Physiol. Genomics Journal of Applied Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Physiol. Genomics 31: 126-138, 2007. First published June 12, 2007; doi:10.1152/physiolgenomics.00068.2007
1094-8341/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
31/1/126    most recent
00068.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de la Vega, E.
Right arrow Articles by Degnan, B. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de la Vega, E.
Right arrow Articles by Degnan, B. M.
Received 23 March 2007; accepted in final form 8 June 2007.
Physiological Genomics 31:126-138 (2007)
1094-8341/07 $8.00 © 2007 American Physiological Society

Stress-induced gene expression profiling in the black tiger shrimp Penaeus monodon

Enrique de la Vega1,2, Michael R. Hall1, Kate J. Wilson1, Antonio Reverter3, Rick G. Woods4 and Bernard M. Degnan2

1 Australian Institute of Marine Science, Townsville, Queensland, Australia
2 School of Integrative Biology, The University of Queensland, Brisbane, Queensland, Australia
3 Bioinformatics Group, Commonwealth Scientific and Industrial Research Organisation Livestock Industries, Queensland Bioscience Precinct, St. Lucia, Queensland, Australia
4 The Queensland Institute of Medical Research, Royal Brisbane Hospital, Herston, Queensland, Australia


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cultured shrimp are continuously exposed to variable environmental conditions that have been associated with stress and subsequent outbreaks of disease. To investigate the effect of environmental stress on Penaeus monodon gene expression, a 3,853 random cDNA microarray chip was generated with clones originating from six stress-enriched hemocyte libraries generated by suppression subtractive hybridization and a normal hemocyte cDNA library. Changes in temporal gene expression were analyzed from shrimp exposed to hypoxic, hyperthermic, and hypoosmotic conditions; 3.1% of the cDNAs were differentially expressed in response to at least one of the environmental stressors, and 72% of the differentially expressed clones had no significant sequence similarity to previously known genes. Among those genes with high identity to known sequences, the most common functional groups were immune-related genes and non-long terminal repeat retrotransposons. Hierarchical clustering revealed a set of cDNAs with temporal and stress-specific gene expression profiles as well as a set of cDNAs indicating a common stress response between stressors. Hypoxic and hyperthermic stressors induced the most severe short-term response in terms of gene regulation, and the osmotic stress had the least variation in expression profiles relative to the control. These expression data agree with observed differences in shrimp physical appearance and behavior following exposure to stress conditions.

cDNA microarray; stress response; immune genes; retrotransposons


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
CULTURED SHRIMP ARE CONTINUOUSLY exposed to variations in environmental conditions such as water temperature, dissolved oxygen, and salinity levels, some of which can cause stress. In shrimp, the relationship between environmental conditions and a modification of the immune system is largely unknown (18, 26, 37, 38, 48). To date, research has been focused mainly on measuring susceptibility to bacterial pathogens and quantitative analysis of a limited number of immune parameters, such as total and differential hemocyte counts, modulation of the prophenoloxidase activity, and release of reactive oxygen intermediates (reviewed in Refs. 25, 37). In contrast, work on the relationship between environmental conditions or stressors and susceptibility to viral infections in shrimp is limited, although increasing in recent years, with one report describing the relationship between salinity and infectious hypodermal and hematopoietic necrosis virus (IHHNV) in Litopenaeus vannamei (5), one report describing the effect of acute salinity change on white spot syndrome virus (WSSV) outbreaks in Fenneropenaeus chinensis (40), one report describing the effect of Taura syndrome virus infection on salinity tolerance in L. vannamei (42), a few reports on the effect of water temperature on the pathogenicity (22, 24, 65) and replication (23) of WSSV in L. vannamei, one report describing the effect of water temperature on IHHNV replication in L. vannamei (45), and one report describing the effect of different types of stress on gill-associated virus (GAV) titer in Penaeus monodon (13).

A number of genes involved in the shrimp innate immune system have been identified recently and characterized, including those involved in hemolymph coagulation (28, 60), pattern recognition receptors (56), antibacterial proteins and peptides (3, 15, 58), and a limited number of proteins with antiviral properties (43, 68). Despite significant advances in the identification and characterization of these genes and the wide range of physiological information available on the link between environmental stress and some indicators of immune vigor (e.g., total hemocyte counts, prophenoloxidase activity, and quantitation of reactive oxygen species), the influence of different stressors on gene expression in shrimp has been understudied.

The analysis of differential gene expression in shrimp exposed to environmental stress can reveal adaptive mechanisms to stress. Stress-induced expression of immune response genes may influence the shrimp's capacity to fight pathogens. Microarray analysis allows the expression profiling of thousands of transcripts simultaneously, some of which can be of unknown sequences (41, 57). To address the genetic changes underlying the effect of environmental stress on the general health status and the immune vigor of shrimp, we developed a cDNA microarray for P. monodon hemocytes and used it to interrogate the effect of short-term environmental stress exposure on gene expression in the shrimp. This study describes, to our knowledge, the first utilization of cDNA microarray technology for the study of temporal changes in gene expression profiles of shrimp exposed to different types of stress.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Environmental Stress Exposures for Construction of cDNA Libraries
Shrimp exposure to environmental stress was performed as described previously (13). Briefly, three different types of environmental stress were applied in parallel to the shrimp on days 1 and 8 of postacclimation. There was a 7-day recovery period between stress exposures. For the hyperthermic treatment, water temperature was raised from 29 to 35°C (at a rate of 2°C/h), and shrimp were kept at this elevated water temperature for 24 h, after which the temperature was reduced slowly to 29°C (at a rate of 1°C/h). For osmotic exposure, the salinity in three additional tanks was reduced from 35 to 10 parts per thousand (ppt), and shrimp were kept under these hypoosmotic conditions for 8 h, at which time the salinity was returned to normal. For hypoxic exposure, the oxygen level in three of the tanks was reduced to 1 part per million (ppm), and shrimp were kept under these hypoxic conditions for 8 h, at which time the air stones were returned into the tanks. Control tanks were kept at 29°C, 35 ppt salinity, and under continuous aeration.

Shrimp (average weight 15.9 g) were sampled from the experimental tanks (3 animals/tank) immediately following the second exposure on day 8, before restoration of the normal state. The weight and molting stage were recorded for each individual shrimp as previously described (55), and hemolymph (~400 µl) was withdrawn from the ventral sinus with a 1-ml syringe and a 21-gauge needle containing 500 µl of ice-cold anticoagulant solution (64) and transferred into a 1.5-ml centrifuge tube. Hemocytes were separated from the plasma by centrifugation at 750 g for 4 min, the plasma was discarded, and the hemocytes were snap-frozen in liquid nitrogen.

Environmental Stress Exposures for Microarray Analysis
For microarray analysis, a second group of juvenile P. monodon (average weight 14.9 g) was collected from a local shrimp farm, acclimated for 5 days in 12 100-liter tanks, and exposed to the same 3 environmental stressors described above with some minor modifications. The stress exposure was carried out only once, and the hyperthermic exposure was carried out at 35.5°C.

Shrimp were sampled after acclimation (time 1; T1) (1 animal/tank), immediately following the first challenge on day 1 (T2) (2 animals/tank), after a 24-h short recovery (T3) (2 animals/tank), and at the end of the experiment (T4) (day 10 postacclimation) (all remaining animals). Hemocyte sampling was carried out as described above.

Total RNA Extraction
RNA extraction was performed as previously described (13). Briefly, hemocytes were homogenized in TRIzol-LS (Invitrogen), and RNA from individual shrimp was isolated according to the manufacturer's instructions. Total RNA was DNase treated using DNA-free (Ambion) according to the manufacturer's instructions, ethanol precipitated, and quantified spectrophotometrically. Total RNA integrity was checked by agarose gel electrophoresis.

Construction of cDNA Libraries
Suppression subtractive hybridization (SSH) (17) was used to generate cDNA libraries enriched for genes that are regulated under different types of environmental stress exposure. Double-stranded cDNA was prepared from 2 µg of poly(A) RNA extracted from shrimp hemocytes (equal amounts of RNA from 3 shrimp were pooled together for cDNA preparation) and digested with RsaI, and tester and driver cDNAs were generated for treatments and controls using the PCR-Select cDNA Subtraction kit (Clontech) according to the manufacturer's instructions. The PCR-amplified cDNA fragments generated by SSH were then ligated into the plasmid pGEM-T Easy (Promega) and used to transform High Efficiency JM109 competent cells (Promega) according to the manufacturer's instructions. Six different SSH cDNA libraries were constructed, enriching for both up- and downregulated genes during osmotic, hypoxic, and hyperthermic environmental stress exposure. A total of 1,248 random recombinant colonies were handpicked, cored, and further amplified in 96-well plates with LB broth with 100 µg/ml ampicillin.

A nonnormalized full-length enriched cDNA hemocyte library in {lambda}TriplEx2 was also constructed from hemocytes of a single hyperthermically stressed shrimp using the SMART cDNA library construction kit (Clontech) following the manufacturer's instructions for long-distance PCR. The cDNA ligated into {lambda}TriplEx2 arms was packaged using the Gigapack III Gold packaging system (Stratagene) and the phage used to infect Escherichia coli, strain XL1-blue cells. A total of 2,688 random recombinant plaques were cored and the phage eluted in SM buffer (Stratagene).

PCR Amplification and Construction of cDNA Microarray
The insert from each of the 3,936 cDNA fragments from the SSH libraries and the nonsubtracted hemocyte library was PCR amplified using either plasmid TriplEx2 (pTriplEx2) sequencing forward and reverse primers (Clontech) for clones from the pTriplEx2 cDNA library or SSH nested primers-1 and -2R (Clontech) for clones from the SSH libraries. PCR products were assessed by agarose gel electrophoresis (27), isopropanol precipitated, washed with 70% ethanol, and resuspended in 60 µl of 4x SSC-0.1% sarkosyl. The 4,032 DNA elements composed of PCR products from the different cDNA libraries and a series of controls [complementary data are publicly available at the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus, project no. GPL4890] were spotted in duplicate onto polylysine-coated glass slides using a GMS 417 Arrayer robot (Genetic Microsystems). Postprocessing of the slides was accomplished as previously described (66).

Microarray Processing
RNA from different individual shrimp exposed to the same treatment was pooled together into 23 different groups (Table 1) and then amplified by use of the Message Amp II amplified RNA (aRNA) amplification kit (Ambion), following the manufacturer's instructions for one-round amplification.


View this table:
[in this window]
[in a new window]

  		Table 1. Pool composition for aRNA amplification and microarray hybridization

 
Labeled cDNA was prepared as described by Woods et al. (66), with some modifications. Briefly, 2 µg of aRNA were reverse transcribed using an oligo(dT18) primer (50 µM), 1.25 mM dNTP mix (Biotech International) with a ratio of 4:1 aminoallyl-dUTP to dTTP (Sigma), 10 mM DTT, 600 units of Superscript III (Invitrogen), and 60 units of RNAseOUT (Invitrogen) in a 30-µl reaction volume. RNA was hydrolyzed by the addition of 3 µl of 1 M NaOH and 3 µl of 0.5 M EDTA, followed by heating to 65°C for 15 min. The solution was neutralized by the addition of 3 µl of 1 M HCl. Following synthesis, aminoallyl-labeled cDNA (aa-cDNA) was purified using the Millipore Montage PCR cleanup kit, following the manufacturer's instructions, and dried down, and the pellet was resuspended in 4.5 µl of freshly prepared 100 mM sodium carbonate (pH 9.0). A total of 100 ng of Cy3 or Cy5 reactive dyes (resuspended in 4.5 µl of DMSO) (Amersham) were added to the respective aa-cDNAs, and the coupling reaction was allowed to proceed at room temperature for 1.5 h in the dark. Unincorporated dyes were removed from the reaction by the addition of 4 M hydroxylamine. Cy3- and Cy5-labeled cDNA was then combined and purified using a Qiagen QIAquick PCR cleanup kit.

Human Cot1 DNA (10 µg, Invitrogen) and 2 µg of poly(dA) (Roche) were added to the purified probe mix and dried before resuspension in 30 µl of 4x SSC (0.6 M sodium chloride and 60 mM sodium citrate), 40% deionized formamide, and 0.1% SDS. Following incubation at 95°C for 5 min, and then 45°C for 90 min, the labeled probe mix was loaded onto the microarray chip, and hybridization was carried out for 16 h at 45°C. Following hybridization, the microarray chip was washed twice with 0.2x SSC-0.05% SDS and two more times with 0.2x SSC. Slides were then dried by centrifugation at 600 rpm for 5 min. Finally, Cy3 and Cy5 fluorescence hybridized to the cDNA elements spotted onto the microarray chip was detected with a GMS-418 Array Scanner (Genetic Microsystems) using Imagene 4.2 software (BioDiscovery).

An even-loop hybridization design (34) with 2 independent (replicate) loops and 12 slides per loop was used to perform a time series comparison of the gene expression profiles in juvenile shrimp subjected to different types of stress (Fig. 1). Both loops contained different pooled aRNA sample sets to provide biological replication and estimation of population variability (31–33).


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 1. Schematic diagram of the even-loop design for the time series analysis. Each circle represents a pool of samples, described in Table 1. Each arrow represents an individual hybridization on a single slide, with the tail of the arrow being the RNA labeled with Cy5 and the head representing the RNA labeled with Cy3. This loop analysis was performed twice for a total of 24 separate hybridizations.

 
Data Acquisition, Filtering, and Normalization
Only genes with a signal-to-noise ratio (computed by dividing the background-corrected intensities by the standard deviation of the background pixels) >1.0, scored in both dyes and with at least 24 readings across the 24 slides, were used in this study. Signal intensities were base-2 logarithm (log2) transformed and normalized by a multivariate mixed-model approach, as previously described (51), by the following equation

Formula 1(1)
where y is the vector of all the intensity signals (N = 202,636), background corrected and log2 transformed; X is the incidence matrix relating signals in y to systematic fixed effects in ß, including array slide, printing block, and fluorescent dye channel; ZE1 is the incidence matrix relating y to random effects in c corresponding to the clones (N = 2,841) printed on the array that passed the filtering criteria; ZE2 is the incidence matrix relating y to random effects in a corresponding to the three-way interaction of clone by array and printing block; ZE3 is the incidence matrix relating y to random effects in d corresponding to the interaction of clone by the two fluorescent dye channels; ZE4 is the incidence matrix relating y to random effects in t corresponding to the interaction of clone by the 12 experimental treatment conditions; and e is the random error associated with signals in y.

The variance components were estimated by restricted maximum likelihood, using VCE4 software (available at http://www.tzv.fal.de/~eg/vce4/vce4.html). The fitting of the above model provided best linear unbiased prediction (BLUP) solutions that were the basis for the normalization methods.

Identification of Differentially Expressed Genes
Five different contrasts of interest were analyzed: contrast 1, gene expression pattern in the control group at different time points (T1, T2 + T3, and T4); contrast 2, gene expression pattern in the osmotically stressed group at different time points (T1, T2, T3, and T4); contrast 3, gene expression pattern in the hypoxically stressed group at different time points (T1, T2, T3, and T4); contrast 4, gene expression pattern in the hyperthermically stressed group at different time points (T1, T2, T3, and T4); and contrast 5, gene expression profile between different types of stress and controls at T2 (immediately following stress).

The list of differentially expressed genes for each particular contrast of interest was determined by comparing the solutions of the BLUP model between sampling points or varieties (from now on referred to as variety) for each particular gene (i) according to the multivariate equation described below, where for each gene in i, the difference between the solutions in A (variety 1; e.g., hyperthermic stress at T2) and B (variety 2; e.g., controls at T2) provides a measure of the expression differences between A and B (Eq. 2)

Formula 2(2)
Large positive di values are likely to belong to genes whose expression is upregulated in variety A and downregulated in variety B, whereas large negative values are likely to belong to genes whose expression is downregulated in variety A and upregulated in variety B. The gene solutions were compared among the varieties in each contrast of interest, and the 3.1% most extreme values (either negative or positive) were selected as differentially expressed genes. This cut-off value of 3.1% was selected from the total variance attributed to the gene by variety interaction (see RESULTS).

To compensate for the false discovery rate that results from analyzing multiple contrasts, the list of differentially expressed genes selected for each individual contrast was divided by 5 (total no. of contrasts), selecting the genes with the most extreme changes in expression. This adjusted list of differentially expressed genes was used for further analysis.

Cluster Analysis
Gene expression patterns were identified by unsupervised hierarchical cluster analysis using a colored representation of the data matrix (20) and PermutMatrix software (8) with the Euclidean distance clustering algorithm.

Array Sensitivity
The sensitivity of the microarray experiment was defined as the minimum detectable concentration at which the probability of erroneously detecting a differential gene expression (type I error) equals that of not detecting a genuine differential expression (type II error) and was determined as previously described (52).

Sequencing
cDNA clones that appeared to be differentially expressed were further characterized by unidirectional sequencing. For the differentially expressed genes from the {lambda}-cDNA library, {lambda}TriplEx2 clones were converted into pTripleEx2 by the in vivo excision and circularization of a complete plasmid from the recombinant phage. Plasmids were isolated from single colonies of the subsequent cultures using the Qiagen Miniprep kit according to the manufacturer's instructions. Plasmid samples were directly sequenced from the 5'-end with the Clontech {lambda}TripleEx2F sequencing primer (Clontech), using an ABI3700 sequencer and Big Dye terminator reactions. Clones (PCR products) from the SSH libraries were unidirectionally sequenced using the Clontech SSH nested 1 primer and the DYEnamic ET Terminator cycle sequencing kit (Amersham) in a MegaBase 1000 (Amersham).

Sequences obtained were visually examined for possible sequencing errors; vector and adaptor sequences were removed, and the resulting sequences were checked for redundancy using Sequencher 4.1.4 software (Gene Codes). Sequences with >90% similarity over at least 20 continuous bases were grouped into the same contig and regarded as redundant. The final sequences were exported as text files to Biomanager at the Australian National Genome Information Service (ANGIS; http://www.angis.org.au) and submitted for basic local alignment search tool (BLAST)X, BLASTN, and FASTA searches (1, 47) for matches to known sequences in SwissProt, SpTrEMBL, and GenBank databases. Database searches were limited to expressed sequence tags (ESTs) >200 bp in length, and matches with E-values <10–5 were considered significant. Open reading frame finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html) was used to identify the size of potential open reading frames in the differentially expressed genes.

Quantitative Real-Time PCR for Selected Genes
To confirm the microarray results, real-time quantitative (q)RT-PCR was used to evaluate the expression levels of six different genes in hemocytes of shrimp subjected to environmental stress. Experimental cDNAs from hemocytes of shrimp exposed to different stressors (same RNA samples used in the microarray hybridization) were generated as previously described (13). Briefly, 400 ng of DNase-treated total RNA were reverse transcribed with 2.5 µM random hexamer primers in a 20-µl reaction volume. Following reverse transcription, all cDNAs were diluted fourfold in nuclease-free water and stored at –20°C.

Real-time qRT-PCR reactions were performed in duplicate, and each of them contained 2 µl of diluted cDNA (20 ng of total RNA equivalents), 1x Platinum SYBR Green I qPCR Supermix buffer (Invitrogen), and 300 nM forward and reverse primers (Table 2) in a 15-µl reaction volume. The amplification profile consisted of an initial denaturation step at 95°C for 2 min and then 40 cycles of 95°C for 10 s, 60°C for 15 s, and 72°C for 20 s (acquiring), followed by a melt from 72 to 95°C with a 5-s hold for each step in a Rotor-Gene 3000-G thermocycler. The resulting profile was analyzed using Rotor-Gene 6.0.14 software (Corbett Research). Amplification efficiencies for all qRT-PCR primers were determined as described previously (49). The relative mRNA expression levels for all of these genes in shrimp subjected to different types of stress were determined with Q-Gene software (46) using ED501 (GenBank accession no. EG026283) as the internal reference (normalizer) gene. This gene, the putative function of which in shrimp could not be assigned because of a lack of similarity to other genes in the public databases (no hits by BLAST analyses), was initially identified with the microarray analysis as constantly expressed in shrimp hemocytes by selecting those genes with the smallest variation (standard deviation) in expression levels between treatments and varieties. Its expression levels in hemocytes of shrimp exposed to different treatments were further analyzed by real-time qRT-PCR.


View this table:
[in this window]
[in a new window]

  		Table 2. Oligonucleotide primer sequences for real-time quantitative PCR

 
To test the biased effects of sample pooling and aRNA amplification in the isolation of differentially expressed genes, both pooled amplified aRNA and pooled unamplified RNA (same pools used for the aRNA amplification) and individual unamplified RNA samples from the same experimental animals were analyzed for their relative hemocyanin expression using real-time qRT-PCR. cDNAs from the pooled aRNA samples were produced as described above.

Biological Interpretation
The biological interpretation of the results was determined by classifying genes that are differentially expressed into functional classes as described in the Gene Ontology database (http://www.geneontology.org).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Shrimp Survival and General Condition During the Experiment
There were no significant differences in survival among treatments (Kaplan-Meier, chi square = 1.8, P = 0.178), with all tanks showing survival close to 100%. Typical signs of stressed shrimp such as reduction in food consumption and red coloration were observed only during the stress exposure for the hypoxic and thermal stress treatments. No red coloration, lack of appetite, or tail degeneration was observed in the hypoosmotic treatment during exposure or in any of the experimental shrimp once the stress was removed.

Generation and Characterization of the P. monodon cDNA Microarray
A shrimp cDNA microarray chip containing 4,032 spots in duplicate was produced by printing a series of PCR products from P. monodon hemocyte cDNA libraries (6 SSH cDNA libraries and 1 full-length cDNA library in {lambda}TripleEx2) and a series of controls. Some clones that produced poor-quality PCR products were reamplified, and both the original as well as the reamplification products were included on the microarray. Therefore, some clones were printed four times rather than twice, resulting in a microarray chip containing 3,853 individual clones (from now on referred to as elements).

Of the 3,853 elements in the array, 2,841 (73%) passed the filtering and editing criteria. Additional information for data quality such as median-to-mean correlation (62) and scatter and M-A plots (19, 67) for each hybridization is available as supplemental data at the NCBI Gene Expression Omnibus (GPL4890 and GSE7456) and at ftp://ftp.aims.gov.au/pub/SHRIMP%20MICROARRAY/GEXEX_Database/Introduction.htm.

After log2 transformation and normalization of data by multivariate mixed-model ANOVA, the variance component for each random factor in the model, as well as the proportion of total variance accounted for by each factor, was estimated. The mixed-model analysis estimated a residual of 4.80 and a 95.4% goodness of fit. The total variance accounted for by genes was 89.1%, for the gene x array and printing block interaction 3.1%, for the gene x dye interaction 0.1%, and for the gene x variety (treatment-day) interaction 3.1%.

The 3.1% total variance attributed to the gene-by-variety interaction is equivalent to the amount of total variation that can be attributed to the differentially expressed genes (53). As a result, it was determined that 3.1% of the genes were differentially expressed.

The list of differentially expressed cDNA clones was determined from the BLUP solutions for each gene in different varieties being contrasted. After compensation for multiple testing, 145 clones were selected as differentially expressed.

The P. monodon hemocyte cDNA microarray had a sensitivity of 35 transcripts per million (tpm) determined from the point of equilibrium between the distribution of transcript abundance and the expression intensity readings observed for the differentially expressed clones.

Expression Response to Environmental Stress
Sequence analysis of the 145 clones that are differentially expressed in response to environmental stress revealed that 72% of these clones had no significant similarity to any previously reported sequences, and an additional 11% had significant similarities to genes in the public databases with unknown function (Fig. 2). Among those clones for which a putative function could be assigned, the most common groups were those with immune-related functions and polymerase (POL) from non-long terminal repeat (non-LTR) retrotransposons (Table 3).


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 2. Functional distribution of differentially expressed genes by similarity analysis. A: immune related. B: protein synthesis. C: mitochondrial. D: retrotransposons. E: microsatellites. F: unknown function. G: no match.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Expression patterns of differentially expressed genes with sequence similarity to sequences in the public databases

 
Hypoxic stress.
Two different contrasts were involved in the identification of differentially expressed clones in response to hypoxic stress. For the time series contrast, 75 clones were differentially expressed. Among the few of these clones for which a putative function could be assigned through sequence similarities, crustin (C0960) was downregulated and transglutaminase was upregulated immediately following stress (T1 to T2). In addition, a human-like protease inhibitor with a whey acidic protein domain (clones P0810, P1609, and P1646) and a P. japonicus-like protease inhibitor (clone C0968) were downregulated between T3 and T4.

For the hypoxic-to-control contrast at T2, 26 clones were identified as differentially expressed. From those for which a putative function could be assigned, hemocyanin (clones C0065, C1551, and C1613) and crustin (C0960) were downregulated, whereas 16S mitochondrial genes (clones P1214 and P2202) were upregulated, in response to hypoxic stress.

Osmotic stress.
For the time series contrast, 69 cDNA clones were identified as differentially expressed in response to osmotic stress. Among those clones for which a putative function could be assigned, five corresponded to hemocyanin (clones C0065, C1551, C1552, C1596, and C1613), which was upregulated between T1 and T2; one corresponded to transglutaminase (P1837), which was upregulated during the short-term recovery (T2 to T3); and three had matches to the POL region of non-LTR retrotransposons (clones C0959 and C1849, similar to a non-LTR retrotransposons in Drosophila melanogaster, and C1666, similar to MosquI-Aa2 in Aedes aegypti). These non-LTR retrotransposon-like clones were downregulated immediately following hypoosmotic exposure (T1 to T2) followed by an upregulation during the short recovery (T2 to T3).

For the osmotic-to-control contrast at T2, a strong downregulation in gene expression was observed in 44 of the 45 differentially expressed cDNA clones. Among those clones for which a putative function could be assigned, transglutaminase (P1837) and three non-LTR retrotransposon genes (C0959, C1666, and C1849) were downregulated. Hemocyanin was the only gene whose expression levels were upregulated in response to osmotic stress.

Hyperthermic stress.
For the time series contrast, 68 cDNA clones were identified as differentially expressed. Four of these clones corresponded to hemocyanin (C1551, C1552, C1596, and C1613), which was downregulated during the short recovery period (T2 to T3) followed by an upregulation during the long recovery period (T3 to T4). One clone (C0858), similar to the POL region of a non-LTR retrotransposon from D. melanogaster, was downregulated in response to thermal stress (T1 to T2). Two additional retrotransposons (C0032, similar to a non-LTR retrotransposon from D. melanogaster, and C0771, similar to an I factor from D. teissieri) were upregulated only between T3 and T4 (long recovery period). Lysozyme (clone C0041) was downregulated in response to hyperthermic exposure (T1 to T2) followed by an upregulation during the short recovery period (T2 to T3). Ribosomal S-24 protein (clone C0453) was upregulated during the long recovery phase (T3 to T4).

For the hyperthermic-to-control contrast at T2, 30 clones were identified as differentially expressed, 14 of which were downregulated in response to stress. Downregulated genes for which a putative function could be assigned included lysozyme (C0041) and three clones with similarity to non-LTR retrotransposons (C0858, C0032, and C0071).

Controls.
For the control group, 49 clones were identified as differentially expressed through time. Among those genes for which a function could be assigned, a non-LTR retrotransposon (C0032) was downregulated between T3 and T4, hemocyanin (C0065, C1551, C1596, and C1613) was upregulated between T1 and T2T3 (pool of samples from T2 and T3), a protease inhibitor (P0510) was upregulated between T3 and T4, and 16S mitochondrial sequences (P1214 and P2202) were upregulated between T3 and T4.

Common Stress Response Between Treatments
Gene expression analysis immediately following stress exposure (T2) revealed a common stress response among various treatments. For hypoxic and hyperthermic exposures, there were eight differentially expressed cDNA clones having the same patterns of expression. Four clones (C0386, C0406, C1008, and C1012) were downregulated in both treatments, and four clones (P0552, P0930, P1741, and P2868) were upregulated in both treatments. The expression levels for these eight cDNA clones were not modified following the osmotic exposure.

There was also a common stress response between osmotic and hyperthermic exposures, with six clones (C0959, C1666, C1849, P1520, P2234, and P2312) being downregulated in response to stress for both treatments (see GoFig. 4B).


Figure 3
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 3. Unsupervised hierarchical cluster (Euclidean distance clustering algorithm) for the 3 treatments and control groups. Analysis of 145 differentially expressed clones in the time course series. Columns indicate separate time points, and every row displays the expression profile of a single clone. Dendograms at top cluster the time points based on the relatedness of their gene expression patterns. Values from color scale correspond to the log2-transformed signal intensity (min, minimum; max, maximum).

 

Figure 4
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 4. Unsupervised hierarchical cluster for different treatments immediately following stress exposure (T2) (see MATERIALS AND METHODS for descriptions of T1–T4). A: analysis of 145 differentially expressed clones for the different treatments. Columns indicate separate treatments, and every row displays the expression profile of a single clone. Dendogram at left clusters the clones based on the relatedness of their gene expression patterns. Dendogram at top clusters the treatments based on the relatedness of their gene expression patterns. B: hierarchical cluster analysis for cDNA clones involved in the common stress response. C, control group; O, osmotic stress; H, hypoxic stress; T, hyperthermic stress. Values from color scale correspond to the log2-transformed signal intensity.

 
There was no common stress response between osmotic and hypoxic treatments.

Hierarchical Cluster Gene Expression Analysis in Response to Stress
To analyze and interpret the data, stress-responsive genes were clustered for each treatment (osmotic, hypoxic, hyperthermic, and controls) according to their temporal expression patterns using a hierarchical clustering program (Fig. 3). This hierarchical clustering of the temporal gene expression patterns for each treatment revealed significant differences in the way various types of environmental stressors affect transcription through time in shrimp hemocytes. Transcriptional analysis through hierarchical clustering according to treatment immediately following stress (Fig. 4A) resulted in two separate clusters: one for the control and hypoosmotic stress treatment and the other for the hypoxic and hyperthermic treatments, revealing global similarities in transcription between those treatments within a group or cluster.

Confirmation of Microarray Data by Real-Time qRT-PCR
Clone ED501, identified by microarray analysis as constitutively expressed between treatments, was selected as a normalizer for real-time qRT-PCR analysis of the differentially expressed genes. The constant expression of clone ED501 in hemocytes of shrimp exposed to different types of environmental stress was confirmed by real-time qRT-PCR (data not shown).

Six different putative genes classified as differentially expressed by microarray analysis were selected, and their differential expression in shrimp exposed to environmental challenges was evaluated by real-time qRT-PCR. All six genes analyzed (hemocyanin, lysozyme, 2 different retrotransposons, a clone with a significant similarity to an Anopheles gambiae gene with unknown function, and 1 clone with no significant match to sequences in the public databases) were confirmed as being differentially expressed in response to environmental stress (Fig. 5, A–H).


Figure 5
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 5. Differential expression of various shrimp genes in response to environmental stress as determined by microarray (A, C, E, G, I) and quantitative (q)RT-PCR analyses (B, D, F, H, J). MosquI-Aa2 non-long terminal repeat (non-LTR) retrotransposon-like gene (ED363) expression in response to various types of stress as determined from microarray (A) and qRT-PCR analysis (B). Differential expression of an unknown function gene (contig F) in response to hypoxic stress as determined by microarray (C) and qRT-PCR analysis (D). Differential expression of lysozyme in response to hyperthermic stress as observed with microarray (E) and qRT-PCR analysis (F). Differential expression of a non-LTR retrotransposon-like gene (clone C0858) in response to different types of stress as observed by microarray (G) and qRT-PCR analysis (H). Differential expression of an unknown function gene (contig AQ) in response to various types of stress as determined from microarray (I) and qRT-PCR analysis (J). Bars represent mean expression values ± SD.

 
A similar gene expression pattern was observed for hemocyanin with both microarray and real-time qRT-PCR analysis. The pooling of samples and linear amplification of aRNA did not cause a significant bias in the relative transcript abundance for hemocyanin (Fig. 6, A–C).


Figure 6
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 6. Differential expression of hemocyanin in response to environmental stress. A: microarray gene expression data for pooled aRNA samples in the time series experiment. Bars represent mean normalized intensity values for 4 separate clones with significant similarity to Penaeus monodon hemocyanin (±SD). B: qRT-PCR data for hemocyanin expression in the pooled aRNA samples (2 separate pools of 3 shrimp each) during the time series experiment. Bars represent the mean normalized expression values for 2 separate pools of 3 shrimp/pool. C: qRT-PCR data for hemocyanin expression using 6 individual nonamplified RNA samples per time point. Control samples at T2 and T3 are represented as 2 separate bars, although they correspond to the same group of samples because of the pooling before aRNA amplification and microarray analysis (refer to Table 1).

 
The same significant differences in transcript abundance among treatments and/or time points were observed for all genes, regardless of the analysis method (microarray or real-time qRT-PCR), except for the hyperthermic pattern in the MosquI-Aa2 non-LTR retrotransposon (ED363), in which a significant downregulation was observed between T1 and T2 with microarray analysis but not with real-time qRT-PCR (Fig. 5).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
In this study, a first-generation cDNA microarray from P. monodon hemocytes was developed to study the impact of three types of stress, osmotic, hypoxic, and hyperthermic, on gene expression in shrimp. Because the emphasis was on the impact of stress on the immune response genes, circulating hemocytes, which play a major role in shrimp immunity (63), were used for construction of cDNA libraries and microarray analysis in this study.

Technical Aspects of the Microarray
Various parameters of microarray quality control, such as spot morphology (62), dye intensity biases or M-A plots (19), and low level of variability between replicate samples of the same experimental condition and in self hybridizations, revealed the high quality of the data being generated with this shrimp microarray. A computational method to determine microarray sensitivity without the need to use spiked controls (52) revealed a sensitivity of 35 tpm. This value is below the 250 tpm observed in a mouse microarray (7), the 50 tpm observed in a human cell microarray (39), and values between 100 and 400 tpm reported by Reverter et al. (51) but higher than the Affymetrix short-oligo arrays with reported sensitivities in the range of 3–20 tpm (10). The lower the value in transcripts per million, the more sensitive a microarray is and the higher the likelihood of identifying differentially expressed clones from among rare gene transcripts (Reverter et al., Ref. 52). Hence, the relatively low values or "good" sensitivity obtained with the shrimp array is expected to allow the identification of differential expression even for rare transcripts.

Biological Interpretation of Microarray Analysis
The low number of differentially expressed genes for which a putative function could be assigned limited a biological interpretation to their expression patterns. Nevertheless, cluster analysis revealed some characteristic stress-specific responses. A significant overall reduction of mRNA transcript abundance (referred to as downregulation) in gene expression was observed in the osmotic stress with 44 of 45 differentially expressed clones being downregulated following the hypoosmotic exposure. It has been shown in shrimp that high-energy expenditure is required under hypoosmotic conditions to maintain a hyperosmotic extracellular fluid (50). Therefore, this global repression in gene expression may suggest an energy-saving mechanism to compensate for the amount of energy required for osmoregulation. A similar downregulation of gene expression associated with a reduction in energy consumption has been previously suggested in zebrafish exposed to hypoxia (61).

Hemocyanin is the only gene upregulated (significant increase in mRNA transcript abundance) in response to the osmotic challenge, suggesting the vital importance of this protein in the physiological adaptation to hypoosmotic exposure. An increase in the concentration of hemolymph protein, of which hemocyanin is the main component (36), has been observed in crustaceans as a result of exposure to low salinities (59). A rapid twofold increase of hemocyanin concentration in the circulating hemolymph was observed in the crab Carcinus maenas when exposed to a hypoosmotic stress by reduction of water salinity by one-half (4). In the shore crab C. aestuarii, a reduction in water salinity from 35 to 11 ppt induced a transitory 30% increase in hemolymph copper, protein, and hemocyanin concentrations within 6 h following stress exposure (16). The increase in hemocyanin mRNA transcription in osmotically stressed shrimp identified here by microarray analysis and real-time qRT-PCR and our previous observations by suppression subtractive hybridization using independent shrimp populations suggest that hemocyanin upregulation is conserved in the hypoosmotic stress response of crustaceans. Although the implications of an increase of hemocyanin concentration in the hemolymph remain unclear, it could have significance far beyond oxygen transport, as it has been suggested that hemocyanin is involved in amino acid and protein storage (4) and enzymatic activities and may even have antiviral properties (68). In fact, the finding of high levels of hemocyanin transcripts in shrimp hemocytes following hypoosmotic stress is novel. Hepatopancreas is the main site of hemocyanin production (36), and it is commonly believed that hemocyanin transcription is limited to this organ (54) despite previous observations of hemocyanin transcripts among hemocyte EST collections in the public databases (e.g., GenBank accession nos. CK591319, CK591488). Regulated hemocyanin transcription in hemocytes could be an indication of a novel physiological function of hemocytes that should be further investigated.

Among those genes downregulated in response to a hypoosmotic exposure are three clones with significant similarity to the POL region of non-LTR retrotransposons. The differential expression of these types of elements in response to environmental stress exposures had been observed previously by suppression subtractive hybridization and real-time qRT-PCR analyses using independent shrimp populations exposed to the same types of stress (E. de la Vega, B. M. Degnan, M. R. Hall, and K. J. Wilson, unpublished data). The fact that similar observations were obtained with microarray analysis suggests their importance in the shrimp's stress response to environmental change.

An exposure to hypoxic conditions results in a complex gene expression pattern through time with about the same number of genes being up- and downregulated. Because most of the differentially expressed clones could not be annotated, the interpretation of the results is very limited. From those cDNA clones for which a putative function could be assigned, a few immune-related genes are regulated in response to hypoxic exposure. Transglutaminase, an important gene involved in blood coagulation (28), is upregulated immediately following hypoxic stress. An increase in transglutaminase transcription has been previously observed in surviving shrimp following a Vibrio infection, which is an indication of its importance in the immune response (14). The regulation of other immune-related genes was affected by hypoxic exposure, with a downregulation of crustin immediately following stress and downregulation of two types of protease inhibitors, one with a Kunitz inhibitor domain and the other with a whey acidic protein (WAP) domain, during the long-term recovery phase after hypoxic exposure. Because of the involvement of these proteins in various aspects of microbial control such as blood coagulation, prophenoloxidase cascades, and selective pathogen digestion among others (30), their downregulation in response to hypoxic stress might increase the shrimp's susceptibility to microbial infections. Moreover, the fact that both crustin (3) and one of the differentially expressed proteinase inhibitors contain WAP domains (12), and that both proteins were downregulated in response to hypoxic exposure, suggests that the WAP family of proteins could be involved in the hypoxic stress response in shrimp.

For the hyperthermic stress exposure, a series of novel stress response genes were identified for which a putative function could be assigned to three clones with significant similarity with POL in non-LTR retrotransposons. The expression of these non-LTR retrotransposons was downregulated immediately following the hyperthermic stress and upregulated during the short recovery period. The downregulation of these non-LTR retrotransposons was also observed following an osmotic stress, suggesting their involvement in a common stress response.

There was a common stress response between osmotic and hyperthermic treatments, with the downregulation of the same six clones in both treatments, and between hypoxic and hyperthermic stress, with eight genes having the same patterns of expression between the two treatments. A common stress response has been observed in yeast, where 10% of all genes are equally regulated between different types of environmental stress (9), and in Drosophila, where 38% of the differentially expressed genes display similar responses to oxidative and endoplasmic reticulum stresses (21). Moreover, certain similarities are observed between the common environmental response described in yeast (9) and flies, with heat shock genes, genes involved in the detoxification processes, or genes associated with fatty acid metabolism and DNA repair showing similar patterns in all stress conditions studied (21).

No putative DNA repair, heat shock proteins, or any of the previously characterized hypoxia-responsive genes from crabs (6) were identified among the differentially expressed genes in shrimp. However, this first-generation microarray has <4,000 random cDNA clones from six subtracted and one unsubtracted hemocyte library, which results in gene redundancy. By analyzing only mRNA from shrimp hemocytes with a relatively small number of unique probes (unique genes spotted in the microarray), only a small proportion of the shrimp transcriptome is being evaluated. Moreover, a strict filtering criteria and strategy to compensate for the false discovery rate associated with multiple testing resulted in a conservative list of genes with the most "extreme" differential expression. Therefore, the lack of appearance of DNA repair or heat shock proteins (Hsp) among the differentially expressed genes does not mean they are not regulated in response to environmental stress. This is further supported by previous experiments using anti-Hsp70 antibodies that revealed an upregulation of Hsp70 following hyperthermic stress (11, 13).

Cluster analysis based on time points for each particular stressor revealed some significant temporal differences in the shrimp's response toward the different types of environmental stress exposures. For instance, T1 in the hypoxically stressed shrimp formed a separate cluster from T2, T3, and T4, which suggests a long-term stress response following the low oxygen exposure (Fig. 3). For the osmotic challenge, T1 and T3 were clustered together, with T2 and T4 in separate branches of the clustering tree. These findings suggest that the osmotic challenge is characterized more by a short-term response, with gene expression returning to prestress levels within 24 h postexposure (short-term recovery). The hyperthermic exposure was characterized by a late host response, with an increase in the severity from T2 to T3 and then a return to normal 7 days later (T4), as observed in the three separate clusters. Cluster analysis of the control group reveals differences in expression linked with time, which could be a response to the long-term retention in the experimental tanks. The deleterious effects of holding juvenile shrimp in experimental tanks have been observed previously, with an associated increase in GAV concentration and a low-level constant mortality (13). Further characterization of differentially expressed genes in the time series analysis for the control group may therefore provide some indications of the detrimental changes in shrimp physiology that result from long-term holding in tanks.

Cluster analysis between the different treatments immediately following stress (Fig. 4A) revealed what appear to be differing levels of severity in the short-term host response, with hypoxic and hyperthermic stressors causing the most extreme response, and the osmotic stress having the least variation in expression profiles relative to the control. These expression data agree with the clear differences in shrimp physical appearance and behavior (e.g., red coloration and reduction in food consumption for the hypoxic and hyperthermic treatments) observed during the stress period, suggesting a possible association between the severity of the stressor and body coloration, reduction in food consumption, and gene expression profiles.

A striking feature of the microarray analysis was the differential expression of transposable and repetitive elements in stressed shrimp. Exposure to different stressors has been shown to stimulate the expression of transposable elements (TEs) in yeast, plants, insects, and fish (2, 29, 35, 44). The present observations with microarray analysis reveal that stress can also regulate TE transcription in shrimp. At this stage, the physiological implications of stress-regulated TE transcription would be purely speculative, as there is no evidence of transposition or regulation of surrounding genes. Further characterization of these elements by full-length sequencing and genomic stability studies is required to determine the physiological implication of their stress-regulated transcription.

A similar characterization is also required for most other differentially expressed genes identified in the microarray experiment for which a putative function could not be assigned. Full-length cDNA sequencing of those genes is required before classifying them as novel stress-regulated genes, or perhaps as poorly conserved untranslated regions of previously known genes from better-characterized organisms.

In addition, changes in gene expression associated with environmental stress reported here originate from circulating hemocytes and thus could be influenced by variations in relative abundance of different hemocyte populations. Thus future gene expression studies involving hemocytes should include an estimation of relative abundance of various types of hemocytes by techniques such as differential hemocyte counts.

In summary, a series of novel stress-specific and common stress-responsive genes were identified. A large percentage of these stress-responsive genes were non-LTR retrotransposons, microsatellites, and sequences with short predicted open reading frames (data not shown), suggesting the involvement of noncoding and repetitive elements in environmental stress response. In addition, the observation of a stress-specific host response by microarray analysis that included the regulated expression of various immune-related genes further supports previous reports on the close relationship between environmental stress, host response, and susceptibility to disease.


   GRANTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by Australian Research Council ARC Linkage Grant Project No. LP0455813 and the Australian Prawn Farmers Association.


   FOOTNOTES
 
Address for reprint requests and other correspondence: E. de la Vega, Hollings Marine Laboratory, 331 Ft. Jonson Rd., Charleston, SC 29412 (e-mail: delavega{at}musc.edu)

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


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402, 1997.[Abstract/Free Full Text]
  2. Arnault C, Loevenbruck C, Biemont C. Transposable element mobilization is not induced by heat shocks in Drosophila melanogaster. Naturwissenschaften 84: 410–414, 1997.[CrossRef][Web of Science][Medline]
  3. Barlett TC, Cuthbertson BJ, Shepard EF, Chapman RW, Gross PS, Warr GW. Crustins, homologues of an 11.5-kDa antibacterial peptide, from two species of Penaeid shrimp, Litopenaeus vannamei and Litopenaus setiferus. Mar Biotechnol 4: 278–293, 2002.[CrossRef][Medline]
  4. Boone WR, Schoffeniels E. Hemocyanin synthesis during hypo-osmotic stress in the shore crab Carcinus maenas (L.). Comp Biochem Physiol B 63: 207–214, 1979.[CrossRef]
  5. Bray WA, Lawrence AL, Leung-Trujillo JR. The effect of salinity on growth and survival of Penaeus vannamei, with observations on the interaction of IHHN virus and salinity. Aquaculture 122: 133–146, 1994.[CrossRef][Web of Science]
  6. Brown-Peterson NJ, Larkin P, Denslow N, King C, Manning S, Brouwer M. Molecular indicators of hypoxia in the blue crab Callinectes sapidus. Mar Ecol Prog 286: 203–215, 2005.[CrossRef]
  7. Callow MJ, Dudoit S, Gong EL, Speed TP, Rubin EM. Microarray expression profiling identifies genes with altered expression in HDL-deficient mice. Genome Res 10: 2022–2029, 2000.[Abstract/Free Full Text]
  8. Caraux G, Pinloche S. PermutMatrix: a graphical environment to arrange gene expression profiles in optimal linear order. Bioinformatics 21: 1280–1281, 2005.[Abstract/Free Full Text]
  9. Causton HC, Ren B, Koh SS, Harbison CT, Kanin E, Jennings EG, Lee IT, True HL, Lander ES, Young RA. Remodeling of yeast genome expression in response to environmental changes. Mol Biol Cell 12: 323–337, 2001.[Abstract/Free Full Text]
  10. Chudin E, Walker R, Kosaka A, Wu SX, Rabert D, Chang TK, Kreder DE. Assessment of the relationship between signal intensities and transcript concentration for Affymetrix GeneChip Arrays. Genome Biol 3: RESEARCH0005, 2002.[Medline]
  11. Cimino EJ, Owens L, Anderson TA. A newly developed ELISA showing the effect of environmental stress on levels of hsp86 in Cherax quadricarinatus and Penaeus monodon. Comp Biochem Physiol A 132: 591–598, 2003.
  12. Clauss A, Lilja H, Lundwall A. A locus on human chromosome 20 contains several genes expressing protease inhibitor domains with homology to whey acidic protein. Biochem J 368: 233–242, 2002.[CrossRef][Web of Science][Medline]
  13. de la Vega E, Hall MR, Degnan BM, Wilson KJ. Short-term hyperthermic treatment of Penaeus monodon increases expression of heat shock protein 70 (HSP70) and reduces replication of gill associated virus (GAV). Aquaculture 253: 82–90, 2006.[CrossRef][Web of Science]
  14. de Lorgeril J, Saulnier D, Janech MG, Gueguen Y, Bachere E. Identification of genes that are differentially expressed in hemocytes of the Pacific blue shrimp (Litopenaeus stylirostris) surviving an infection with Vibrio penaeicida. Physiol Genomics 21: 174–183, 2005.[Abstract/Free Full Text]
  15. Destoumieux D, Bulet P, Loew D, Van Dorsselaer AV, Rodriguez J, Bachere E. Penaeidins, a new family of antimicrobial peptides isolated from the shrimp Penaeus vannamei (Decapoda). J Biol Chem 272: 28398–28406, 1997.[Abstract/Free Full Text]
  16. Devescovi M, Lucu C. Copper levels during osmotic stress in the shore crab Carcinus aestuarii. Comp Biochem Physiol C 115: 195–200, 1996.[CrossRef][Web of Science]
  17. Diatchenko L, Lau YFC, Campbell AP, Chenchik A, Moqadam F, Huang B, Lukyanov S, Lukyanov L, Gurskaya N, Sverdlov ED, Siebert PD. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci USA 93: 6025–6030, 1996.[Abstract/Free Full Text]
  18. Direkbusarakom S, Danayadol Y. Effect of oxygen depletion on some parameters of the immune system in black tiger shrimp (Penaeus monodon). In: Advances in Shrimp Biotechnology, edited by Flegel TW. Bangkok, Thailand: National Center for Genetic Engineering and Biotechnology, 1998.
  19. Dudoit S, Yang YH, Callow MJ, Speed TP. Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments. Statistical Sinica 12: 111–139, 2002.
  20. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863–14868, 1998.[Abstract/Free Full Text]
  21. Girardot F, Monnier V, Tricoire H. Genome wide analysis of common and specific stress responses in adult Drosophila melanogaster. BMC Genomics 5: 74, 2004.[CrossRef][Medline]
  22. Granja CB, Aranguren F, Vidal OM, Aragon L, Salazar M. Does hyperthermia increase apoptosis in white spot syndrome virus (WSSV)-infected Litopenaeus vannamei? Dis Aquat Organ 54: 73–78, 2003.[CrossRef][Web of Science][Medline]
  23. Granja CB, Vidal OM, Parra G, Salazar M. Hyperthermia reduces viral load of white spot syndrome virus in Penaeus vannamei. Dis Aquat Organ 68: 175–180, 2006.[Web of Science][Medline]
  24. Guan Y, Yu Z, Li C. The effect of temperature on white spot syndrome infections in Marsupenaeus japonicus. J Invertebr Pathol 83: 257–260, 2003.[CrossRef][Web of Science][Medline]
  25. Hall MR, de la Vega E. Physiological response to stress and health implications in Crustacea. In: Styli 2003: Thirty Years of Shrimp Farming in New Caledonia, edited by Goarant C, Harache Y, Herbland A, and Mugnier C. Plouzane, France: INRA Editions, 2004, p. 38–56.
  26. Hall MR, van Ham EH. The effects of different types of stress on blood glucose in the giant tiger prawn, Penaeus monodon. J World Aquac Soc 29: 290–299, 1998.[CrossRef]
  27. Holloway A, Murphy M, Chandrasekharappa S, Monte D. Visualization and assesment of PCR products. In: DNA Microarrays: A Molecular Cloning Manual, edited by Bowtell D, and Sambrook J. New York: Cold Spring Harbor Laboratory, 2003, p. 23–27.
  28. Huang CC, Sritunyalucksana K, Soderhall K, Song YL. Molecular cloning and characterization of tiger shrimp (Penaeus monodon) transglutaminase. Dev Comp Immunol 28: 279–294, 2004.[CrossRef][Web of Science][Medline]
  29. Ikeda K, Nakayashiki H, Takagi M, Tosa Y, Mayamma S. Heat shock, copper sulfate and oxidative stress activate the retrotransposon MAGGY resident in the plant pathogenic fungus Magnaporthe grisea. Mol Genet Genomics 266: 318–325, 2001.[CrossRef][Web of Science][Medline]
  30. Jarasrassamee B, Supungul P, Panyim S, Klinbunga S, Rimphanichayakit V, Tassanakajon A. Recombinant expression and characterization of five-domain kazal-type serine proteinase inhibitor of black tiger shrimp (Penaeus monodon). Mar Biotechnol 7: 46–52, 2005.[CrossRef][Medline]
  31. Kendziorski CM, Irizarry RA, Chen KS, Haag JD, Gould MN. On the utility of pooling biological samples in microarray experiments. Proc Natl Acad Sci USA 102: 4252–4257, 2005.[Abstract/Free Full Text]
  32. Kendziorski CM, Zhang Y, Lan H, Attie AD. The efficiency of pooling mRNA in microarray experiments. Biostat 4: 465–477, 2003.[CrossRef]
  33. Kerr MK. Design considerations for efficient and effective microarray studies. Biometrics 59: 822–828, 2003.[CrossRef][Web of Science][Medline]
  34. Kerr MK, Churchill GA. Experimental design for gene expression microarrays. Biostatistics 2: 183–201, 2001.[Abstract]
  35. Krasnov A, Koskinen H, Afanasyev S, Molsa H. Transcribed Tc1-like transposons in salmonid fish. BMC Genomics 6: 107, 2005.[CrossRef][Medline]
  36. Lehnert SA, Johnson SE. Expression of hemocyanin and digestive enzyme messenger RNAs in the hepatopancreas of the black tiger shrimp Penaeus monodon. Comp Biochem Physiol B 133: 163–171, 2002.[CrossRef][Medline]
  37. Le Moullac G, Haffner P. Environmental factors affecting immune responses in Crustacea. Aquaculture 91: 121–131, 2000.
  38. Le Moullac G, Soyez C, Saulnier D, Ansquer D, Avarre JC, Levy P. Effect of hypoxic stress on the immune response and the resistance to vibriosis of the shrimp Penaeus stylirostris. Fish Shellfish Immunol 8: 621–629, 1998.[CrossRef][Web of Science]
  39. Lin JY, Pollack JR, Chou FL, Rees CA, Christian AT, Bedford JS, Brown PO, Ginsberg MH. Physical mapping of genes in somatic cell radiation hybrids by comparative genomic hybridization to cDNA microarrays. Genome Biol 3: 261–267, 2002.
  40. Liu B, Yu Z, Song X, Guan Y, Jian X, He J. The effect of acute salinity change on white spot syndrome (WSS) outbreaks in Fenneropenaeus chinensis. Aquaculture 253: 163–170, 2006.[CrossRef][Web of Science]
  41. Lockhart DJ, Winzeler EA. Genomics, gene expression and DNA arrays. Nature 45: 827–836, 2000.
  42. Lotz JM, Salazar LA, Soto MA. Effect of chronic Taura syndrome virus infection on salinity tolerance of Litopenaeus vannamei. Dis Aquat Organ 65: 75–78, 2005.[CrossRef][Web of Science][Medline]
  43. Luo T, Zhang X, Shao Z, Xu X. PmAV, a novel gene involved in virus resistance of shrimp Penaeus monodon. FEBS Lett 551: 53–57, 2003.[CrossRef][Web of Science][Medline]
  44. Mhiri C, Morel JB, Vernhettes S, Casacuberta JM, Lucas H, Grandbastien MA. The promoter of the tobacco Tnt1 retrotransposon is induced by wounding and by abiotic stress. Plant Mol Biol 33: 557–266, 1997.
  45. Montgomery-Brock D, Tacon AGJ, Poulos B, Lightner D. Reduced replication of infectious hypodermal and hematopoietic necrosis virus (IHHNV) in Litopenaeus vannamei held in warm water. Aquaculture 265: 41–48, 2007.[CrossRef][Web of Science]
  46. Muller PY, Janovjak H, Miserez AR, Dobbie Z. Processing of gene expression data generated by quantitative real-time RT-PCR. Biotechniques 32: 1–7, 2002.
  47. Pearson WR, Lipman DJ. Improved tools for biological sequence analysis. Proc Natl Acad Sci USA 85: 2444–2448, 1988.[Abstract/Free Full Text]
  48. Perazzolo LM, Gargioni R, Ogliari P, Barracco MA. Evaluation of some hemato-immunological parameters in the shrimp Farfantepenaeus paulensis submitted to environmental and physiological stress. Aquaculture 214: 19–33, 2002.[CrossRef][Web of Science]
  49. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30: e36, 2002.[Abstract/Free Full Text]
  50. Pillai BR, Diwan AD. Effects of acute salinity stress on oxygen consumption and ammonia excretion rates of the marine shrimp Metapenaeus monoceros. J Crust Biol 22: 45–52, 2002.[CrossRef][Web of Science]
  51. Reverter A, Barris W, McWilliam S, Byrne Ka Wang YH, Tan SH, Hudson N, Dalrymple BP. Validation of alternative methods of data normalization in gene co-expression studies. Bioinformatics 21: 1112–1120, 2005.[Abstract/Free Full Text]
  52. Reverter A, McWilliam SM, Barris W, Dalrymple BP. A rapid method for computationally inferring transcriptome coverage and microarray sensitivity. Bioinformatics 21: 80–89, 2005.[Abstract/Free Full Text]
  53. Reverter A, Wang YH, Byrne K, Tan SH, Harper GS, Lehnert SA. Joint analysis of multiple cDNA microarray studies via multivariate mixed models applied to genetic improvement of beef cattle. J Anim Sci 82: 3430–3439, 2004.[Abstract/Free Full Text]
  54. Robalino J, Bartlett T, Shepard E, Prior S, Jaramillo G, Scura E, Chapman RW, Gross PS, Browdy CL, Warr GW. Double-stranded RNA induces sequence-specific antiviral silencing in addition to nonspecific immunity in a marine shrimp: convergence of RNA interference and innate immunity in the invertebrate antiviral response? J Virol 79: 13561–13571, 2005.[Abstract/Free Full Text]
  55. Robertson L, Bray W, Leung-Trujillo J, Lawrence A. Practical molt staging of Penaeus setiferus and Penaeus stylirostris. J World Aquac Soc 18: 180–185, 1987.[CrossRef]
  56. Roux MM, Pain A, Klimpel KR, Dhar AK. The lipopolysaccharide and B-1,3-glucan binding protein gene is upregulated in white spot virus-infected shrimp (Penaeus stylirostris). J Virol 76: 7140–7149, 2002.[Abstract/Free Full Text]
  57. Shalon D, Smill SJ, Brown PO. A DNA microarray system for analyzing complex DNA samples using two-color fluorescent probe hybridization. Genome Res 6: 639–645, 1996.[Abstract/Free Full Text]
  58. Sotelo-Mundo RR, Islas-Osuna MA, de la Re Vega E, Hernandez-Lopez J, Vargas-Albores F, Yepiz-Plascencia G. cDNA cloning of the lysozyme of the white shrimp Penaeus vannamei. Fish Shellfish Immunol 15: 325–333, 2003.[CrossRef][Web of Science][Medline]
  59. Spicer JI, Baden SP. Environmental hypoxia and haemocyanin between-individual variability in Norway lobsters Nephrops norvegicus. Mar Biol (Berl) 139: 727–734, 2001.
  60. Sritunyalucksana K, Cerenius L, Soderhall K. Molecular cloning and characterization of prophenoloxidase in the black tiger shrimp, Penaeus monodon. Dev Comp Immunol 23: 179–186, 1999.[CrossRef][Web of Science][Medline]
  61. Ton C, Stamatiou D, Liew CC. Gene expression profile of zebrafish exposed to hypoxia during development. Physiol Genomics 13: 97–106, 2003.[Abstract/Free Full Text]
  62. Tran PH, Pfeiffer DA, Shin Y, Meek LM, Brody JP, Cho KWY. Microarray optimizations: increasing spot accuracy and automated identification of true microarray signals. Nucleic Acids Res 30: e54, 2002.[Abstract/Free Full Text]
  63. van de Braak CB, Botterblom MH, Huisman EA, Robmbout JHW, van der Knaap WP. Preliminary study on haemocyte response to white spot syndrome virus infection in black tiger shrimp Penaeus monodon. Dis Aquat Organ 51: 149–155, 2002.[Web of Science][Medline]
  64. Vargas-Albores F, Guzman-Murillo MA, Ochoa JL. An anticoagulant solution for haemolymph collection and prophenoloxidase studies of Penaeid shrimp (Penaeus californiensis). Comp Biochem Physiol A 106: 299–303, 1993.
  65. Vidal OM, Granja CB, Aranguren F, Brock JA, Salazar M. A profound effect of hyperthermia on survival of Litopenaeus vannamei juveniles infected with white spot syndrome virus. J World Aquac Soc 32: 364–372, 2001.[CrossRef]
  66. Woods RG, Roper KE, Gauthier M, Bebell LM, Sung K, Degnan BM, Lavin MF. Gene expression during early ascidian metamorphosis requires signaling by Hemps, an EGF-like protein. Development 131: 2921–2933, 2004.[Abstract/Free Full Text]
  67. Yang YH, Speed T. Design of microarray expression experiments. Part 4. Representing and evaluating slide data. In: DNA Microarrays: A Molecular Cloning Manual, edited by Bowtell D and Sambrook J. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 2003, p. 544–551.
  68. Zhang X, Huang C, Qin Q. Antiviral properties of hemocyanin isolated from shrimp Penaeus monodon. Antiviral Res 61: 93–99, 2004.[CrossRef][Web of Science][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
31/1/126    most recent
00068.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de la Vega, E.
Right arrow Articles by Degnan, B. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de la Vega, E.
Right arrow Articles by Degnan, B. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.