which is similar in structure and properties to ricin and is obtained from the seeds of Abrus precatorius (jequirity bean). Abrin is highly toxic, with an estimated human fatal dose of 0.1–1 ␮g/kg, and has caused death after accidental and intentional poisoning. Abrin is a potent biological toxin warfare agent. There are no chemical antidotes available against the toxin. Neurological symptoms like delirium, hallucinations, reduced consciousness and generalized seizures were reported in human poisoning cases. Death of a patient with symptoms of acute demyelinating encephalopathy with gastrointestinal bleeding due to ingestion of abrin seeds was reported in India. The aim of this study was to examine both dose and time-dependent transcriptional responses induced by abrin in the adult mouse brain. Mice (n = 6) were exposed to 1 and 2 LD50 (2.83 and 5.66 ␮g/kg respectively) dose of abrin by intraperitoneal route and observed over 3 days. A subset of animals (n = 3) were sacrificed at 1 and 2 day intervals for microarray and histopathology analysis. None of the 2 LD50 exposed animals survived till 3 days. The histopathological analysis showed the severe damage in brain and the infiltration of inflammatory cells in a dose and time dependent manner. The abrin exposure resulted in the induction of rapid immune and inflammatory response in brain. Clinical biochemistry parameters like lactate dehydrogenase, aspartate aminotransferase, urea and creatinine showed significant increase at 2-day 2 LD50 exposure. The whole genome microarray data revealed the significant regulation of various pathways like MAPK pathway, cytokine-cytokine receptor interaction, calcium signaling pathway, Jak-STAT signaling pathway and natural killer cell mediated toxicity. The comparison of differential gene expression at both the doses showed dose dependent effects of abrin toxicity. The real-time qRT-PCR analysis of selected genes supported the microarray data. This is the first report on host-gene response using whole genome microarray in an animal model after abrin exposure. The data generated provides leads for developing suitable medical counter measures against abrin poisoning. © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Ribosome-inactivating proteins (RIPs) are a group of proteins that share the property of damaging ribosomes in an irreversible manner acting catalytically (Stirpe, 2004). Examples of plant RIPs include abrin, ricin, gelonin, momordin, mistletoe lectin, etc. (Narayanan et al., 2005). Abrin is a heterodimeric glycoprotein

Abbreviations: CNS, central nervous system; GSH, glutathione; PBS, phosphate buffered saline; qRT-PCR, quantitative real-time polymerase chain reaction; MAPK, mitogen activated protein kinase; KEGG, Kyoto Encyclopedia of Genes and Genomes; DAVID, Database for Annotation, Visualization and Integrated Discovery. ∗ Corresponding author. Tel.: +91 751 2233495; fax: +91 751 2341148. E-mail address: [email protected] (P.V.L. Rao). 1 Address: Centre de Recherche des Cordeliers, Equipe 16 – Immunopathology and Therapeutic Immuno-intervention, Universite Pierre et Marie Curie, UMR S 872, 15 rue de I’ Ecole de Medicine, Paris F-75006, France. 0300-483X/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tox.2012.05.005

found in the beans of Abrus precatorius plant commonly known as jequirity bean or rosary pea. Abrin and ricin are structurally and functionally related protein toxins. Toxic protein abrin and ricin are among the most poisonous substances known only next to botulinum toxin. Abrin is a potent biological warfare agent because of its low cost of isolation and ease of use either by aerosolisation as a dry powder or liquid droplets, or by addition to food and water as a contaminant. Abrin belongs to the family of type II ribosome inactivating proteins (RIPs) comprising of A and B subunits cross-linked by a single disulfide bond. The active A-chain moiety has an enzymatic function. The toxicity of A-chain is due to its RNA-N-glycosidase activity, by which it brings about depurination of adenine at 4324 in the 28S rRNA. The end result of this activity is complete inhibition of cellular translation (Irvin, 1995). Polypeptide B chain (38 kDa) is a galactose-specific lectin that facilitates the binding of abrin to cell membranes, while the other chain A enters the cytoplasm. Once in

34

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

the cell, the A chain acts on the 60S ribosomal subunit, preventing binding of elongation factor 2, thus inhibiting protein synthesis and leading to cell death (Sandvig et al., 1976; Irvin, 1995). Though RIPs were identified more than few decades ago, there is still speculation about their biological functions. Recent studies have shown clearly RNAse and DNAse activities of RIPs including ricin and abrin (Barbieri et al., 1997, 2000). There are recent reports on the DNA damage effects of RIPs like ricin and (Rao et al., 2005) abrin (Shih et al., 2001; Bhaskar et al., 2008). Abrin is highly toxic, with an estimated human fatal dose of 0.1–1 ␮g/kg, and has caused death after accidental and intentional poisoning. Abrin can be extracted from jequirity beans using a relatively simple and cheap procedure. This satisfies one criterion of a potential chemical warfare agent, although the lack of large-scale production of jequirity seeds means that quantity is unavailable for ready mass production of abrin for weapons (Dickers et al., 2003). There are several reported cases of ricin poisoning and few cases of abrin poisonings (Rauber and Heard, 1985). A number of poisoning cases due to abrin were reported from India (Sahni et al., 2007; Subrahmanyan et al., 2008; Sahoo et al., 2008; Pillay et al., 2005). Clinical features of abrin poisoning commonly include nausea, vomiting, diarrhea, and abdominal pain. Gastrointestinal bleeding may ensue, with bloody diarrhea (Davis, 1978; Frohne et al., 1984) and/or hematemesis (Dickers et al., 2003). Neurological symptoms like delirium, hallucinations, reduced consciousness and generalized seizures were reported (Frohne et al., 1984). Death of a patient with symptoms of acute demyelinating encephalopathy with gastrointestinal (GI) bleeding due to ingestion of abrin seeds was reported in India (Sahni et al., 2007; Sahoo et al., 2008). The measurement of gene expression levels upon exposure to a chemical can both provide information about the mechanism of action of toxicants and signature pattern of expression (Hamadeh et al., 2002; Lettieri, 2006). Toxicogenomics aim to apply both mRNA and protein expression technology to study chemical effects in biological systems. The development of high quality, commercially available gene arrays has allowed this technology to become a standard tool in molecular toxicology. No reports are available on whole genome array after exposure to abrin. However, there are few reports on genomic expression profiles of another RIP ricin whose mode of action is similar to abrin. The pulmonary genomic profile of BALB/c mice inhalationally exposed to lethal dose of ricin was examined using cDNA arrays (DaSilva et al., 2003). cDNA array results have identified a number of differentially expressed genes responsible for a variety of activities, such as inflammatory processes, tissue and DNA repair, cell migration and structure, cell growth and differentiation and apoptosis (DaSilva et al., 2003). The present study is thus designed to understand the consequences and the molecular mechanism of abrin toxicity in CNS using cDNA array technology. Swiss albino mice were exposed to 1 (2.83 ␮g/kg) and 2 LD50 dose of abrin (5.66 ␮g/kg body weight) by intraperitoneal route. Dose and time dependent effects of abrin on gene expression profile of brain tissue were examined using mouse whole genome array. Considering the number of neurological symptoms reported in poisoning cases, brain was selected for gene expression analysis in the present study. The differential genomic profile correlated with the pathological changes observed in the brain, with the appearance of indicators of inflammation, and an increase in infiltrating leukocytes. The abrin exposure resulted in the induction of a rapid immune and inflammatory response in brain. The comparison of differential gene expression at both the doses showed dose dependent effects of abrin toxicity. The real-time qRT-PCR analysis of few selected genes supported the microarray data. This is the first report of gene expression profiling after abrin exposure.

2. Materials and methods 2.1. Animals Swiss albino mice randomly bred in Institute’s animal facility, weighing between 24 and 26 g were used in this study. The animals were maintained on standard conditions of temperature and humidity. The animals were fed standard pellet diet (Ashirwad Brand, Chandigarh, India). Food and water were given ad libitum. The animals were handled according to the guidelines of CPCSEA (Committee for the Purpose of Control and Supervision on Experiments on Animals) and Institutional Animal Ethics Committee approved the experiment. 2.2. Chemicals Abrin was isolated from A. precatorius seeds as described elsewhere (Kumar et al., 2008). The purified abrin was lyophilized and stored at −80 ◦ C and reconstituted as and when required in PBS. The whole genome mouse 4x44K microarrays and RNA 6000 Nano Lab Chip were purchased from Agilent (Germany) and processed at Genotypic Technology (Bangalore, India). Low RNA Input Fluorescent Linear Amplification Kit was purchased from Agilent (Santa Clara, CA). QIAamp viral RNA mini kit, total RNA extraction kit, RNAlater, Quanti Tect primer assay kit and Quanti Fast one-step real time RT-PCR kit were purchased from Qiagen (Hilden, Germany). 2.3. Animal exposure Animals were divided into three groups of six animals each. Groups 1 and 2 were administered a single dose of 1.0 and 2.0 LD50 of abrin (2.83 and 5.66 ␮g/kg body weight respectively) by intraperitoneal route. Three animals from each group were killed at day 1 and day 2 to harvest brain tissue. The animals were perfused with cold PBS before harvesting the brain tissue. Control animals received the same volume of PBS as the experimental group. 2.4. Biochemical assays The clinical biochemical parameters were assessed in a separate study with six mice for each treatment group (1 and 2 LD50) and time point (1 and 2 days). For biochemical studies, blood was drawn from orbital plexus before sacrificing the animals. Serum harvested from each mouse was used to determine the biochemical markers alanine aminotransferase (ALT), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), urea, uric acid, alkaline phosphatase (ALP), creatinine and albumin. Serum variables were measured using commercially available diagnostic kits following manufacturer’s protocol. 2.5. Sample acquisition, RNA isolation and quality control Brain tissues were collected from control and treated groups of mice at different days of post-exposure. The sections of cerebral cortex of brain tissue were stored in RNAlater (Qiagen, Hilden, Germany) at −70 ◦ C until processed for RNA extraction. Total RNA was extracted using the Qiagen (GmbH, Hilden) RNEasy® Mini kit according to the instructions of the manufacturer. RNA quality and integrity was assessed using RNA 6000 Nano Lab Chip on the 2100 Bioanalyzer (Agilent, Germany) following the manufacturer’s protocol. RNA samples with RIN (RNA Integrity Number) ≥8 were used in all experiments. The pooled RNA samples were used for microarray analysis to reduce the effects of biological variation and to easily find the substantive differences (Kendziorski et al., 2005). However, individual RNA was also verified with real time RT-PCR without pooling and no significant variation in gene expression was observed. 2.6. cDNA microarrays and hybridization Low RNA Input Fluorescent Linear Amplification Kit (Agilent, Santa Clara, CA) was used for labeling. Briefly, both first and second strand cDNA were synthesized by incubating 500 ng of pooled total RNA with 1.2 ␮l of oligo dT-T7 promoter primer in nuclease-free water at 65 ◦ C for 10 min followed by incubation with 4.0 ␮l of 5× first strand buffer, 2 ␮l of 0.1 M DTT, 1 ␮l of 10 mM dNTP mix, 1 ␮l of 200 U/␮l MMLV-RT, and 0.5 ␮l of 40 U/␮l RNaseOUT, at 40 ◦ C for 2 h. Immediately following cDNA synthesis, the reaction mixture was incubated with 2.4 ␮l of 10 mM Cyanine3-CTP or 2.4 ␮l of 10 mM Cyanine-5-CTP (Perkin-Elmer, Boston, MA), 20 ␮l of 4× Transcription buffer, 8 ␮l of NTP mixture, 6 ␮l of 0.1 M DTT, 0.5 ␮l of RNaseOUT, 0.6 ␮l of inorganic pyrophosphatase, 0.8 ␮l of T7 RNA polymerase, and 15.3 ␮l of nuclease-free water at 40 ◦ C for 2 h. Qiagen’s RNeasy mini spin columns were used for purifying amplified cRNA samples. The quantity and specific activity of cRNA was determined by using NanoDrop ND-1000 UV-VIS Spectrophotometer version 3.2.1. Samples with specific activity >8 were used for hybridization. 825 ng of each Cyanine 3 or Cyanine 5 labeled cRNA in a volume of 41.8 ␮l were combined with 11 ␮l of 10× blocking agent and 2.2 ␮l of 25× fragmentation buffer (Agilent), and incubated at 60 ◦ C for 30 min in dark. The fragmented cRNA was mixed with 55 ␮l of 2× hybridization buffer (Agilent). About 110 ␮l of the resulting mixture was applied to the Agilent Whole Genome Mouse 4X 44k Gene Expression Microarray (AMADID: 14868, Agilent Technologies) and hybridized in a two-color comparative format at

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

35

Table 1 Effect of 1 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route on serum biochemical markers after 1 and 2 days post-exposure. Control

1.0 LD50

2.0 LD50

1 day LDH (IU/L) GPT (IU/L) GOT (IU/L) Urea (mg/dl) Uric acid (mg/dl) ALP (IU/L) Creatinine (␮mol/l) Albumin (g/dl)

98.48 5.70 7.96 32.1 5.21 17.48 24.53 2.81

± ± ± ± ± ± ± ±

9.0 0.34 0.36 0.65 0.51 2.1 0.6 0.14

403.6 8.18 14.26 45.5 6.77 24.58 43.70 4.24

2 day ± ± ± ± ± ± ± ±

12.4* 0.37 2.7* 4.1 0.27* 2.0 1.3* 0.16*

491.1 13.16 28.18 51.32 5.77 27.95 49.55 3.24

1 day ± ± ± ± ± ± ± ±

16.6* 0.16 0.36* 2.6* 0.20 4.3* 0.8* 0.13

460.3 9.44 18.88 44.43 8.23 23.4 50.58 4.62

2 day ± ± ± ± ± ± ± ±

24.7* 0.33 0.4* 3.1 0.25* 1.2 1.5* 0.12*

546.4 16.32 33.25 54.32 7.92 30.23 57.55 4.86

± ± ± ± ± ± ± ±

12.7* 0.21 0.54* 5.3* 0.32* 2.4* 1.2* 0.14*

Values are mean ± SE of four mice per group. * Significantly different from control group at p < 0.05 by Dunnet’s test. 65 ◦ C for 17 h in an Agilent microarray hybridization oven (SureHyb: G2534A). After hybridization, slides were washed with Agilent Gene expression wash buffer I for 1 min at room temperature followed by a 1 min wash with Agilent gene expression wash buffer II at 37 ◦ C. Slides were finally rinsed with acetonitrile for cleaning up and drying. Hybridized arrays were scanned at 5 ␮m resolution on an Agilent DNA Microarray Scanner, Model G2565BA. Data extraction from images was done using Feature Extraction software of Agilent. 2.7. Microarray data analysis Feature extracted data was analyzed using GeneSpring Gx v 11.0 software from Agilent. Normalization of the data was done using per spot per chip intensity dependent lowess normalization. Further quality control of normalized data was done using correlation based condition tree to eliminate experimental error. Genes that had ≥2 (up regulated) and ≤−2 (down regulated) fold change were filtered from the data and selected for further analysis. Differentially regulated genes were clustered using gene tree to identify significant gene expression patterns. Ontology based biological analysis was done using Gene Ontology browser in GeneSpring Gx. Statistically significant genes were categorized into KEGG pathways using DAVID (Dennis et al., 2003). 2.8. Real time qRT-PCR The quantitative real-time RT-PCR was carried out to validate the microarray data with three biological replicates without pooling of RNA, using gene-specific primers from Quanti Tect primer assay kit (Qiagen Germany). Quanti Fast one-step RT-PCR kit (Qiagen, Germany) was used for real time PCR and RNA polymerase II (RPII) was used as an endogenous reference gene. Briefly, the reaction mixture consisted of 12.5 ␮l of 2× QuantiFast SYBR Green RT-PCR Master Mix, 2.5 ␮l of 10× Quantitect primer mix, 0.25 ␮l of Quantifast RT Mix, 100 ng (2 ␮l) of template RNA and 8.75 ␮l nuclease free water in 25 ␮l reaction volume. The Roche light cycler 480 was used to monitor the SYBR Green signal at the end of each extension period for 40 cycles. The thermal profile consisted of 10 min of reverse transcription at 50 ◦ C one cycle and 5 min of polymerase activation at 95 ◦ C, followed by 40 cycles of PCR at 95 ◦ C for 10 s, 60 ◦ C for 30 s for combined annealing/extension. The relative quantification levels in expression were determined using the 2nd derivative maximum analysis with the determination of the crossing points for each transcript. Crossing point values for each gene were normalized to the respective crossing point values for the reference gene RP II. Data are presented as normalized ratios of genes along with standard error using the Roche Applied Science E-method (Tellmann and Olivier, 2006). 2.9. Histopathological analysis Control and abrin treated animals were killed at 1 and 2 day post-exposure. Tissue samples of brain was dissected out and fixed in Bouin’s solution. After fixation, small pieces were processed by automated tissue processor (Leica TP1020) dehydrated and embedded in paraffin wax. Multiple sections of 12-␮m thickness were prepared using automatic microtome (Microm HM360) and stained with hematoxylin and eosin in Leica Autostain-XL. Microscopic observation was performed on sections of cerebral cortex under LEICA DMLB microscope and photographs were taken using Leica DC 500 camera. 2.10. Statistical analysis For microarray, sequential Student’s t test (treated versus control) was used to identify genes differentially expressed (p ≤ 0.05) for each group and the experiment was repeated once. The quantitative real-time RT-PCR data and serum biochemical variables data were analyzed by one-way ANOVA followed by Dunnet’s test for comparison between control and abrin treated groups. The level of significance was set at p ≤ 0.05. The data were expressed as mean ± SE of three animals per group. The real-time RT-PCR experiments were repeated twice.

3. Results 3.1. Abrin toxicity Abrin induced a dose dependent decrease in body weight. The food and water intake was decreased. LD50 of abrin used in the study was determined for 7 day period .The animals exposed to 1 LD50 showed slow body movement and loss of body weight and started dying after 5th day of exposure. The surviving animals of 1 LD50 recovered by 14 days. The symptoms were severe in 2 LD50 exposure with bloody diarrhea, piloerection, paralysis and mortality after 50–60 h of post-exposure. None of the animals exposed to 1 and 2 LD50 died within the time points selected for the study. 3.2. Effect on clinical biochemical variables The serum biochemical parameters measured include LDH, ALT, AST, urea, uric acid, ALP, creatinine and albumin and the results are summarized in Table 1. LDH, AST and creatinine levels showed time and dose dependent increase compared to control. Whereas, the effect was significant only at 2 days post-exposure for ALP and urea at both the doses. There was no effect on ALP at any of the dose and time point. 3.3. Histopathological analysis of brain tissue Histopathological analysis of cerebral cortex was performed from control and abrin treated animals to determine the pathological changes after abrin exposure (Fig. 1A–F). The control animals showed normal arrangement of neurons in various layers with blood vessels and glial cells in brain. In 1 LD50 exposure, perivascular cuffing with a large number of leukocyte infiltration and congestion of blood vessels was observed at 1 day with edema and gliosis at day 2. However, in 2 LD50 exposure, cortex edema, mild leukocyte infiltration and mild gliosis was observed at 1 day and neurodegenaration with a large number of necrotic neurons was observed at day 2 (Fig. 1E and F). 3.4. Microarray analysis of brain tissue after abrin exposure The host response was studied at transcriptional level by using whole genome microarray in mouse brain. The data from controland treated animals were compared in both the doses after 1 and 2 day post-exposure. The differential gene expression showed highest number of genes (1195) significantly regulated at day 1 in 2 LD50 dose (Table 2). A Venn diagram showing differentially expressed genes (≤+2.0 fold change or ≤−2.0-fold change; p ≤ 0.05) unique or common at the time points selected is shown in Fig. 2A–D. When 1 day 1 LD50 was compared with day 2 a total of 197 genes showed similar regulation (Fig. 2A). Highest number of 545 genes showed similar regulation in their expression when both the doses were

36

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

Fig. 1. Histopathology of the mouse brain after abrin exposure. Hematoxylin/eosin-stained section of cerebral cortex of mice at different days post abrin exposure by intraperitoneal route. Brain section showing (A) normal arrangement of neurons in various layers with blood vessels and glial cells; (B) cerebellum showing pyknosis of pyramidal neurons at 1 day (1 LD50); (C) perivascular cuffing (thick black arrow) and congestion of blood vessel at 1 day (1 LD50); (D) edema (thick black arrow) and gliosis at 2 day (1 LD50); (E) cortex edema, leukocyte infiltration (thick black arrow) and mild gliosis at 1 day (2 LD50); and (F) neuro degeneration (thick black arrow) and necrotic neurons (thin black arrow) at 2 day (2 LD50) (scale bar = 50 ␮M).

Table 2 Differential gene expression in mouse brain after 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route. Dose

Up regulated

Down regulated

1 LD50 1 day 1 LD50 2 day 2 LD50 1 day 2 LD50 2 day

393 321 523 397

471 393 672 376

compared at day 2 (Fig. 2C). However, only 100 genes showed similar regulation at 2 LD50 when compared with days 1 and 2 (Fig. 2D). Gene Ontology classifications were assigned to all significant genes and classification showed significant regulation of various pathways in both the doses (Table 3). Some of the pathways were differentially regulated in both the doses. The metabolic pathways, NK cell mediated toxicity and cytokine–cytokine receptor interaction were significantly up regulated at 2 day in 2 LD50 dose exposure only (Table 4). More pathways were up regulated at 1 day 2 LD50 compared to 1 LD50. Cancer and MAPK signaling pathways were common at both the doses at 1 day. But other pathways

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

37

Table 3 Differential expression of genes involved in various biological processes in mouse brain after 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route. Biological process

Accession

Gene symbol

Description

Fold change 1 day

2 day

1 LD50 (2.83 ␮g/kg body weight) Transport NM 01970 NC NM 025684 Transport NM 024412 NM 027759 NC NM 007575 Transport BC006868 Cell differentiation NM 025312 Pattern recognition NC NM 026648 Ubiquitin cycle NM 177076 AK076230 Biological process NM 011325 Transport NM 018832 NC AK052813 RNA transport NM 028625 NC AK018619 NC NM 008715 RNA processing Apoptosis NM 178373 Ubiquitin cycle AK087860 Transport NM 013927 Synaptogenesis AK032385

Clcnkb Nepn Clcnka Fsip1 Ciita Dock7 Sostdc1 Lrrc50 Fbxl13 Tra2a Scnn1b Pdzx Ckap5 Lce1a2 Eps15 Ints6 Cidec Rnf20 Cngb3 Erc2

Chloride channel Kb Nephrocan Chloride channel Ka Fibrous sheath-interacting protein 1 Class II transactivator Dedicator of cytokinesis 7 Sclerostin domain containing 1 Leucine rich repeat containing 50 F-box and leucine-rich repeat protein 13 Inferred: RIKEN cdna1500010G04 gene, Sodium channel, nonvoltage-gated 1 beta PDZ domain containing, X chromosome Unclassifiable, Late cornified envelope 1A2 Unclassifiable Integrator complex subunit 6 Cell death-inducing DFFA-like effector c RING finger protein 20 Cyclic nucleotide gated channel beta 3 Similar to ELKS [Homo sapiens]

−0.51 −0.03 0.20 0.81 −1.09 0.91 3.87 2.37 2.00 −0.23 −0.51 −0.09 3.10 −1.34 −0.54 −0.48 1.10 −0.85 −0.48 0.49

3.77 3.25 2.63 2.57 2.35 2.26 1.33 1.08 1.06 −2.08 −2.08 −2.09 −2.23 −2.28 −2.37 −2.44 −2.50 −2.93 −3.25 −6.17

2 LD50 (5.66 ␮g/kg body weight) NM 001080813 Transport Translation NM 172750 Transport NM 170597 NM 001081370 NC AK082520 NC BC034290 NC NM 172966 Ubiquitin cycle NM 025413 NC Protein targeting AK007076 Protein phosphorylation AK052494 Translation NM 007908 Electron transport NM 025489 AK033795 Protein phosphorylation Apoptosis NM 153135 Transport NM 181317 Metabolic process BC051059 Cell cycle BC037635 Protein phosphorylation XM 978094 Transport NM 021301 NC NM 175413 Protein phosphorylation AK005289 Protein phosphorylation NM 177081 Transport AK051911 NM 173052 Protein catablolism AK044856 Transport NC NM 029974 Ubiquitin cycle AK030317 NM 019659 Transport XM 990077 NC NM 172922 Protein phosphorylation AK081332 Oxidative stress AK084681 Protein localization AK076990 Protein kinases AK078837 Transport

Rab11fip1 Adprhl1 Creg2 Shank2 Klhl7 Lrrc29 Sh3rf2 Lce1g Mxra8 D8Ertd82e Eef2k Dnajc5b Myo3b Unc5d Kcns2 Pigg Ube2i Parp8 Slc15a2 Lrrc39 Stk25 Ptpn7 Gga2 Serpinb1b Flna Dcst1 Usp53 Kcnj1 Phf20l1 Ankk1 Dgkk Akap10 Mapk10 Tmed3

RAB11 family interacting protein 1 ADP-ribosylhydrolase like 1 Cellular repressor of E1A-stimulated genes 2 SH3/ankyrin domain gene 2 Hypothetical BTB/POZ domain containing protein Leucine rich repeat containing 29 SH3 domain containing ring finger 2 Late cornified envelope 1G Hypothetical RNA-binding region RNP-1 DNA segment, Chr 8, ERATODoi 82 Eukaryotic elongation factor-2 kinase Dnaj (Hsp40) homolog Unclassifiable, full insert sequence Unc-5 homolog D K+ voltage-gated channel Phosphatidylinositol glycan anchor biosynthesis, class G Ubiquitin-conjugating enzyme E2I Poly (ADP-ribose) polymerase family, member 8 Solute carrier family 15 (H+/peptide transporter) Leucine rich repeat containing 39 Serine/threonine protein kinase 25 Protein tyrosine phosphatase, non-receptor type 7 Unclassifiable, full insert sequence Serine (or cysteine) peptidase inhibitor, clade B Calponin homology DC-STAMP domain containing 1 Hypothetical protein, full insert sequence Potassium inwardly-rectifying channel PHD finger protein 20-like 1, transcript variant 10 Ankyrin repeat and kinase domain containing 1 Hypothetical protein, full insert sequence A kinase (PRKA) anchor protein 10 Mitogen activated protein kinase 10 Unclassifiable, full insert sequence

1.48 1.14 1.08 0.50 −0.14 0.52 −3.39 −0.89 −0.97 −3.09 0.71 0.02 −2.98 −1.69 −1.10 −0.05 0.42 −1.93 −2.95 −0.91 −1.10 −3.05 −0.92 −1.36 −0.92 −0.84 −0.90 −0.89 −0.89 −3.56 −2.36 −1.12 −0.86 −0.92

2.19 2.13 2.08 2.07 2.03 2.02 −2.02 −2.04 −2.05 −2.05 −2.07 −2.08 −2.09 −2.10 −2.11 −2.14 −2.16 −2.16 −2.23 −2.32 −2.35 −2.42 −2.61 −2.63 −2.68 −2.69 −2.72 −2.73 −2.79 −3.07 −3.15 −3.31 −3.99 −4.22

were both dose and time specific (Table 4). The pathways, which were down regulated, are shown in Table 5. More pathways were down regulated at 1 day 1 LD50 compared to 2 LD50 dose. Statistically significant genes were categorized into KEGG pathways using DAVID (Table 6). Pathways involving significant number of genes include cytokine–cytokine receptor interaction, chemokine signaling pathway, cell adhesion molecules and MAPK signaling pathway (Table 6). Differential expression of genes involved in various biological processes like immune response, inflammation, cell adhesion, signal transduction, are listed in Table 6. Most of the genes were involved in inflammation, complement pathway and other adaptive immune response such as Igl-V1, Hfe, Igj, H2-Ab1,

H2-Aa, Cxcl1, Cxcl9, Cxcl10, Cxcl13, Ccl8, Mapk13, Htr5b, Cd74, C3, and C4b. Interestingly, genes involved in cell proliferation were highly up regulated in 1 LD50 exposure only (Table 7). 3.5. Dose dependent differential expression of genes in brain after abrin exposure The host transcriptional profile in brain after abrin exposure with two different doses showed expression of different sets of genes. The immunoglobulin genes were up regulated on day 2 in 1 LD50 exposure (Table S1); however, these genes are not significantly regulated in the 2 LD50 exposure. The immune response

38

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

Table 4 Significantly up regulated pathways in mouse brain after 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route. Pathway

No. of genes. 1 LD50

Pathways in cancer MAPK signaling pathway Asthma Systemic lupus erythematosus Complement and coagulation cascades Allograft rejection Graft-versus-host disease Type 1 diabetes mellitus Antigen processing and presentation Autoimmune thyroid diseases Neuroactive ligand–receptor interaction Regulation of actin cytoskeleton Cell adhesion molecules (CAMs) Metabolic pathways Natural killer cell mediated toxicity Cytokine–cytokine receptor interaction

2 LD50

1 day

2 day

1 day

2 day

9 (0.007) 7 (0.011) – – – – – – – – – – – – – –

– – 7 (0.001) 17 (0.005) 9 (0.021) – – – – – – – – – – –

14 (0.009) 9 (0.003) – – – 15 (0.0002) 14 (0.0005) 14 (0.0001) 18 (0.001) 15 (0.003) 12 (0.009) 9 (0.041) 21 (0.043) – – –

– – – – – – – – – – – – – 26 (1 E−09) 3 (0.022) 9 (0.036)

Figures in parenthesis are p-values.

Table 5 Significantly down regulated pathways in mouse brain after 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route. Pathway

No. of genes 1 LD50

MAPK signaling pathway Pathways in cancer Endocytosis Cytokine–cytokine receptor interaction Calcium signaling pathway Jak-STAT signaling pathway Parkinson’s disease Metabolic pathways Regulation of actin cytoskeleton Huntington’s disease Alzheimer’s disease Wnt signaling pathway Focal adhesion

2 LD50

1 day

2 day

1 day

2 day

8 (0.006) 11 (0.006) 5 (0.01) 9 (0.019) 6 (0.027) 4 (.0034) 3 (0.038) – – – – – –

– – – 6 (0.015) – – – 24 (1 E−09) 5 (0.032) 4 (0.037) 5 (0.041) – –

– – 8 (0.01) 13 (0.017) – – – – – – – 5 (0.016) 10 (0.045)

– 10 (0.011) – 9 (0.049) – – – – 5 (0.017) – – – –

Figures in parenthesis are p-values.

Table 6 Representation of statistically significant genes categorized into different pathways by the KEGG classification system. Mice were exposed to 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin byintraperitoneal route. Pathway

No. of genesa

p-Value

Fold enrichment

Benjam

Cytokine–cytokine receptor interaction Chemokine signaling pathway Intestinal immune network for IgA production NOD-like receptor signaling pathway Graft-versus-host disease Allograft rejection Type I diabetes mellitus Autoimmune thyroid disease Primary immunodeficiency Cell adhesion molecules (CAMs) ECM–receptor interaction Natural killer cell mediated cytotoxicity Toll-like receptor signaling pathway Complement and coagulation cascades Asthma Focal adhesion Hematopoietic cell lineage MAPK signaling pathway Leukocyte transendothelialmigration Axon guidance

25 19 8 8 7 7 7 7 5 10 7 8 7 6 4 10 6 12 7 7

8.02.0E−9 6.12.4E−7 2.62.6E−4 2.66.2E−4 2.22.4E−3 2.22.4E−3 2.23.7E−3 2.27.1E−3 1.61.0E−2 3.21.1E−2 2.21.4E−2 2.62.6E−2 2.23.0E−2 1.93.3E−2 1.34.3E−2 3.24.7E−2 1.95.0E−2 3.85.1E−2 2.26.4E−2 2.29.2E−2

4.3 4.3 6.2 5.4 5.0 5.0 4.6 4.0 5.8 2.7 3.5 2.7 2.9 3.3 5.0 2.1 3.0 1.9 2.4 2.2

1.9E−7 1.1E−5 8.4E−3 1.5E−2 4.6E−2 4.6E−2 5.7E−2 9.3E−2 1.2E−1 1.1E−1 1.3E−1 2.1E−1 2.2E−1 2.2E−1 2.6E−1 2.6E−1 2.6E−1 2.6E−1 3.0E−1 3.9E−1

a Genes heading indicates number of genes mapped to an ontology category. p-Value derived from Fisher exact test and Benjam indicates p-value after application of Benjamini multiple test correction.

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

39

Table 7 Differential expression of genes in mouse brain after 1 LD50 (2.83 ␮g/kg) and 2 LD50 (5.66 ␮g/kg) of abrin exposure by intraperitoneal route. Accession no.

Gene symbol

Description

Fold change 1 LD50

2 LD50

1 day

2 day

1 day

2 day

Autoimmune regulator Tnf superfamily, member 8 2 -5 -oligoadenylatesynthetase 2 Immunoglobulin lambda chain, variable 1 Hemochromatosis protein Immunoglobulin joining chain Histocompatibility 2, class II antigen A, beta 1 Histocompatibility 2, class II antigen A, alpha Histocompatibility 2, t region locus 24

2.53 3.71 −1.50 −0.83 3.34 0.01 −0.60 0.47 −1.47

1.50 3.56 −2.12 4.29 2.79 4.17 2.91 4.27 −2.66

3.90 4.26 −0.45 2.93 2.03 3.85 3.56 4.36 −2.92

3.53 2.71 −2.07 1.74 1.73 1.47 1.10 −0.33 −0.64

Cxcl1 Cxcl9 Cxcl10 Cxcl13 Chi3l3 Ccl8 Prok2 Tac1

Chemokine ligand 1 Chemokine ligand 9 Chemokine ligand 10 Chemokine ligand 13 Chitinase 3-like 3 Chemokine ligand 8 Prokineticin 2 Tachykinin 1

2.03 −1.32 1.72 −0.47 0.07 −0.34 −2.43 −2.70

2.05 3.00 2.82 4.04 3.74 2.95 −1.05 −0.57

4.67 3.13 4.67 2.86 3.82 3.27 −2.40 −2.94

4.28 −0.16 1.95 −2.17 3.80 −0.06 −1.36 −1.89

Cell adhesion NM 010493 NM 011582 NM 175485 ENSMUST00000050239

Icam1 Thbs4 Prtg Pcdh11x

Intercellular adhesion molecule Thrombospondin 4 Protogenin homolog Protocadherin 11

2.69 3.05 0.31 −1.38

0.96 1.44 2.89 −2.54

2.42 2.61 2.73 −0.50

3.30 2.24 1.65 −2.39

Chemotaxis NM 013650 NM 0091145 NM 009911

S100a8 S100a9 Cxcr4

Calcium binding protein A8 Calcium binding protein A9 Chemokine (C-X-C motif) receptor 4

1.13 1.63 1.31

3.44 3.22 2.31

2.19 2.91 3.27

3.44 3.22 −0.53

Signal transduction NM 011950 NM 008152 NM 017466 NM 010483 NM 011113 NM 009924 NM 009519

Mapk13 Gpr65 Ccrl2 Htr5b Plaur Cnr2 Wnt11

Mitogen activated protein kinase 13 G-protein coupled receptor 65 Chemokine receptor-like 2 5-hydroxytryptamine (serotonin) receptor 5B Plasminogen activator, urokinase receptor Cannabinoid receptor 2 (macrophage) Wingless-related MMTV integration site 11

2.10 1.12 1.14 3.13 2.21 0.85 1.57

1.71 3.54 2.19 0.68 0.60 2.18 4.77

0.97 2.19 2.71 3.32 2.06 1.18 2.25

2.72 2.50 3.84 3.24 3.42 2.08 2.83

Cell proliferation BC030896

Pdgfd

Platelet-derived growth factor, D polypeptide

4.36

3.81

−0.41

2.75

Biological process NM 201639 NM 010545 NM 01970

Dmn Cd74 Clcnkb

Desmuslin Cd74 antigen Chloride channel Kb

5.59 −0.86 −0.51

3.77 3.03 3.77

5.73 3.39 3.45

4.67 0.17 3.12

Extra cellular matrix BC030317 NM 029796

Lrrc17 Lrg1

Leucine rich repeat containing 17 Leucine-rich alpha-2-glycoprotein 1

2.61 2.11

3.73 1.64

2.62 2.41

2.51 2.24

Complement pathway NM 009778 NM 009780 NM 007576

C3 C4b C4bp

Complement component 3 Complement component 4B Complement component 4 binding protein

0.34 1.13 2.76

2.80 2.34 0.60

2.70 2.42 3.15

0.67 1.63 0.07

Immune response NM 009646 NM 009403 NM 145227 AK008094 AJ306425 NM 152839 NM 207105 NM 010378 NM 008207

Aire Tnfsf8 Oas2 Igl-V1 Hfe Igj H2-Ab1 H2-Aa H2-T24

Inflammation NM 008176 NM 008599 NM 021274 NM 018866 NM 009892 NM 021443 NM 015768 NM 009311

genes like Csf3, Ada, Cfb showed up regulation on day 2 in 2 LD50 (Table S1). Most of the genes related to cell adhesion (Table S2), chemotaxis (Table S3) and inflammation (Table S4) were up regulated on day 2 in 2 LD50 exposure and only few genes showed up regulation in 1 LD50. Chemokine receptor (CxCr4) showed time dependent up regulation at 1 LD50 whereas, CxCl12 showed time dependent down regulation (Table S3). The in depth analysis of significantly regulated genes showed expression of various cell surface receptors in both the doses (Table S5). Most of these genes belong to the olfactory receptor family. Other receptors like Pira3, Cd244, and Rtp1 showed up regulation in 1 LD50 exposure only. Serotonin and calcitonin receptor showed down regulation in 2 LD50 exposure only. The genes related to signal transduction were also significantly regulated at both the doses (Table S6). Genes like Vav2 and

Angpt 1 showed up regulation on day 2 in 1 LD50 and desmokynin and Map3k6 were up regulated in 2 LD50 exposure (Table S6). The exposure with both the doses showed differential expression of transcriptional regulatory genes. However, the genes were not similar in both the doses (Table S7). The genes related to the proteolysis gene family were also found to be significantly regulated in both the doses (Table S8). 3.6. Validation of microarray data by real-time qRT-PCR The validation of microarray data was performed by qRTPCR with the expression analysis of selected representative genes involved in immune response (CxCl9, CxCl10), inflammatory response (CxCl1), signal transduction (Cclr2), cell adhesion (ICAM1)

40

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

Fig. 2. Comparison of gene expression at different doses and time points after abrin exposure. A Venn diagram depicting the differentially expressed genes (≥2.0-fold change or ≤−2.0-fold change; p ≤ 0.05) unique or common at (A) 1 LD50 1 day versus 2 day; (B) 2 LD50 1 day versus 2 day; (C) 1 LD50 1 day versus 2 LD50 1 day; (D) 1 LD50 2 day versus 2 LD50 2 day. Green and red indicates down regulated and up regulated genes respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

and cell receptor (CalCr). The data demonstrate that the overall results of qRT-PCR were consistent with those of the microarray analysis. The qRT-PCR results reveal the same relative regulation of transcription as the microarray data, and confirm that expression of the selected genes was significantly regulated in response to abrin exposure (Fig. 3). Fold increase in the case of Icam1 and CxCl1 was more pronounced by qRT-PCR method than by microarray. 4. Discussion Abrin and ricin are structurally and functionally related protein toxins commonly called as ribosome inactivating proteins. Abrin

shows significant similarities to ricin at the sequence level as well as the structural level, but abrin is many times more potent than ricin. Abrin and ricin toxins are known to cause severe intestinal toxicity with symptoms of abdominal pain, vomiting and bloody diarrhea. At cellular level, abrin inhibits protein synthesis, thereby causing cell death. Previous studies with abrin showed vascular leak syndrome as one of the feature of abrin poisoning, in which there is endothelial cell damage, and an increase in capillary permeability with protein leakage and tissue edema (Dickers et al., 2003). Few reports are available on direct toxic effect of abrin on the central nervous system (Deshpande et al., 1961; Frohne et al., 1984), but this has not been substantiated. Death of a patient with symptoms of acute demyelinating encephalopathy due to ingestion of abrin seeds was reported (Sahni et al., 2007). Gene expression analysis using DNA microarrays provide sensitive indicators of exposure to toxicants and data concerning biochemical mechanisms and signaling pathways, allowing generation of transcriptional signatures (Morgan, 2002). No antidote or vaccine is currently available for abrin. The host defensive mechanism and the activation of gene regulatory circuits in response to abrin toxicity is not explored yet and relatively little is known about its mechanism of toxicity in lethal cases. Information on host response is critical for the development of an effective prophylactic and therapeutic countermeasures to abrin. Our study examined both dose and time-dependent transcriptional responses induced by abrin in the adult mouse after intraperintoneal exposure. In this study, abrin was isolated from the seeds and purified using column chromatography before intraperitoneal administration to mice. After intraperitoneal exposure, animals started showing toxicity symptoms on 2 day postexposure. At 2 LD50 dose of abrin all the animals succumbed to death between 50 and 60 h of post-exposure. The brain histopathology results showed abrin induced leukocyte infiltration. Congestion of blood vessels with edema and gliosis and degeneration of number of neurons at day 2. This is in agreement with human fatal case of poisoning where MRI of the brain scan showed evidence of encephalitis (Sahni et al., 2007). Demyelination is immunemediated and abrin is a well-known immune modulator and stimulator (Griffiths, 2011). Our unpublished results have shown the alteration of blood–brain barrier permeability after abrin exposure in mice. The clinical biochemical parameters like LDH, AST, ALP and creatinine levels showed significant elevation compared to control indicating liver and kidney damage. Cytokine-induced neutrophil chemo attractant-1 (CINC-1) is the major chemokine involved in neutrophil recruitment to the brain and spinal cord. CINC-1 functions as an acute phase protein and after induction of a focal inflammatory lesion in the brain, there is rapid hepatic and serum CINC-1 induction, which is associated with increases in neutrophil numbers within the liver and within the circulation (Campbell et al., 2003). An earlier report suggests that autophagy is important for the toxin-induced cell lysis (Sandvig and van Deurs, 1992). The increased expression of Aire in highly lethal dose suggests that this autoimmune regulator may be involved in the abrin induced cell lysis via autophagy (Shi et al., 2010). The microarray results clearly showed a number of differentially expressed genes responsible for various activities, such as immune response, cell adhesion, chemotaxis, inflammatory processes, transcription and signal transduction. There are few reports on the mechanism of abrin-induced apoptosis in vitro (Griffiths et al., 1987; Narayanan et al., 2004) and DNA damage (Bhaskar et al., 2008). No published information is available on gene expression profiling after abrin intoxication. We hypothesize that the genes which were expressed in 2 LD50 dose but which were not regulated or showed opposite regulation in 1 LD50 dose may have significant role in abrin induced severe toxicity and neuronal consequences leading to death of the animal. The significant expression

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

40

8

2 LD50

6

4

2 LD50

20

10

2

0

1 LD50

30

Fold Change / Cxcl1

Fold Change / Cxcl10

1 LD50

41

0

1 2 1 2 Time [Day Post Treatment]

7

8

1 LD50

2 LD50

2 1 2 1 Time [Day Post Treatment] 1 LD50

2 LD50

Fold Change /Icam1

Fold Change / Ccrl2

6 5 4 3 2

6

4

2

1 0

0

2 1 2 1 Time [Day Post Treatment]

1 2 1 2 Time [Day Post Treatment] 8

1 LD50

10

2 LD50

1 LD50

2 LD 50

Fold Change / Calcr

Fold Change / Cxcl9

6 4 2 0

5

0

-5

-2 -4

1 2 1 2 Time [Day Post Treatment] Microarray Data

-10

2 1 2 1 Time [Day Post Treatment] Real Time PCR Data

Fig. 3. Validation of microarray data of selected genes by qReal-Time RT-PCR. The mRNA expression levels of selected genes were determined to validate the expression data of microarray analysis. Polr2a (RNA polymerase II) was used as housekeeping gene. The values are mean ± SE of three biological replicates without pooling RNA. *Significantly different from control mouse at p < 0.05 by Dunnet’s test.

of immunoglobulin genes and the presence of large number of infiltrating cells in 1 LD50 exposure compared to 2 LD50 exposures showed strong cellular response in 1 LD50 and could be an important factor in the survival of these animals.

In the present study intracellular adhesion molecule, ICAM1 levels were significantly up regulated at 2 LD50 exposure on both the days. ICAM1 expression is critical in T cells and other cell types for development of demyelinating disease. ICAM1 also plays novel role

42

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43

in modulating GSH metabolism and raise the hypothesis that this adhesion molecule controls endothelial redox status under basal and inflammatory conditions (Pruitt et al., 2007). Our earlier study has shown abrin induced oxidative stress leading to GSH depletion and DNA damage in U937 cells (Bhaskar et al., 2008). A possible neuroprotective role for CxCl1 during the course of autoimmune demyelination is also reported. The opposite regulation of the chemotactic genes such as Cxcl9, Cxcl10, and Cxcr4 in the two lethal doses indicates the important role of the cellular infiltration in abrin induced severe toxicity. The entry of lymphocytes into the central nervous system is normally shielded by the blood–brain barrier, which becomes compromised in several pathological conditions. Natural killer (NK) cells are recruited to the inflamed brain by CX3 C-chemokine ligand 1 (CX3 CL1) produced by neurons. Brainspecific cell types and neurotransmitters alter the features of NK cells that migrate from the periphery (Shi et al., 2011). The inverse regulation of these chemotactic genes may be playing important role in non-lethal dose rather than at lethal dose. These chemotactic genes are playing important role in the cellular trafficking (Adler and Rogers, 2005) and this cellular infiltration might be playing key role in the protection against severe toxicity. It can also be hypothesized that the robust expression of these genes on very first day in 2 LD50 exposures was responsible for the aggravated infiltration of the immune cells, which induces the immunopathology and severe neuronal consequences such as in the case of viral encephalitis (Gupta and Lakshmana Rao, 2011). The evidence of hemorrhage with increased protein and lymphocytes in cerebrospinal fluid was reported in an human abrin poisoning case (Pillay et al., 2005). Recent reports suggest that the chemokine system plays a crucial role in brain development and function by interacting with neurotransmitter system. The importance of the chemokine system in the brain homeostatic and perturbed situation is currently gaining importance (Adler and Rogers, 2005). Chemokines CxCl1 was significantly up regulated at both 1 and 2 LD50 dose of abrin at both the time points. Chemokines CxCl1 and CxCl2 are up regulated in the central nervous system (CNS) during multiple sclerosis and its animal model, experimental autoimmune encephalomyelitis (EAE) (Carlson et al., 2008). Blockade or genetic silencing of CxCr2, a major receptor for these chemokines in mice, abrogates blood–brain barrier (BBB) breakdown, CNS infiltration by leukocytes and in the clinical deficits during the presentation and relapse of EAE (Carlson et al., 2008). Results of our study had shown time dependent up and down regulation of CxCr4 and CxCl12 respectively after abrin exposure. CxCl12 and its receptor CxCr4 are required for normal brain development, based on the abnormal neuronal organization in the cerebellum of CxCr4 or CxCl12 deficient mice (Ma et al., 1998). It is likely that a role for CxCl12 persists through adult life of mouse by virtue of the role of this chemokine in promoting neuronal survival and apoptosis. The significant regulation of the pathways that are involved in neurodegenerative disorders such as Parkinson’s, Alzheimer’s and Huntington’s disease during abrin exposure may be because of the similar kind of host response to these pathological conditions in CNS. However, this observation raised question for the use of abrin in the development of experimental models with neurodegenerative disorders. Regulation of genes involved in development and growth like Wnt11 and Pdgfd is showing the survival response from host. The cannabinoid receptors Cnr2 are now known to attenuate microglial activation and protection in neurodegeneration (Palazuelos et al., 2009; Atwood and Mackie, 2010). It can be hypothesized from the data that host responded to both the doses toward survival but with varying degree of expression and there were some specific set of genes in highly lethal dose that played detrimental role. The results of present study show that abrin induces neuro-inflammation after

1 or 2 days of exposure. However, it is not clear whether the neuroinflammation is due to direct effect after entry of abrin into brain or it is due to abrin-induced alterations in host periphery such as inflammation, modulation of blood–brain barrier and cellular trafficking in the brain. There was no microarray data available for the RIPs induced host responses in the brain of an experimental animal model. In the absence of previous information on whole genome microarray data after abrin exposure in any animal model, our results are thus compared with the available case reports and the clinical symptoms. In conclusion, results of our study showed that abrin exposure resulted in rapid immune and inflammatory response in brain. A dose and time dependent transcriptional changes was caused by abrin exposure. Microarray data analysis indicated the involvement of immunologically important genes influencing neuroinflammation, cell migration and chemotaxis. Immunopathological changes like demyelination may lead to severe neurotoxicity. Further studies are required to identify whether this expression profile in brain is unique to abrin intoxication and also gene expression changes in other target organs so that new strategies can be developed for possible medical countermeasures against abrin intoxication. Conflict of interest None declared. Acknowledgements We thank Dr. M.P. Kaushik, Director, Defence Research and Development Establishment for offering all facilities and support required for this study. Mr. Nimesh Gupta is recipient of DRDO Senior research fellowship. We are thankful to Dr. Vinay Lomash for histopathology analysis. This work was supported by the grant from Ministry of Defence, Government of India. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tox.2012.05.005. References Adler, M.W., Rogers, T.J., 2005. Are chemokines the third major system in the brain? J. Leukoc. Biol. 78, 1204–1209. Atwood, B.K., Mackie, K., 2010. CB2 : a cannabinoid receptor with an identity crisis. Br. J. Pharmacol. 160, 467–479, http://dx.doi.org/10.1111/j.14765381.2010.00729.x. Barbieri, L., Valbonesi, P., Bonora, E., Gorini, P., Bolognesi, A., Stirpe, F., 1997. Polynucleotide:adenosine glycosidase activity of ribosome-inactivating proteins: effect on DNA, RNA and poly(A). Nucleic Acids Res. 25, 518–522. Barbieri, L., Bolognesi, A., Valbonesi, P., Polito, L., Olivieri, F., Stirpe, F., 2000. Polynucleotide:adenosine glycosidase activity of immunotoxins containing ribosome-inactivating proteins. J. Drug Target. 8, 281–288. Bhaskar, A.S.B., Deb, U., Kumar, O., Lakshmana Rao, P.V., 2008. Abrin induced oxidative stress mediated DNA damage in human leukemic cells and its reversal by N-acetylcysteine. Toxicol. In Vitro 22, 1902–1908. Campbell, S.J., Hughes, P.M., Iredale, J.P., Wilcockson, D.C., Waters, S., Docagne, F., Perry, V.H., Anthony, D.C., 2003. CINC-1 is an acute-phase protein induced by focal brain injury causing leukocyte mobilization and liver injury. FASEB J. 17, 1168–1170. Carlson, T., Kroenke, M., Rao, P., Lane, T.E., Segal, B., 2008. The Th17-ELR+CXC chemokine pathway is essential for the development of central nervous system autoimmune disease. J. Exp. Med. 205, 811–823. DaSilva, L., Cote, D., Roy, C., Martinez, M., Duniho, S., Pitt, M.L., Downey, T., Dertzbaugh, M., 2003. Pulmonary gene expression profiling of inhaled ricin. Toxicon 41, 813–822. Davis, J.H., 1978. Abrus precatorius (rosary pea). The most common lethal plant poison. J. Florida Med. Assoc. 65, 188–191. Dennis Jr., G., Sherman, B.T., Hosack, D.A., Yang, J., Gao, W., Lane, H.C., Lempicki, R.A., 2003. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 4 (5), P3. Deshpande, V.R., Dubey, P.N., Joshi Rao, M.K., 1961. Toxicity of Abrus precatorius. Indian J. Med. Sci. 15, 195–197.

A.S.B. Bhaskar et al. / Toxicology 299 (2012) 33–43 Dickers, K.J., Bradberry, S.M., Rice, P., Griffiths, G.D., Vale, J.A., 2003. Abrin poisoning. Toxicol. Rev. 22, 137–142. ¨ H.J., 1984. Die Paternostererbse: keineswegs harmFrohne, D., Schmoldt, A., Pfander, los. Dtsch. Apoth. Ztg. 124, 2109–2113. Griffiths, G.D., 2011. Understanding ricin from a defensive viewpoint. Toxins 3, 1373–1392. Griffiths, G.D., Leek, M.D., Gee, D.J., 1987. The toxic plant proteins ricin and abrin induce apoptotic changes in mammalian lymphoid tissues and intestine. J. Pathol. 151, 221–229. Gupta, N., Lakshmana Rao, P.V., 2011. Transcriptomic profile of host response in Japanese encephalitis virus infection. Virol. J. 8, 92. Hamadeh, H.K., Amin, R.P., Paules, R.S., Afshari, C.A., 2002. An overview of toxicogenomics. Curr. Issues Mol. Biol. 4, 45–56. Irvin, J.D., 1995. In: Chessin, M., De Borde, D., Zipf, A. (Eds.), Antiviral Proteins in Higher Plants. CRC Press, Inc., Boca Raton, FL, pp. 5–94. Kendziorski, C., Irizarry, R.A., Chen, K.S., Haag, J.D., Gould, M.N., 2005. On the utility of pooling biological samples in microarray experiments. Proc. Natl. Acad. Sci. USA 102, 4252–4257. Kumar, O., Kannoji, A., Jayaraj, R., Vijayaraghavan, R., 2008. Purification and characterization of abrin toxin from white Abrus precatorius seeds. J. Cell Tissue Res. 8, 1243–1248. Lettieri, T., 2006. Recent applications of DNA microarray technology to toxicology and ecotoxicology. Environ. Health Perspect. 114, 4–9. Ma, Q., Jones, D., Borghesani, P.R., Segal, R.A., Nagasawa, T., Kishimoto, T., Bronson, R.T., Springer, T.A., 1998. Impaired B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in CXCR4- and SDF-1-deficient mice. Proc. Natl. Acad. Sci. USA 95, 9448–9453. Morgan, K.T., 2002. Gene expression analysis reveals chemical-specific profiles. Toxicol. Sci. 67 (2), 155–156. Narayanan, S., Surolia, A., Karande, A.A., 2004. Ribosome inactivating protein and apoptosis: abrin causes cell death via mitochondrial pathway in Jurkat cells. Biochem. J. 377 (1), 233–240. Narayanan, S., Surendranath, K., Bora, N., Surolia, A., Karande, A.A., 2005. Ribosome inactivating proteins and apoptosis. FEBS Lett. 579, 1324–1331. Palazuelos, J., Aguado, T., Pazos, M.R., Julien, B., Carrasco, C., Resel, E., Sagredo, O., Benito, C., Romero, J., Azcoitia, I., Fernández-Ruiz, J., Guzmán, M., Galve-Roperh, I.,

43

2009. Microglial CB2 cannabinoid receptors are neuroprotective in Huntington’s disease excitotoxicity. Brain 132 (11), 3152–3164. Pillay, V.V., Bhagyanathan, P.V., Krishnaprasad, R., Rajesh, R.R., Vishnupriya, N., 2005. Poisoning due to white seed variety of Abrus precatorius. J. Assoc. Physicians India 53, 317–319. Pruitt, H.M., Langston, W., Kevil, C.G., Patel, R.P., 2007. ICAM-1 cross-linking stimulates endothelial glutathione synthesis. Antioxid. Redox Signal. 9 (January (1)), 159–164. Rao, P.V.L., Jayaraj, R., Bhaskar, A.S.B., Kumar, O., Bhattacharya, R., Saxena, P., Dash, P.K., Vijayaraghavan, R., 2005. Mechanism of ricin induced apoptosis in human cervical cancer cells. Biochem. Pharmacol. 69, 855–865. Rauber, A., Heard, J., 1985. Castor bean toxicity re-examined: a new perspective. Vet. Hum. Toxicol. 27, 490–502. Sahni, V., Agarwal, S.K., Singh, N.P., Sikdar, S., 2007. Acute demyelinating encephalitis after jequirity pea ingestion (Abrus precatorius). Clin. Toxicol. (Phila) 45, 77–79. Sahoo, R., Hamide, A., Amalnath, S.D., Narayana, B.S., 2008. Acute demyelinating encephalitis due to Abrus precatorius poisoning – complete recovery after steroid therapy. Clin. Toxicol. 46, 1071–1073. Sandvig, K., van Deurs, B., 1992. Toxin-induced cell lysis: protection by 3methyladenine and cycloheximide. Exp. Cell Res. 200, 253–262. Sandvig, K., Olsnes, S., Pihl, A., 1976. Kinetics of binding of the toxic lectins abrin and ricin to surface receptors of human cells. J. Biol. Chem. 251, 3977–3984. Shi, L., Hu, L.H., Li, Y.R., 2010. Autoimmune regulator regulates autophagy in THP-1 human monocytes. Front. Med. China 4 (September (3)), 336–341. Shi, F.D., Ljunggren, H.G., La Cava, A., Van Kaer, L., 2011. Organ-specific features of natural killer cells. Nat. Rev. Immunol. 11 (10), 658–671. Shih, S.F., Wu, Y.H., Hung, C.H., Yang, H.Y., Lin, J.Y., 2001. Abrin triggers cell death by inactivating a thiol specific antioxidant protein. J. Biol. Chem. 276, 21870–21877. Stirpe, F., 2004. Ribosome-inactivating proteins. Toxicon 44, 371–383. Subrahmanyan, D., Mathew, J., Raj, M., 2008. An unusual manifestation of Abrus precatorius poisoning: a report of two cases. Clin. Toxicol. (Phila) 46, 173–175. Tellmann, G., Olivier, G., 2006. Light Cycler 480 Real-Time PCR System: innovative solutions for relative quantification. Biochemica 4, 16–18.