Microarray analyses identify molecular biomarkers of Atlantic salmon macrophage and hematopoietic kidney response to Piscirickettsia salmonis infection

Matthew L. Rise1, Simon R. M. Jones2, Gordon D. Brown3, Kristian R. von Schalburg3, William S. Davidson4 and Ben F. Koop3

1 Great Lakes Wisconsin Aquatic Technology and Environmental Research (WATER) Institute, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin
2 Pacific Biological Station, Fisheries and Oceans Canada, Nanaimo
3 Centre for Biomedical Research, University of Victoria, Victoria
4 Department of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, British Columbia, Canada


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Piscirickettsia salmonis is the intracellular bacterium that causes salmonid rickettsial septicemia, an infectious disease that kills millions of farmed fish each year. The mechanisms used by P. salmonis to survive and replicate within host cells are not known. Piscirickettsiosis causes severe necrosis of hematopoietic kidney. Microarray-based experiments with QPCR validation were used to identify Atlantic salmon macrophage and hematopoietic kidney genes differentially transcribed in response to P. salmonis infection. Infections were confirmed by microscopy and RT-PCR with pathogen-specific primers. In infected salmon macrophages, 71 different transcripts were upregulated and 31 different transcripts were downregulated. In infected hematopoietic kidney, 30 different transcripts were upregulated and 39 different transcripts were downregulated. Ten antioxidant genes, including glutathione S-transferase, glutathione reductase, glutathione peroxidase, and cytochrome b558 {alpha}- and ß-subunits, were upregulated in infected macrophages but not in infected hematopoietic kidney. Changes in redox status of infected macrophages may allow these cells to tolerate P. salmonis infection, raising the possibility that treatment with antioxidants may reduce hematopoietic tissue damage caused by this rickettsial infection. The downregulation of transcripts involved in adaptive immune responses (e.g., T cell receptor {alpha}-chain and C-C chemokine receptor 7) in infected hematopoietic kidney but not in infected macrophages may contribute to infection-induced kidney tissue damage. Molecular biomarkers of P. salmonis infection, characterized by immune-relevant functional annotations and high fold differences in expression between infected and noninfected samples, may aid in the development of anti-piscirickettsial vaccines and therapeutics.

Salmo salar; salmonid rickettsial septicemia; antioxidant


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
THE BACTERIUM Piscirickettsia salmonis is a gram-negative, obligate intracellular bacterium that causes salmonid rickettsial septicemia, an emerging infectious disease. P. salmonis is pathogenic to coho (Oncorhynchus kisutch), Atlantic (Salmo salar), and chinook (Oncorhynchus tshawytscha) salmon, and rainbow trout (Oncorhynchus mykiss) (19, 27). In 1989, outbreaks of P. salmonis infection at some sites were associated with the mortality of up to 90% of farmed salmon in Chile (6, 16, 17). The organism continues to kill millions of farmed salmonids each year in Chile (38). P. salmonis or rickettsia-like organisms have also been identified in farmed salmonids in Atlantic and Pacific Canada (8, 12), Ireland (37), and Norway (33). Piscirickettsiosis is therefore a serious threat to the global salmon aquaculture industry. Clinical signs (symptoms) of piscirickettsiosis include severe necrosis and inflammation of hematopoietic kidney (19), anemia, enlarged spleen, and liver lesions (6). A rickettsia-like organism isolated from diseased kidney was shown using Koch’s postulates to be the causative agent of piscirickettsiosis (16, 19). Infections with rickettsia-like organisms have been identified in non-salmonids, including tilapia (11), blue-eyed plecostomus (24), and white sea bass (10).

P. salmonis occurs within membrane-bound vacuoles in various cell types (31) including phagocytes. The mechanisms used by the bacterium for survival and replication within host cells are not known. Microarray-based gene expression studies of cultured mammalian macrophages infected with intracellular, nonrickettsial bacterial pathogens identified suites of responsive host genes, pointing to cellular processes (e.g., inflammation, signal transduction, apoptosis) that may be altered to prevent pathogen destruction (13, 15). The work contained herein, utilizing genomics resources developed by the Genomic Research on Atlantic Salmon Project (GRASP) (36), is the first microarray-based study that considers and compares the gene expression responses of isolated macrophages and their tissue of origin, the hematopoietic salmon kidney, following infection with P. salmonis (Fig. 1, A and B). A comparison of in vivo and in vitro expression studies leads to hypotheses regarding host molecular pathways potentially altered by P. salmonis infection. Candidate molecular biomarkers of P. salmonis infection, derived from lists of informative, functionally annotated transcripts with relatively high levels of induction or suppression, may aid in the development of anti-piscirickettsial vaccines and therapeutics.



View larger version (68K):
[in this window]
[in a new window]
 
Fig. 1. A: overview of macrophage (i) and hematopoietic (head) kidney (ii) microarray studies. Arrows between samples symbolize microarrays, with the base of each arrow on the Cy3-labeled target and the arrowhead on the Cy5-labeled target. B: Venn diagrams show intersection between the macrophage and head kidney microarray studies, with 7 transcripts greater than 2-fold upregulated in both studies, and 4 transcripts greater than 2-fold downregulated in both studies. C: RT-PCR with Piscirickettsia salmonis-specific primers (30) confirms infection of cells and tissues. D: P. salmonis within vacuole (arrow) of Atlantic salmon macrophage. Giemsa-stained imprint of hematopoietic kidney. Scale bar is 10 µm.

 

    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
P. salmonis infections in the head kidney study.
P. salmonis was isolated from a moribund farmed Atlantic salmon near Vancouver Island, British Columbia, Canada into CHSE-214 cells following amplification in naive Atlantic salmon parr (22). Infected cells were incubated at 15°C in minimal essential medium supplemented with 10% FBS (MEM-10). Following the development of complete cytopathic effect, 1.0-ml aliquots of culture supernatant, further supplemented with 10% DMSO, were stored in liquid nitrogen.

To confirm P. salmonis in Atlantic salmon tissues and in cell culture, DNA was extracted from kidney and from culture supernatants (Qiagen DNEasy kit) for use as template in a nested polymerase chain reaction (PCR). The primers (EubA, EubB, PS2S, and PS2A2) and methods followed Mauel et al. (30). In addition, the organism was identified in tissue sections and imprints using an indirect fluorescent antibody test (IFAT) described by Almendras et al. (2).

Atlantic salmon parr received an intraperitoneal injection of infected tissue culture supernatant diluted 1:10 with MEM-10. The titer of P. salmonis in the inoculum was determined on CHSE-214 cells using the method of Kärber (23). Control salmon received an intraperitoneal injection with MEM-10 alone. After 14 days, the head kidney was dissected from 10 control and 10 infected salmon. The organs were plunged into liquid nitrogen immediately following dissection and subsequently stored at –80°C.

P. salmonis infections in cultured macrophages.
Separate suspensions of hematopoietic kidney (head kidney) leukocytes were obtained from six naive Atlantic salmon (~300 g) that had been reared in freshwater. Briefly, the head kidney from each fish was pressed through a sterile 100-µm mesh nylon screen in the presence of ice-cold RPMI 1640 medium supplemented with 25 mM HEPES, 0.3 µg/ml amphotericin B, 10 U/ml heparin, and 0.1% FBS (cRPMI). The cell suspension was layered onto a 2.5-ml bed of 54% Percoll and centrifuged at 4°C for 30 min at 650 g. Viable leucocytes were counted by hemocytometer in the presence of trypan blue and adjusted in cRPMI to 5.0 x 106 cells/ml.

Sterile, round, glass coverslips were aseptically placed into three wells in 12-well polystyrene tissue culture plates. Leukocyte cultures were incubated overnight at 15°C, and nonadherent cells were removed by washing three times with cold, unsupplemented RPMI. The medium in three of the four wells containing cells from each salmon (including the coverslipped well) was replaced with 1.0 ml of P. salmonis culture supernatant diluted 1:10 with cRPMI, containing 5% fetal bovine serum (FBS) (cRPMI-5). The medium in the fourth well was replaced with 1.0 ml cRPMI-5. Plates were incubated in a sealed humid chamber at 15°C.

Cells from each fish were harvested after 6 h and 24 h incubation. Three harvesting methods were followed. 1) Cells in the uninfected well and in one infected well were lysed with 250 µl MicroPoly (A)Pure lysis solution (Ambion). The lysate was transferred to a sterile tube and stored at –80°C. 2) Cells from the second infected well were scraped from an infected well and centrifuged for 4 min at 12,000 g, and the cell pellet fixed in 2% glutaraldehyde. Fixed cells were processed for routine transmission electron microscopy. 3) Cells from the third infected well and adherent to the glass coverslip were fixed in 100% alcohol and stained with Giemsa stain. The dried coverslip was inverted and mounted to a glass slide using Permount.

Microarray fabrication and quality control.
Microarray construction and initial testing were previously described (36). Briefly, 3,557 clones from 18 high-complexity salmonid cDNA libraries were selected with an emphasis on immune-relevant transcripts. Clones were robotically re-arrayed from daughter glycerol stock 384-well plates into 96-well plates prefilled with 7% glycerol in Luria broth + ampicillin, incubated overnight at 37°C, and checked for uniform optical density. Plasmid inserts were PCR amplified in a Tetrad PTC-200 thermocycler (MJ Research) using 1 µl overnight culture, 0.2 µM M13/pUC forward primer (5'-CCCAGTCACGACGTTGTAAAACG-3'), 0.2 µM M13/pUC reverse primer (5'-AGCGGATAACAATTTCACACAGG-3'), 2 mM MgCl2, 10 mM Tris·HCl, 50 mM KCl, 250 µM dNTPs, 1 U AmpliTaq (PerkinElmer), and nuclease-free H2O (Invitrogen) to 100 µl. PCR conditions were as follows: 2 min 95°C denaturation; 35 cycles of 95°C 30 s, 60°C 45 s, 72°C 3 min; 7 min 72°C. Five microliters of each PCR product was run on a 1% agarose gel to assess yield and quality. Of 3,557 salmonid clones, there were 3,312 strong, single bands (93%), 170 absent (5%), and 75 multiple bands (2%). Five exogenous genome (Arabidopsis) cDNAs were amplified from the following clones kindly provided by the Arabidopsis Information Resource (http://www.arabidopsis.org/): rubisco activase, protochlorophyllide reductase, chlorophyll a/b-binding protein CP29, PSII oxygen-evolving complex protein 2, and tonoplast intrinsic protein root-specific RB7. Arabidopsis cDNAs were spotted in quadruplicate on each microarray and used for thresholding (determining number of transcripts present). PCR products were robotically cleaned (Qiagen) and consolidated into 384-well plates, lyophilized by SpeedVac, and resuspended in 15 µl of 3x SSC.

All cDNAs were printed as double, side-by-side spots on aminosilane slides (ez-rays, Matrix) with a BioRobotics MicroGrid II microarray printer (Apogent Discoveries). MicroSpot 10K quill pins (BioRobotics) in a 48-pin tool were used to deposit ~0.5 nl (0.2 ng cDNA) per spot onto the slide. Resulting microarrays have a 4-by-12 subgrid layout with 182 spots per subgrid, each spot having approximate diameter and pitch of 100 µm and 250 µm, respectively. The slides were cross-linked in a UV Stratalinker 2400 (Stratagene) at 120 mJ. Spot morphology was assessed by visual inspection or SYBR Green 1 (Molecular Probes) staining. To check clone tracking, 42 high-quality sequences were obtained from randomly selected wells of the consolidated plates used for microarray printing. All 42 had BLAST identifiers matching gene IDs predicted from the re-array spreadsheet, indicating accurate clone tracking throughout microarray fabrication.

RNA amplification, and microarray hybridization and analysis.
These microarray experiments were designed to comply with MIAME guidelines (7). MicroPoly(A)Pure kits (Ambion) were used to prepare mRNA from P. salmonis-infected and control Atlantic salmon macrophages and head kidney. To have adequate quantities of template for microarray target syntheses, infected and control (24 h incubation) macrophage mRNA samples were subjected to two rounds of amplification using the MessageAmp aRNA kit (Ambion) according to the manufacturer’s instructions. Initial macrophage mRNA isolations yielded 0.9–1.8 µg of mRNA per sample. Four hundred nanograms of macrophage mRNA was used as template in the first round of amplification, yielding 8.1 µg of infected macrophage amplified RNA (aRNA) and 7.8 µg of control macrophage aRNA. One microgram of first-round macrophage aRNA was used as template in the second round of amplification, yielding 81.2 µg of infected macrophage aRNA and 83.3 µg of control macrophage aRNA. Head kidney mRNA samples were prepared from pooled (n = 10) infected or control kidney. Since the amounts of kidney mRNAs available were inadequate for running a microarray experiment (1.5 µg of control and 7.6 µg of infected head kidney mRNA), these samples were subjected to one round of amplification. Two hundred nanograms of infected or control head kidney mRNA was used as template in aRNA synthesis reactions. Two infected head kidney samples were pooled to yield 8.9 µg of aRNA, and a single control head kidney sample contained 8.4 µg of aRNA. Infected and control macrophage and head kidney aRNA samples were visualized on agarose gels to check quality and quantity (data not shown). Except for the amounts of aRNA and reverse transcription (RT) primer used, macrophage and head kidney target labeling reactions and microarray hybridizations were identical. Macrophage target syntheses used 5 µg of second-round aRNA and 5 µg of random primers (Roche), and head kidney target syntheses used 2 µg of aRNA and 2 µg of random primers (Roche). RT primers were mixed with aRNA samples, and nuclease-free H2O (Invitrogen) was added to make 10 µl total volume. Samples were incubated at 70°C 4 min, then placed on ice. RT reactions included 50 mM Tris·HCl, pH 8.3, 75 mM KCl, 3 mM MgCl2, 10 mM DTT; 0.5 mM each of dATP, dCTP, and dGTP; 0.2 mM dTTP, 0.05 mM Cy3-dUTP or Cy5-dUTP (Amersham), 40 U RNAguard ribonuclease inhibitor (Amersham), and 400 U SuperScript II RNase H reverse transcriptase (Invitrogen). RT reactions were incubated at 42°C for 2.5 h, followed by addition of 0.025 U RNase A (Sigma) and 1.5 U RNase H (New England BioLabs) and incubation at 37°C for 30 min. Labeled targets were cleaned using the QIAquick PCR purification kit (Qiagen) and precipitated, with 300 mM sodium acetate, 1 µl of 20 µg/µl glycogen (Invitrogen), and 2.5 vol of 100% ethanol, at –20°C overnight. Each labeled target was recovered by centrifugation (14,000 rpm, 4°C, 1 h), washed in 70% ethanol, air-dried, and resuspended in 60 µl of hybridization buffer: 50% deionized formamide, 5x SSC, 0.1% SDS, 5 µl of 5 µg/µl oligo dT (5'T18[Q-N]3'), 5 µl of 2 µg/µl BSA (Pierce), and 2 µl of 10 µg/µl sonicated human placental DNA (Sigma). Targets were incubated at 96°C for 3 min and then at 65°C until applied to microarrays. Microarrays were prepared for hybridization by washing twice for 5 min each wash in 0.1% SDS, washing five times for 1 min each time in MilliQ H2O, immersing 3 min in 95°C MilliQ H2O, and drying by centrifugation (2,000 rpm, 5 min, in 50-ml conical tube). Microarray hybridizations were run in the dark under HybriSlips hybridization covers (Grace Biolabs) in slide hybridization chambers (Corning) submerged in a 45°C water bath for 16 h. Coverslips were floated off at 45°C in 1x SSC, 0.2% SDS, and arrays were washed once for 10 min at 45°C in 1x SSC, 0.2% SDS, three times for 4 min each time at room temperature in 0.1x SSC, 0.1% SDS, and four times for 4 min each time at room temperature in 0.1x SSC. Slides were dried by centrifugation as before.

Fluorescent images of hybridized arrays were acquired immediately at 10 µm resolution using ScanArray Express (PerkinElmer). The Cy3 and Cy5 cyanine fluors were excited at 543 nm and 633 nm, respectively, and the same laser power (90%) and photomultiplier tube (PMT) settings were used for all slides in a study (PMT 72 Cy3, PMT 63 or 64 Cy5 for macrophage study; PMT 75 Cy3, PMT 63 or 64 Cy5 for head kidney study). Fluorescent intensity data were extracted from TIF images using ImaGene 5.5 software (BioDiscovery). Quality statistics were compiled in Excel from raw ImaGene fluorescence intensity report files. Elements were sorted (7,356 salmonid spots representing 3,557 different cDNAs, 20 Arabidopsis spots representing 5 different cDNAs, and 1,356 other control spots), for each microarray’s Cy3 and Cy5 data. Median signal values from the 20 Arabidopsis control spots were used to calculate threshold (mean plus 3 standard deviations), and mean numbers of salmonid elements passing threshold were determined for Cy3 and Cy5 data separately. All quality statistics for microarrays used in this report, as well as all scanned microarray TIF images, an ImaGene grid, the gene identification file, and ImaGene quantified data files, are available online from the University of Victoria Centre for Biomedical Research (http://web.uvic.ca/cbr/grasp). In addition, all microarray raw (TIF images) and extracted (ImaGene files) data have been deposited at the National Center for Biotechnology Information (NCBI) in the Gene Expression Omnibus database (GEO; http://www.ncbi.nlm.nih.gov/geo/). The GEO platform accession number is GPL966, and the sample accession numbers are GSM15897 (slide CL012.003), GSM15898 (CL012.004), GSM15899 (CL012.005), GSM16101 (DD035.031), GSM16389 (DD035.032), GSM16390 (DD035.033), and GSM16391 (DD035.034). Data transformation (background correction, and setting background corrected values < 0.01 to 0.01), normalization (Lowess), and analysis (formation and manipulation of fold change transcript lists) were performed in GeneSpring 5 (Silicon Genetics).

Real-time quantitative PCR (QPCR).
Poly(A) RNA (mRNA) was prepared from flash-frozen P. salmonis-infected (pooled, n = 10) and noninfected (pooled, n = 10) head kidney and from P. salmonis-infected and noninfected macrophages (24 h incubation time), using MicroPoly(A) Pure kits (Ambion). Head kidney mRNA was subjected to one round of amplification; microarray targets were synthesized from first-round aRNA, and QPCR cDNA templates were synthesized from mRNA. Macrophage mRNA was subjected to two rounds of amplification; microarray targets were synthesized from second-round aRNA, and QPCR templates were synthesized from first-round aRNA. Two hundred nanograms of mRNA each from infected and control head kidney was reverse transcribed using 200 ng oligo dT primer (5'T18[Q-N]3'). Two hundred nanograms of first-round aRNA each from infected and control macrophage was reverse transcribed using 200 ng random primers (Roche). RT reactions included 40 U RNase inhibitor (Promega), 500 µM dNTPs, 10 mM DTT, and 400 U SuperScript II RNase H reverse transcriptase (Invitrogen) with the manufacturer’s buffer. First-strand cDNAs were diluted 1:20 and used as templates for QPCR analysis. QPCR used two PCR primers per transcript of interest (TOI) and SYBR Green I dye. Ubiquitin was selected as the normalizer gene, since its expression was stable (background-corrected, Lowess normalized Cy5/Cy3 ratios were within 2-fold lines of scatter plots) in all microarrays used in this report. QPCR primer pairs for TOI and ubiquitin were designed from EST FASTA files using Primer3 (from the Whitehead Institute for Biomedical Research, http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) and the following guidelines: product size 150–250 bp, Tm 60 ± 1°C, at least three of the 3' terminal six bases G/C. For each TOI, sequences of the forward and reverse QPCR primers are given in the table in Fig. 2. Primers used for ubiquitin were: 5'-ATGTCAAGGCCAAGATCCAG-3' and 5'-ATAATGCCTCCACGAAGACG-3'. Reactions (50 µl total volume) containing 2 µl diluted template, 200 nM each primer, and 1x Brilliant SYBR Green QPCR Master Mix (Stratagene) were run in triplicate using the Mx3000P real-time PCR system (Stratagene) and the following cycling parameters: 95°C for 9 min; then 40 cycles of 95°C for 15 s, 52°C for 30 s, 72°C for 45 s. Controls (no template) were run for all primer pairs.



View larger version (74K):
[in this window]
[in a new window]
 
Fig. 2. Quantitative PCR (QPCR) confirmation of microarray results. 1Real-time QPCR and microarray results for 19 transcripts of interest, normalized to ubiquitin. Average () ratios (fold changes) and standard errors of the mean (SEM) are shown. Numbers of replicates: macrophage microarray, n = 3; head kidney microarray, n = 4; all QPCR, n = 3. Macrophage mRNAs were subjected to two rounds of amplification to provide adequate amounts of amplified RNA (aRNA) for microarray experiments; microarray targets were synthesized from second-round aRNA, and QPCR cDNA templates were synthesized from first-round aRNA. Kidney mRNAs were subjected to one round of amplification; microarray targets were synthesized from aRNA, and QPCR cDNA templates were synthesized from mRNA. GenBank accession numbers (acc.) are for ESTs corresponding to informative microarray features; n/a = not available (EST not yet submitted). Gene names of the most significant BLASTX hits are given, and additional annotation of these salmonid ESTs is given in Tables 14. 2–5Salmonid microarray features from 2Table 1, 3Table 2, 4Table 3, and 5Table 4. Shading indicates either greater than twofold upregulated (black, or "+") or downregulated (gray, or "–") in infected samples relative to noninfected samples.

 
Triplicate data (Ct values) for each TOI with each treated (infected) template were first normalized to ubiquitin (TOI minus average ubiquitin Ct for the same template). Data (normalized Ct values) from infected and control head kidney or macrophage templates were compared (converted to fold differences) using Mx3000P software and the relative quantification method (34) and assuming 100% efficiencies. All Ct values are available as online supplemental data (available at http://web.uvic.ca/cbr/grasp; also, see Supplemental Table S6B, available at the Physiological Genomics web site).1 Melting curves of the QPCR products for all TOI except inhibitor of nuclear factor {kappa}B {alpha} (I{kappa}B{alpha}) showed single peaks, and end-point analysis of all QPCR products on agarose gel showed strong, single bands of the expected sizes. With all templates, I{kappa}B{alpha} QPCR product melting curves showed a weak secondary peak that was undetectable on agarose gel.

RT-PCR confirmation of P. salmonis infection.
Two hundred nanograms of P. salmonis-infected and control macrophage mRNA was reverse transcribed with 200 ng random primer (Roche), and 5 µl (~33 ng) of cDNA was used as template in PCR. Two hundred nanograms of P. salmonis infected and control head kidney mRNA were reverse transcribed using MessageAmp aRNA kit reagents and instructions (Ambion). Head kidney cDNAs were purified and eluted in 100 µl nuclease-free water, and 1 µl (~2 ng) was used as template in PCR. Fifty-microliter PCR reactions were run using P. salmonis-specific primers PS2S and PS2A2 (30), 1x Q solution (Qiagen), dNTPs (200 µM each), 1x PCR buffer (Qiagen), 1.25 U HotStarTaq DNA polymerase (Qiagen), and the following cycling parameters: 95°C for 15 min; then 40 cycles of 94°C for 1 min, 50°C for 2 min, 72°C for 2 min. Positive controls (using P. salmonis DNA) and negative controls (no template) were run. Ten microliters of each RT-PCR product or negative control, and 2.5 µl of positive control reaction, were visualized by agarose gel electrophoresis (Fig. 1C).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Microscopic and RT-PCR confirmation of infection.
In the macrophage study, adherent head kidney leukocytes displayed morphological characteristics of macrophages. In addition, lymphocytes, polymorphonuclear leukocytes, and erythrocytes were evident on stained coverslips. Cultured cells were exposed to ~104.8 TCID50 ("tissue culture infectious dose 50") of cultured P. salmonis, and infection was confirmed by RT-PCR with P. salmonis-specific primers (30). The expected, 467-bp band was amplified from supernatants of infected cultures but not from uninfected culture supernatants (data not shown). A strong, 467-bp product was obtained using cDNAs generated from 24-h infected macrophages, whereas a faint 467-bp band was obtained from 6-h infected macrophages (Fig. 1C). No bands were obtained from control, noninfected macrophages (Fig. 1C). Light microscopy confirmed infections within macrophages (data not shown). In the head kidney study, salmon exposed to ~104.8 TCID50 of cultured P. salmonis began to die 8 days after injection. The microorganism was recognizable in stained histological preparations of organs from mortalities and in organ sections examined by IFAT. Head kidney was obtained from infected salmon 14 days following infection and from control salmon. All remaining infected salmon died by 21 days postinjection, and no unexpected mortalities were observed in control salmon. A strong, 467-bp RT-PCR product was obtained using infected head kidney cDNA and P. salmonis-specific primers (Fig. 1C) (30). No bands were obtained from control, noninfected head kidney (Fig. 1C). P. salmonis was observed microscopically within macrophages in stained imprints and IFAT preparations (not shown) of kidney from infected (Fig. 1D) but not control salmon (data not shown).

Effects of P. salmonis infection on host macrophage and head kidney gene expression.
Differential gene expression in P. salmonis-infected and noninfected Atlantic salmon macrophages (24 h incubation), and in P. salmonis-infected and noninfected Atlantic salmon head kidney (14 days postinjection), was determined using a microarray containing 3,557 different cDNAs selected from high-complexity, salmonid (primarily Atlantic salmon) cDNA libraries (36). Since the strong, P. salmonis-specific RT-PCR product in 24-h infected macrophages suggested a level of infection that may be altering the host cell transcriptome to allow intracellular survival of the pathogen (Fig. 1C), this time point was chosen for microarray analysis. On the salmonid elements, average Cy3 mean signal/mean background (s/b) was 13.0 (SE 1.2) and average Cy5 s/b was 10.8 (SE 1.3) in the macrophage study, while average Cy3 s/b was 11.2 (SE 0.1) and average Cy5 s/b was 12.3 (SE 0.5) in the head kidney study (Supplemental Table S1; at http://web.uvic.ca/cbr/grasp). In the macrophage study, an average of 67.3% (SE 2.1) of Cy3-labeled transcripts passed threshold and were therefore deemed "present" (see MATERIALS AND METHODS) and an average of 69.8% (SE 3.8) of Cy5-labeled transcripts passed threshold; in the head kidney study, these values were 73.0% (SE 0.9) and 80.6% (SE 1.1), respectively (Supplemental Table S1; http://web.uvic.ca/cbr/grasp).

Salmonid cDNAs having significant (E < 10–5) BLASTX or BLASTN hits against the current GenBank nr or nt databases, as well as unknowns (cDNAs with no significant BLAST hits), are described for transcripts responding reproducibly (>2-fold up- or downregulated in all slides of a study) to P. salmonis infection (Tables 14). Changes in transcription of informative genes and the degree of similarity (length and percent identity over aligned region) between the translation of each salmonid cDNA’s expressed sequence tag (EST) and its top (most negative E-value) BLASTX hit, or between ESTs and their top BLASTN hits, are also shown (Tables 14) and serve to identify potentially informative transcripts.


View this table:
[in this window]
[in a new window]
 
Table 1. Transcripts greater than 2-fold upregulated in P. salmonis-infected Atlantic salmon macrophages in all 3 slides of study

 

View this table:
[in this window]
[in a new window]
 
Table 4. Transcripts greater than 2-fold downregulated in P. salmonis-infected Atlantic salmon head kidney in all 4 slides of study

 
In the macrophage microarray study, 71 different transcripts were reproducibly upregulated in P. salmonis-infected primary cultures of Atlantic salmon head kidney macrophages (Fig. 1A, Table 1). Informative transcripts in Tables 14 are listed by descending signal values, rather than fold change values, in the appropriate channel (e.g., infected for upregulated in infected macrophages, or noninfected for downregulated in infected macrophages) of one slide in the study. The most abundant upregulated transcript in infected macrophages was the 60S ribosomal protein L35 (Table 1). Multiple entries with the same gene name in an informative transcript list (different microarray features with identical top BLAST hits) represent single genes or closely related paralogs and are indicated by subscripts in Tables 14. For example, there were five nephrosin, three I{kappa}B{alpha}, and two matrix metalloproteinase microarray features in the original transcript list from which Table 1 was derived (see Supplemental Table S2 at http://web.uvic.ca/cbr/grasp). The presence of multiple entries of genes in a given informative transcript list provides an internal validation of microarray results. From microarray data, the most highly induced (>6.2-fold) macrophage transcripts with significant BLASTX hits were immune-responsive gene 1 (65.8-fold, SE 3.2), amebocyte aggregation factor (31.6-fold, SE 5.8), chloride intracellular channel 2 (18.1-fold, SE 0.2), transferrin (12.0-fold, SE 0.3), nuclear receptor coactivator 4 (11.3-fold, SE 0.3), granzyme-like protein 1 (9.7-fold, SE 1.8), similar to sequestosome 1 (9.1-fold, SE 0.5), C-type lectin 2-1 (8.0-fold, SE 0.9), I{kappa}B{alpha} (8.0-fold, SE 0.1), Ca2+-transporting ATPase (7.5-fold, SE 0.1), MAP1 light chain 3-like protein 2 (7.4-fold, SE 1.4), HES1 (7.2-fold, SE 0.9), and matrix metalloproteinase (6.3-fold, SE 0.3) (Table 1). QPCR confirmation of microarray results with these samples showed a 97.5-fold induction of immune-responsive gene 1 (SE 0.6), a 111.0-fold induction of amebocyte aggregation factor (SE 2.0), a 32.1-fold induction of chloride intracellular channel 2 (SE 0.9), a 3.9-fold induction of nuclear receptor coactivator 4 (SE 0.5), a 26.7-fold induction of C-type lectin 2-1 (SE 0.2), a 17.3-fold induction of I{kappa}B{alpha} (SE 1.3), a 14.4-fold induction of HES1 (SE 0.3), and a 70.4-fold induction of matrix metalloproteinase (SE 5.1) in infected macrophages (table in Fig. 2).

Thirty-one different transcripts were reproducibly downregulated in P. salmonis-infected macrophages (Fig. 1A, Table 2). Transcripts in Table 2 are listed by descending control macrophage signal values from one slide in the study. In infected macrophages, the most abundant downregulated transcript was {alpha}1-microglobulin (Table 2). There were two K18 simple type I keratin, two 14.5-kDa translational inhibitor, and two kidney dicarbonyl reductase entries in the transcript list from which Table 2 was derived (Supplemental Table S3, http://web.uvic.ca/cbr/grasp). From microarray data, the most highly suppressed (>4.2-fold) macrophage transcripts with significant BLASTX hits were selenoprotein P (30.7-fold, SE 3.6), 14.5-kDa translational inhibitor (26.8-fold, SE 19.2), collagen {alpha}1 (7.1-fold, SE 2.2), cysteine proteinase (5.8-fold, SE 2.2), {alpha}1-microglobulin (5.4-fold, SE 0.4), retinoid X receptor-{alpha} (4.7-fold, SE 0.2), cysteine-rich protein 1 (4.5-fold, SE 0.3), N-formylpeptide receptor-like 2 (4.4-fold, SE 0.1), and lysophospholipase (4.3-fold, SE 0.4) (Table 2). QPCR with these samples showed a 43.8-fold suppression of selenoprotein P (SE 3.3), a 4.9-fold suppression of retinoid X receptor-{alpha} (SE 0.1), a 4.7-fold suppression of N-formylpeptide receptor-like 2 (SE 0.1), and a 4.7-fold suppression of lysophospholipase (SE 0.3) in infected macrophages (table in Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Transcripts greater than 2-fold downregulated in P. salmonis-infected Atlantic salmon macrophages in all 3 slides of study

 
In the head kidney study, 30 different transcripts were reproducibly upregulated in P. salmonis-infected tissues (Fig. 1A, Table 3). Transcripts in Table 3 are listed by descending infected head kidney signal values from one slide in the study. In infected head kidney, the most abundant upregulated transcript was C-type lectin 2-1 (Table 3). There were two C-type lectin 2-1-like transcripts with significant BLASTX hits, two additional C-type lectin 2-1-like transcripts with significant BLASTN hits only, and two amebocyte aggregation factor entries in the transcript list that gave rise to Table 3 (Supplemental Table S4, http://web.uvic.ca/cbr/grasp). From microarray data, the most highly induced (>4.2-fold) head kidney transcripts with significant BLASTX hits were immune-responsive gene 1 (20.4-fold, SE 2.8), argininosuccinate synthase (7.1-fold, SE 3.8), amebocyte aggregation factor (7.1-fold, SE 0.9), differentially regulated trout protein 1 (6.9-fold, SE 1.0), lysophospholipase (6.3-fold, SE 1.2), prostaglandin D synthase (6.3-fold, SE 0.9), and 5-aminolevulinate synthase (4.3-fold, SE 0.3) (Table 3). QPCR with these samples showed a 13.4-fold induction of immune-responsive gene 1 (SE 0.8), a 10.6-fold induction of amebocyte aggregation factor (SE 1.4), a 3.6-fold induction of lysophospholipase (SE 0.1), and a 9.4-fold induction of 5-aminolevulinate synthase (SE 0.4) in infected head kidney (table in Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 3. Transcripts greater than 2-fold upregulated in P. salmonis-infected Atlantic salmon head kidney in all 4 slides of study

 
Thirty-nine different transcripts were reproducibly downregulated in P. salmonis-infected head kidney (Fig. 1A, Table 4). Transcripts in Table 4 are listed by descending control head kidney signal values from one slide in the study. In infected head kidney, the most abundant downregulated transcript was selenoprotein P (Table 4). There were four prostaglandin D synthase, three ependymin, two glutathione S-transferase, two retinol-binding protein, two reverse transcriptase-like, and two progastricsin entries in the transcript list that gave rise to Table 4 (Supplemental Table S5, http://web.uvic.ca/cbr/grasp). From microarray data, the most highly suppressed (>4.2-fold) head kidney transcripts with significant BLASTX hits were zinc finger protein (9.6-fold, SE 0.7), serine protease inhibitor Kazal type 5 (8.0-fold, SE 2.0), ependymin (5.7-fold, SE 0.8), prostaglandin D synthase (5.1-fold, SE 0.2), 3ß-hydroxysteroid dehydrogenase (5.0-fold, SE 0.5), reverse transcriptase-like (4.9-fold, SE 0.7), and B-cell translocation gene 1 (4.4-fold, SE 0.8) (Table 4). QPCR with these samples showed a 4.2-fold suppression of selenoprotein P (SE 0.1), a 5.7-fold suppression of ependymin (SE 0.4), and a 4.3-fold suppression of B-cell translocation gene 1 (SE 0.1) in infected head kidney (table in Fig. 2).

Overlap and QPCR confirmation of macrophage and head kidney microarray studies.
Seven transcripts (immune-responsive gene 1, amebocyte aggregation factor, an unknown similar to a previously identified S. salar kidney cDNA sequence, C-type lectin 2-1, an unknown with no significant BLAST hits, type 1 cytokeratin enveloping layer, and O-methyltransferase) were reproducibly greater than twofold upregulated in both infected macrophages and infected head kidney (Fig. 1B, Tables 1 and 3). Four transcripts ({alpha}1-microglobulin, selenoprotein P, B-cell translocation gene 1, and hemopexin-like) were reproducibly greater than twofold downregulated in both infected macrophages and infected head kidney (Fig. 1B, Tables 2 and 4). The selection of transcripts of interest for QPCR confirmation was based on their potential utility as molecular biomarkers of P. salmonis infection. Preference was given to those S. salar transcripts with significant and functionally annotated BLASTX hits and relatively high fold differences in expression between infected and noninfected samples (Fig. 1, Tables 14, and table in Fig. 2). The characterization and evaluation of potential biomarker transcripts lacking significant BLASTX hits (i.e., with only significant BLASTN hits, or with no significant BLAST hits) is ongoing. QPCR was performed in triplicate by relative quantification for 19 transcripts of interest normalized to ubiquitin (table in Fig. 2; Supplemental Tables S6A and S6B). QPCR products of the predicted sizes were obtained for all transcripts of interest and ubiquitin, and no PCR products were observed in no-template controls. Microarray results were considered confirmed if, for a TOI, the direction of change in expression (e.g., >2-fold upregulated in infected macrophages relative to noninfected macrophages) shown by QPCR matched that seen in microarray experiments (table in Fig. 2 and Supplemental Table S6A).


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The purpose of this study was to determine the effects of P. salmonis infection on gene expression in Atlantic salmon macrophages and head kidney. Two cDNA microarray-based experiments studied 1) differences in gene expression of infected vs. noninfected head kidney macrophages in culture and 2) differences in gene expression of intact head kidney taken from infected vs. noninfected fish (Fig. 1A). Each study provided insight into the effects of P. salmonis infection on host gene expression at a cellular or tissue level. The in vivo and in vitro expression studies have 11 informative transcripts (7 upregulated, 4 downregulated) in common (Fig. 1B, Tables 14). Lists of salmon transcripts induced or suppressed by P. salmonis infection provided insight into the host response to infection and the mechanism by which the pathogen evades this response. Functional annotations of regulated host transcripts (Tables 14) revealed cellular activities that may be altered in P. salmonis-infected macrophages and head kidney. Candidate molecular biomarkers of P. salmonis infection, selected from the most drastically induced or suppressed transcripts that had top BLAST hits with potentially immune-relevant functional annotations, were confirmed using QPCR (table in Fig. 2).

Responses to lipopolysaccharide and oxidative stress.
Activated macrophages respond to bacterial infection and to lipopolysaccharide (LPS) by using NADPH oxidase to reduce oxygen to superoxide (O2), which may then be converted to reactive oxygen intermediates such as free radicals and singlet oxygen (3, 42). This "respiratory burst" protects the host against invading microorganisms by creating a toxic intracellular environment. The host cell protects itself from oxidative damage partly through the synthesis of glutathione, a potent antioxidant (45). The expression of at least 10 genes involved in response to oxidative stress (glutathione S-transferase, glutathione reductase, glutathione peroxidase, cytochrome b558 {alpha}- and ß-subunits, myeloid-specific peroxidase, hydroxyprostaglandin dehydrogenase 15, 20ß-hydroxysteroid dehydrogenase, ribonucleotide reductase r2 class I, and cytochrome P450 4F2) is consistently upregulated in P. salmonis-infected macrophages by up to sixfold (Table 1). Of known antioxidants, only two (selenoprotein P and kidney dicarbonyl reductase) are reproducibly downregulated in infected macrophages (Table 2). Interestingly, both the most highly induced antioxidant (glutathione peroxidase, Table 1) and the most highly repressed antioxidant (selenoprotein P, Table 2 and table in Fig. 2) in P. salmonis-infected macrophages are functionally dependent upon selenium. Selenoprotein P is also downregulated in infected head kidney (Table 4 and table in Fig. 2).

I{kappa}B{alpha} is reproducibly induced in P. salmonis-infected macrophages (Table 1 and table in Fig. 2) but not in infected head kidney (table in Fig. 2). I{kappa}B{alpha} is an endogenous inhibitor of nuclear factor {kappa}B (NF-{kappa}B), an evolutionarily conserved, redox-sensitive transcription factor that regulates the expression of many genes (e.g., cytokines) involved in the innate and adaptive immune responses (43). NF-{kappa}B is often activated in infected host cells. Some pathogenic microorganisms interfere with NF-{kappa}B activation, thereby evading the host immune response and allowing intracellular survival of the pathogen (44). NF-{kappa}B, with other transcription factors, upregulates inducible nitric oxide synthase (iNOS) transcription in LPS-stimulated mouse macrophages (25), increasing the synthesis of reactive nitrogen intermediates that contribute to the killing of phagocytosed bacteria (9). LPS stimulation also induces iNOS transcription in teleost macrophages (40). The upregulation of I{kappa}B{alpha} transcription in P. salmonis-infected macrophages suggests that this pathogen may be interfering with NF-{kappa}B activation, with subsequent repression of genes controlled by NF-{kappa}B (e.g., iNOS) hypothetically forming part of the mechanism by which P. salmonis evades the macrophage immune response. The hypotheses that selenium imbalance and NF-{kappa}B inhibition may play roles in P. salmonis pathogenesis are strengthened by the recent finding that selenium status affects LPS-induced iNOS expression in mouse macrophages, potentially through increased activation of NF-{kappa}B (35).

The generalized upregulation of antioxidant transcripts in P. salmonis-infected macrophages (Table 1) was largely absent in P. salmonis-infected head kidney (Table 3). Glutathione S-transferase was upregulated in infected macrophages (Table 1 and table in Fig. 2), but not in infected head kidney (Table 4 and table in Fig. 2). In human endothelial cells infected with the rickettsial agent of Rocky Mountain spotted fever, decreased levels of reduced glutathione and increased peroxide levels correlated with signs of cell damage (14). Similarly, P. salmonis infection may alter the host antioxidant system, resulting in the cell death and necrosis observed in infected head kidney and other tissues (1). Treatment of rickettsia-infected human endothelial cells with an antioxidant increased intracellular levels of reduced glutathione and prevented oxidative injury (14). The possibility that treatment with antioxidants could decrease cellular damage caused by rickettsial infection in salmonids requires further investigation.

Innate and adaptive immune responses.
In addition to antioxidants, other innate immune-relevant genes were regulated by P. salmonis infection in macrophages and head kidney (Tables 14 and table in Fig. 2). C-type lectin 2-1 and immune-responsive gene 1 (Irg1) were upregulated in infected macrophages (Table 1 and table in Fig. 2) and infected head kidney (Table 3 and table in Fig. 2). MARCO, a macrophage receptor that binds both gram-positive and gram-negative bacteria, was downregulated in infected macrophages (Table 2 and table in Fig. 2). One complement component C3-like transcript was downregulated in infected macrophages (Table 2), while a different complement component C3-like putative paralog was upregulated in infected head kidney (Table 3). Induction of ferritin H-3 and transferrin in infected macrophages suggests altered iron ion homeostasis. Since transferrin appears to be a positive acute phase protein in rainbow trout (5), upregulation of this gene in P. salmonis infected macrophages may signify a host attempt to control bacterial proliferation by limiting available iron. Genes involved in iron homeostasis did not appear to be regulated by P. salmonis infection in head kidney (Tables 3 and 4). Alternatively, transferrin induction may be indicative of an autocrine macrophage activating pathway in salmonids as in cyprinid fish (41). Four immune-relevant transcripts that were reproducibly upregulated in P. salmonis-infected head kidney (C-type lectin 2-1, chemotaxin, differentially regulated trout protein 1, and complement component C3) (Table 3) were also identified in a subtractive library enriched for rainbow trout liver transcripts responding to Vibrio infection (4) and are considered putative acute phase proteins (5). These data suggest that P. salmonis infection induced inflammatory and acute phase responses in head kidney, potentially contributing to cell death and necrosis.

Six transcripts encoding antibody components (immunoglobulin heavy chain constant and variable regions and immunoglobulin light chain) were reproducibly upregulated in P. salmonis-infected macrophage cultures (Table 1). These results were likely due to the presence of lymphocytes in these cell cultures. Interestingly, no immunoglobulin transcripts were induced (Table 3) and one immunoglobulin heavy chain transcript was repressed (Table 4) in P. salmonis-infected head kidney. Several other transcripts involved in adaptive (lymphoid) immune responses were reproducibly downregulated in P. salmonis-infected head kidney (Table 4). These include a T cell receptor {alpha}-chain transcript, a salmon transcript related to a gene expressed in activated human lymphocytes, and a salmon transcript similar to C-C chemokine receptor 7 (a lymphocyte-specific, G protein-coupled receptor for the macrophage inflammatory protein 3-ß chemokine) (Table 4). These results may point to a compromised adaptive immune response in infected head kidney and may reflect in vitro lymphocyte activation by macrophage-derived cytokines.

Evolutionary conservation in transcriptomic responses to infection.
Several similarities were noted between the gene expression responses of P. salmonis-infected salmon macrophages and human macrophages stimulated by LPS or infected by intracellular pathogenic bacteria. The transcription of ferritin H and matrix metalloproteinase 9, among other genes, is upregulated in human macrophages stimulated with LPS (42). Ferritin H-3 and a matrix metalloproteinase were upregulated in P. salmonis-infected salmon macrophages (Table 1 and table in Fig. 2). P. salmonis-infected Atlantic salmon macrophages and mouse macrophages infected with Brucella abortus (a gram-negative, intracellular bacterial pathogen) have several responsive genes in common. C-type lectin, matrix metalloproteinase, Irg1, and I{kappa}B{alpha} were upregulated in both B. abortus-infected mouse macrophages (15) and in P. salmonis-infected salmon macrophages (Table 1 and table in Fig. 2). Of these genes, the expression of C-type lectin, matrix metalloproteinase, and Irg1 was also induced in P. salmonis-infected head kidney (Table 3 and table in Fig. 2). Irg1 was also induced in mouse macrophages infected with Mycobacterium tuberculosis (13). These observations suggest that intracellular bacterial pathogens elicit the expression of conserved immune genes among all vertebrate hosts.

A salmonid transcript belonging to the dermatopontin family was strongly induced in both infected macrophages and infected head kidney (Tables 1 and 3 and table in Fig. 2). The most significant BLASTX hits for this salmonid EST are horseshoe crab (Limulus sp.) hemagglutinin/amebocyte aggregation factor (GenBank nr accession number AAA28272; E value = 9.5e-29), sponge dermatopontin (CAC38786 E = 8.3e-25), and human dermatopontin (CAA80487 E = 9.9e-10). While dermatopontins are abundant in both vertebrate and invertebrate extracellular matrices, their functions remain poorly understood (29, 32). This salmon amebocyte aggregation factor-like cDNA sequence has been previously identified (GenBank nt accession number BG934969), but nothing is known about its expression or function. Limulus amebocytes are immunocompetent hemocytes thought to be functionally homologous to mammalian macrophages, platelets, and B and T lymphocytes (20, 28). The LPS-induced degranulation of Limulus amebocytes, mediated by G protein-linked receptors (39), releases coagulation factors, protease inhibitor, and antimicrobial peptides (18, 21). Amebocyte aggregation factor, also secreted from amebocyte granules, induces the aggregation of amebocytes and the agglutination of erythrocytes (18). These data suggest that amebocyte aggregation factor orthologs may be conserved components of salmon and invertebrate innate immune responses to bacterial invasion. Interestingly, four transcripts involved in G protein-coupled receptor signaling (calmodulin, N-formylpeptide receptor-like 2, complement component C3, and thrombin receptor-like 3) are reproducibly downregulated in infected macrophages (Table 2), but not in infected head kidney (Table 4). Suppression of G protein signaling may be part of the mechanism by which P. salmonis evades macrophage antimicrobial defenses.

Molecular biomarkers for P. salmonis infection.
Nineteen S. salar transcripts, selected because of their relatively high fold differences in expression between infected and noninfected samples as well as the immune-relevant functional annotations of their top BLASTX hits, were confirmed by QPCR (table in Fig. 2). These molecular biomarkers may be useful in research into the molecular pathogenesis of P. salmonis infection. Recent development and testing of vaccines against P. salmonis utilized relative percent survival to gauge vaccine effectiveness (26). We propose that QPCR-based molecular biomarker expression analyses may add sensitivity to current methods of evaluating the efficacies of anti-piscirickettsial vaccines and therapeutics.


    GRANTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This research was supported by Genome Canada, Genome BC, and the Province of British Columbia, and additionally by the Natural Sciences and Engineering Research Council of Canada (to B. F. Koop) and by the Shaw Foundation (to M. L. Rise).


    ACKNOWLEDGMENTS
 
We thank Caren Helbing, Nik Veldhoen, and Scott O’Brien for providing instruments and advice regarding quantitative PCR, and we thank Tom Hansen for assisting in submission of microarray data to Gene Expression Omnibus.

The URL for the Gene Ontology Consortium is http://www.geneontology.org (2001).

The programs PHRED (v. 0.990722.j) and PHRAP (v. 0.990329) are available from the University of Washington Genome Center (http://www.genome.washington.edu/UWGC/).


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: B. F. Koop, Centre for Biomedical Research, Univ. of Victoria, PO Box 3020, Victoria, British Columbia V8W 3N5, Canada (E-mail: bkoop{at}uvic.ca).

10.1152/physiolgenomics.00036.2004.

1 The Supplemental Material for this article (Supplemental Tables S1–S6) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00036.2004/DC1. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Almendras FE and Fuentealba IC. Salmonid rickettsial septicaemia caused by Piscirickettsia salmonis: a review. Dis Aquat Org 29: 137–144, 1997.[ISI]
  2. Almendras FE, Fuentealba IC, Jones SRM, Markham F, and Spangler E. Experimental infection and horizontal transmission of Piscirickettsia salmonis in freshwater-raised Atlantic salmon, Salmo salar L. J Fish Dis 20: 409–418, 1997.[CrossRef][ISI]
  3. Babior BM. NADPH oxidase: an update. Blood 93: 1464–1476, 1999.[Free Full Text]
  4. Bayne CJ, Gerwick L, Fujiki K, Nakao M, and Yano T. Immune-relevant (including acute phase) genes identified in the livers of rainbow trout, Oncorhynchus mykiss, by means of suppression subtractive hybridization. Dev Comp Immunol 25: 205–217, 2001.[CrossRef][ISI][Medline]
  5. Bayne CJ and Gerwick L. The acute phase response and innate immunity of fish. Dev Comp Immunol 25: 725–743, 2001.[CrossRef][ISI][Medline]
  6. Bravo S and Campos M. Coho salmon syndrome in Chile. FHS/AFS Newsl 17: 3, 1989.
  7. Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, Aach J, Ansorge W, Ball CA, Causton HC, Gaasterland T, Glenisson P, Holstege FC, Kim IF, Markowitz V, Matese JC, Parkinson H, Robinson A, Sarkans U, Schulze-Kremer S, Stewart J, Taylor R, Vilo J, and Vingron M. Minimum information about a microarray experiment (MIAME): toward standards for microarray data. Nat Genet 29: 365–371, 2001.[CrossRef][ISI][Medline]
  8. Brocklebank JR, Speare DJ, Armstrong RD, and Evelyn T. Septicemia suspected to be caused by a rickettsia-like agent in farmed Atlantic salmon. Can Vet J 33: 407–408, 1992.
  9. Chakravortty D, Hansen-Wester I, and Hensel M. Salmonella pathogenicity island 2 mediates protection of intracellular Salmonella from reactive nitrogen intermediates. J Exp Med 195: 1155–1166, 2002.[Abstract/Free Full Text]
  10. Chen MF, Yun S, Marty GD, McDowell TS, House ML, Appersen JA, Guenther TA, Arkush KD, and Hedrick RP. A Piscirickettsia salmonis-like bacterium with mortality of white seabass Atractoscion nobilis. Dis Aquat Org 43: 117–126, 2000.[ISI][Medline]
  11. Chern RS and Chao CB. Outbreaks of a disease caused by a rickettsia-like organism in cultured tilapias in Taiwan. Fish Pathol 29: 61–71, 1994.[ISI]
  12. Cusack RR, Groman DB, and Jones SRM. Rickettsial infection in farmed Atlantic salmon in eastern Canada. Can Vet J 43: 435–440, 2002.[ISI][Medline]
  13. Ehrt S, Schnappinger D, Bekiranov S, Drenkow J, Shi S, Gingeras TR, Gaasterland T, Schoolnik G, and Nathan C. Reprogramming of the macrophage transcriptome in response to interferon-{gamma} and Mycobacterium tuberculosis: signaling roles of nitric oxide synthase-2 and phagocyte oxidase. J Exp Med 194: 1123–1139, 2001.[Abstract/Free Full Text]
  14. Eremeeva ME and Silverman DJ. Rickettsia rickettsii infection of the EA.hy 926 endothelial cell line: morphological response to infection and evidence for oxidative injury. Microbiology 144: 2037–2048, 1998.[ISI][Medline]
  15. Eskra L, Mathison A, and Splitter G. Microarray analysis of mRNA levels from RAW264.7 macrophages infected with Brucella abortus. Infect Immun 71: 1125–1133, 2003.[Abstract/Free Full Text]
  16. Fryer JL, Lannan CN, Garcés LH, Larenas JJ, and Smith PA. Isolation of a Rickettsiales-like organism from diseased coho salmon (Oncorhynchus kisutch) in Chile. Fish Pathol 25: 107–114, 1990.[ISI]
  17. Fryer JL, Lannan CN, Giovannoni SJ, and Wood ND. Piscirickettsia salmonis gen. nov, sp nov, the causative agent of an epizootic disease in salmonid fishes. Int J Syst Bacteriol 42: 120–126, 1992.[Abstract]
  18. Fujii N, Minetti CASA, Nakhasi HL, Chen S, Barbehenn E, Nunes PH, and Nguyen NY. Isolation, cDNA cloning, and characterization of an 18-kDa hemagglutinin and amebocyte aggregation factor from Limulus polyphemus. J Biol Chem 267: 22452–22459, 1992.[Abstract/Free Full Text]
  19. Garcés LH, Larenas JJ, Smith PA, Sandino S, Lannan CN, and Fryer JL. Infectivity of a rickettsia isolated from coho salmon Oncorhynchus kisutch. Dis Aquat Org 11: 93–97, 1991.[ISI]
  20. Gupta AP. Insect immunity and other hemocytes: roles in cellular and humoral immunity. In: Immunology of Insects and Higher Arthropods, edited by Gupta AP. Boca Raton, FL: CRC, 1991.
  21. Iwanaga S. The limulus clotting reaction. Curr Opin Immunol 5: 74–82, 1993.[ISI][Medline]
  22. Jones SRM, Markham RJF, Groman DB, and Cusack RR. Virulence and antigenic characteristics of a cultured Rickettsiales-like organism isolated from farmed Atlantic salmon Salmo salar in eastern Canada. Dis Aquat Org 33: 25–31, 1998.[ISI][Medline]
  23. Kärber G. Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche. Arch Exp Pathol Pharmakol 162: 480–483, 1931.
  24. Khoo L, Dennis PM, and Lewbart GA. Rickettsia-like organisms in the blue-eyed plecostomus, Panaque suttoni (Eigenmann and Eigenmann). J Fish Dis 18: 157–164, 1995.[ISI]
  25. Kim JY and Jeong HG. Down-regulation of inducible nitric oxide synthase and tumor necrosis factor-{alpha} expression by bisphenol A via nuclear factor-{kappa}B inactivation in macrophages. Cancer Lett 196: 69–76, 2003.[CrossRef][ISI][Medline]
  26. Kuzyk MA, Burian J, Machander D, Dolhaine D, Cameron S, Thornton JC, and Kay WW. An efficacious recombinant subunit vaccine against the salmonid rickettsial pathogen Piscirickettsia salmonis. Vaccine 19: 2337–2344, 2001.[CrossRef][ISI][Medline]
  27. Lannan CN and Fryer JL. Extracellular survival of Piscirickettsia salmonis. J Fish Dis 17: 545–548, 1994.[ISI]
  28. MacPherson JC and Jacobs RS. An 18.5 kDa protein from the amebocyte of Limulus polyphemus, homologous to the previously described amebocyte aggregation factor, expresses alternative phospholipase A(2) activity. Comp Biochem Physiol B 127: 31–44, 2000.[CrossRef][ISI][Medline]
  29. Marxen JC, Nimtz M, Becker W, and Mann K. The major soluble 19.6 kDa protein of the organic shell matrix of the freshwater snail Biomphalaria glabrata is an N-glycosylated dermatopontin. Biochim Biophys Acta 1650: 92–98, 2003.[ISI][Medline]
  30. Mauel MJ, Giovannoni SJ, and Fryer JL. Development of polymerase chain reaction assays for detection, identification, and differentiation of Piscirickettsia salmonis. Dis Aquat Org 26: 189–195, 1996.[ISI]
  31. Mauel MJ and Miller DL. Piscirickettsiosis and piscirickettsiosis-like infections in fish: a review. Vet Microbiol 87: 279–289, 2002.[CrossRef][ISI][Medline]
  32. Okamoto O, Fujiwara S, Mayumi ABE, and Sato Y. Dermatopontin interacts with transforming growth factor ß and enhances its biological activity. Biochem J 337: 537–541, 1999.[CrossRef][ISI][Medline]
  33. Olsen AB, Melby HP, Speilberg L, Evensen O, and Håstein T. Piscirickettsia salmonis infection in Atlantic salmon Salmo salar in Norway: epidemiological, pathological and microbiological findings. Dis Aquat Org 31: 35–48, 1997.[ISI]
  34. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: e45, 2001.[Abstract/Free Full Text]
  35. Prabhu KS, Zamamiri-Davis F, Stewart JB, Thompson JT, Sordillo LM, and Reddy CC. Selenium deficiency increases the expression of inducible nitric oxide synthase in RAW 264.7 macrophages: role of nuclear factor-{kappa}B in up-regulation. Biochem J 366: 203–209, 2002.[ISI][Medline]
  36. Rise ML, von Schalburg KR, Brown GD, Mawer MA, Devlin RH, Kuipers N, Busby M, Beetz-Sargent M, Alberto R, Gibbs AR, Hunt P, Shukin R, Zeznik JA, Nelson C, Jones SRM, Smailus DE, Jones SJM, Schein JE, Marra MA, Butterfield YSN, Stott JM, Ng SHS, Davidson WS, and Koop BF. Development and application of a salmonid EST database and cDNA microarray: data mining and interspecific hybridization characteristics. Genome Res 14: 478–490, 2004.[Abstract/Free Full Text]
  37. Rodger HD and Drinan EM. Observation of a rickettsia-like organism in Atlantic salmon, Salmo salar L., in Ireland. J Fish Dis 16: 361–369, 1993.[ISI]
  38. Smith PA, Pizarro P, Ojeda P, Contraras J, Oyanedel S, and Larenas J. Routes of entry of Piscirickettsia salmonis in rainbow trout Oncorhynchus mykiss. Dis Aquat Org 37: 165–172, 1999.[ISI][Medline]
  39. Solon E, Gupta AP, and Gaugler R. Signal transduction during exocytosis in Limulus polyphemus granulocytes. Dev Comp Immunol 20: 307–321, 1996.[CrossRef][ISI][Medline]
  40. Stafford JL, Galvez F, Goss GG, and Belosevic M. Induction of nitric oxide and respiratory burst response in activated goldfish macrophages requires potassium channel activity. Dev Comp Immunol 26: 445–459, 2002.[CrossRef][ISI][Medline]
  41. Stafford JL and Belosevic M. Transferrin and innate immune response of fish: identification of a novel highly conserved mechanism of macrophage activation. Dev Comp Immunol 27: 539–554, 2003.[CrossRef][ISI][Medline]
  42. Suzuki T, Hashimoto S, Toyoda N, Nagai S, Yamazaki N, Dong H, Sakai J, Yamashita T, Nukiwa T, and Matsushima K. Comprehensive gene expression profile of LPS-stimulated human monocytes by SAGE. Blood 96: 2584–2591, 2000.[Abstract/Free Full Text]
  43. Taddeo B, Luo TR, Zhang W, and Roizman B. Activation of NF-{kappa}B in cells productively infected with HSV-1 depends on activated protein kinase R and plays no apparent role in blocking apoptosis. Proc Natl Acad Sci USA 100: 12408–12413, 2003.[Abstract/Free Full Text]
  44. Tato CM and Hunter CA. Host-pathogen interactions: subversion and utilization of the NF-{kappa}B pathway during infection. Infect Immun 70: 3311–3317, 2002.[Free Full Text]
  45. Venketaraman V, Dayaram YK, Amin AG, Ngo R, Green RM, Talaue MT, Mann J, and Connell ND. Role of glutathione in macrophage control of mycobacteria. Infect Immun 71: 1864–1871, 2003.[Abstract/Free Full Text]