Mar Biotechnol DOI 10.1007/s10126-015-9640-1
ORIGINAL ARTICLE
Impact of Thermal Stress on Kidney-Specific Gene Expression in Farmed Regional and Imported Rainbow Trout Marieke Verleih 1 & Andreas Borchel 1 & Aleksei Krasnov 2 & Alexander Rebl 1 & Tomáš Korytář 3,5 & Carsten Kühn 4 & Tom Goldammer 1
Received: 8 January 2015 / Accepted: 6 May 2015 # Springer Science+Business Media New York 2015
Abstract Seasonal water temperatures can be stressful for fish in aquaculture and can therefore negatively influence their welfare. Although the kidney is the crucial organ associated with the primary stress response, knowledge about the stressmodulated kidney transcriptome in salmonids is limited. In the present study, we used a comparative microarray approach to characterize the general gene expression profiles of rainbow trout trunk kidney after a 2-week acclimation to mild heat (23 °C) and cold stress (8 °C). Hypothesizing that local adaptation influences stress performance, we aimed to identify differences in the temperature-induced gene expression in the regional trout strain BORN, in addition to a common imported strain. Moderate temperature challenge provoked typical stress response clusters, including heat-shock proteins or cold-inducible factors, in addition to altered energy Electronic supplementary material The online version of this article (doi:10.1007/s10126-015-9640-1) contains supplementary material, which is available to authorized users. * Tom Goldammer [email protected] 1
metabolism in trout kidney. Mild cold, in particular, enhanced renal protein degradation processes, as well as mRNA and protein synthesis, while it also triggered fatty acid biosynthesis. Mild heat led to cytoskeleton-stabilizing processes and might have facilitated cell damage and infection. Furthermore, both breeding lines used different strategies for energy provision, cellular defense, and cell death/survival pathways. As a main finding, the genes involved in energy provision showed generally higher transcript levels at both temperatures in BORN trout compared to imported trout, indicating adjusted metabolic rates under local environmental conditions. Altogether, this study provides a general overview of stress-induced transcriptional patterns in rainbow trout trunk kidney, in addition to identifying genes and networks that contribute to the robustness of the BORN strain. Our analyses suggest SERPINH1 and CIRBP as general marker genes for heat stress and cold stress in trout, respectively. Keywords Robustness . Rainbow trout . Temperature challenge . Kidney . Stress marker . Transcriptome analysis
Introduction Institute for Genome Biology, Leibniz Institute for Farm Animal Biology (FBN), Wilhelm-Stahl-Allee 2, 18196 Dummerstorf, Germany
2
Norwegian Institute of Food, Fisheries and Aquaculture Research (Nofima), Osloveien 1, NO-1430 Ås, Norway
3
Institute of Immunology, Friedrich-Loeffler-Institute (FLI), Federal Research Institute for Animal Health, Südufer 10, 17493 Greifswald, Insel Riems, Germany
4
Institute for Fishery, State Research Centre for Agriculture and Fishery (LFA-MV), Fischerweg 408, 18096 Rostock, Germany
5
Present address: Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce street, Philadelphia, PA, USA
Seasonal fluctuations of biotic and abiotic factors, such as changes in water temperature or quality, inevitably challenge farmed fish in semi-open aquaculture. Fish are poikilothermic animals, and changing temperatures may have fundamental effects on their physiology. Therefore, water temperature is the Babiotic master factor^ for fish (Brett 1971). Several studies proved that water temperatures deviating from their thermal optimum trigger stress-induced modifications of diverse cellular processes (Fry 1971; Bly and Clem 1992; Clarke and Fraser 2004). Local selection and breeding of adapted fish strains that are characterized by low stress susceptibility are
Mar Biotechnol
suitable possibilities that can be used to minimize the influence of stressful conditions on fish welfare. The German rainbow trout line BORN represents a robust breeding strain that is readily adapted to the challenges of the Baltic Sea’s coastal brackish water (Anders 1986; Rebl et al. 2014). Higher survival rates after pathogen-induced stress (Korytář et al. 2009), faster growth, and weight gain are economical key traits of this regional strain. Our observations document that fast rising water temperatures in spring, reaching unusually high values of 22 up to 27 °C, resulted in imported trout refusing food intake and demonstrating a reduced weight gain compared to the level of unaffected ingestion, while the BORN trout still fed and grew. The optimal growth temperature for rainbow trout ranges from 8 to 15°C, while temperatures above 18 °C impair feeding. Values above 25 °C are mostly lethal (Woynarovich et al. 2011). Regarding the different impact of high summer-like temperatures observed on both breeding strains, we assume a different character of thermoregulation, based on divergent transcription patterns. The kidney is the central organ of the primary stress response in higher vertebrates (Reid et al. 1998). The underlying endocrine patterns are evolutionarily conserved (Tort and Teles 2011). The renal steroidogenic cells of teleost fish release the corticosteroid cortisol through the hypothalamicpituitary-interrenal (HPI) axis into the blood circulation system. This leads to an adjusted stress response, metabolic regulation, and immune responses, as well as the expression of heat shock proteins (HSPs) (Pickering 1981; Barton 2002; Leatherland et al. 2010). The kidney of bony fish, along with the gills and intestine, is generally responsible for osmoregulation and excretion (Marshall and Grosell 2006; Engelund and Madsen 2011). Moreover, it is comprised of hematopoietic and lymphoid tissue, exerting immunological functions (Zapata et al. 2006). The latest research on teleost kidney mainly focuses on the clinical aspects of infections, such as the bacterial kidney disease (Schmidt-Posthaus et al. 2013), investigations on the expression of immune relevant genes in head kidney leukocytes (Villarroel et al. 2013) and the regulation of cortisol release (Conde-Sieira et al. 2013). Little is known about renal transcriptome profiles in fish, especially on the impact of environmental stressors, like temperature fluctuations, on functional gene regulation and stress adaptation. To the best of our knowledge, there has only been one report on rainbow trout kidney transcriptome up to now (Krasnov et al. 2005). This study focuses on general transcriptome profiles of rainbow trout trunk kidney after mild heat and cold stress (8 and 23 °C). Microarray analyses provide insights into the molecular mechanisms underlying the thermal stress response of fish. Our previous analysis identified thermal stress responses specific for trout gill, which is directly exposed to the environment (Rebl et al. 2013). In contrast to the gill, the fish kidney is part of the HPI axis
and therefore a crucial tissue directly involved in stress response. We intended to classify characteristic expression profiles in the trunk kidney as well as variations in the strainspecific transcription patterns of the robust regional rainbow trout strain BORN, in addition to an imported strain. The goal of the investigations was to identify kidney-specific, as well as general, markers and expression patterns for genetically determined thermal adaptation of the local breeding line.
Material and Methods Fish origin and Temperature Challenge Preliminary temperature experiments and subsequent transcriptome analyses used eggs of the local rainbow trout strain BORN (Born, Germany), in addition to an imported strain (Steelhead II-WA; Troutlodge, Tacoma, USA). The eggs were transferred into temperature-controlled, flow-through rearing tanks that provided identical rearing environments. The fish were simultaneously grown in freshwater conditions to 7–8month-old trout fingerlings, followed by an adaptation to 800L freshwater glass tanks. Subsequently, the 10-month-old rainbow trout of both strains were transferred into two separate 300-L freshwater tanks: ten animals per strain were mixed together in each tank, and the BORN trout were marked by a dorsal fin cut. The fish used in this study were identical to those used for Rebl et al.’s (2013) gill transcriptome analysis. At the start of the experiment, fish were 21±1 cm in length and 106±14 g in weight. Following an adaptation to 15 °C for 2 weeks, the water temperature was adjusted by 1 °C/day to the final temperatures of 8 and 23 °C. Both temperatures were maintained for another week. Finally, the tissues (trunk kidney, muscle, spleen, gills, and brain) were obtained from all fish of both strains and stored in RNAlater (25 mM Na3C6H5O7; 9.9 mM EDTA; 5.3 M (NH4)2SO4) at −80 °C until RNA extraction. This included tissues from ten control fish adapted to 15 °C to test the robustness of our predicted marker genes. RNA Preparation The tissue samples were homogenized in 1 ml TRIzol (Invitrogen, Karlsruhe, Germany), and the total RNA was extracted and purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany), following the manufacturer’s protocol. The isolated RNA was checked for (i) integrity using agarose gel electrophoresis and (ii) quality using the RNA integrity number (RIN) with the Agilent 2100 Bioanalyzer system. The Agilent 2100 used a Eukaryote total RNA Nano Series II assay (Agilent Technologies, Santa Clara, CA, USA). In addition, the sex of the animals used in this study was
Mar Biotechnol
determined using procedures that were described elsewhere (Rebl et al. 2013). cRNA Synthesis, Array Hybridization, and Data Analysis A comparative one-color microarray analysis of the renal transcriptome in BORN and imported trout, at 8 and 23 °C, was performed using 4×44K Salmon Gene Expression Microarray slides (Agilent Technologies) containing ∼43,800 salmonid 60-mer oligonucleotides. Agilent Service partner Atlas Biolabs (Berlin, Germany) conducted the study’s next steps. Probe preparation and slide hybridization were done following the MIAME guidelines (Wagner et al. 2007), using the protocol for a One-Color Microarray-Based Gene Expression Analysis, Low Input Quick Amp Labeling, v6.5 (http://genomics.agilent.com/). Therefore, pools (8 °C: each pool four individuals per strain; 23 °C: each pool three individuals per strain) of equal RNA amounts (100 ng) isolated from individuals of each treatment and rainbow trout line were transcribed into Cy3-labeled complementary RNA (cRNA) using a Quick Amp Labeling Kit, One Color (Agilent), followed by purification (Quiagen) and fragmentation, according to the manufacturer’s protocol. Finally, 15 ng/ μl of each fragmented cRNA pool (four pools in total) were hybridized to microarrays (65 °C, 10 rpm, 17 h). Each hybridization reaction was repeated as a biological replica (Online Resource 1). After washing in Agilent gene expression buffers 1 and 2 to remove any unbound dye, the slides were scanned using an Agilent Scanner (Model G2505C) and Agilent Scan Software (vA.8.3.1). This was followed by the primary data analysis using Agilent’s Feature Extraction Software (v10.7.3.1). The raw data were normalized using a quantile algorithm before subsequently being log2 transformed. Bad signals and signals below a predetermined threshold were excluded from the data: after the removal of outlier pixels, background correction, correction of spatial effects, etc. Primary tests, including an M versus A analysis, indicated a chip-to-chip comparison of the hybridizations and the variances of the ratios between the chips before and after normalization. Performing a boxplot analysis validated the data quality. Agilent Service partner Atlas Biolabs (Berlin) performed further statistical tests, including a Bonferroni correction of the p value to reduce false positives and a principal component analysis (PCA) of normalized data, as well as the determination of gene expression levels and calculation of fold change values. Annotation was performed using the STARS bioinformatics package and the annotation file BAgilent44kOligo_11_08_2009_FUSED.xls^ from the GRASP (Genomic Research on All Salmon Project) Web site (http://web.uvic.ca/grasp/microarray/). Only differently expressed genes with an absolute fold change of ≥3.0 and a corrected p value of
ORIGINAL ARTICLE
Impact of Thermal Stress on Kidney-Specific Gene Expression in Farmed Regional and Imported Rainbow Trout Marieke Verleih 1 & Andreas Borchel 1 & Aleksei Krasnov 2 & Alexander Rebl 1 & Tomáš Korytář 3,5 & Carsten Kühn 4 & Tom Goldammer 1
Received: 8 January 2015 / Accepted: 6 May 2015 # Springer Science+Business Media New York 2015
Abstract Seasonal water temperatures can be stressful for fish in aquaculture and can therefore negatively influence their welfare. Although the kidney is the crucial organ associated with the primary stress response, knowledge about the stressmodulated kidney transcriptome in salmonids is limited. In the present study, we used a comparative microarray approach to characterize the general gene expression profiles of rainbow trout trunk kidney after a 2-week acclimation to mild heat (23 °C) and cold stress (8 °C). Hypothesizing that local adaptation influences stress performance, we aimed to identify differences in the temperature-induced gene expression in the regional trout strain BORN, in addition to a common imported strain. Moderate temperature challenge provoked typical stress response clusters, including heat-shock proteins or cold-inducible factors, in addition to altered energy Electronic supplementary material The online version of this article (doi:10.1007/s10126-015-9640-1) contains supplementary material, which is available to authorized users. * Tom Goldammer [email protected] 1
metabolism in trout kidney. Mild cold, in particular, enhanced renal protein degradation processes, as well as mRNA and protein synthesis, while it also triggered fatty acid biosynthesis. Mild heat led to cytoskeleton-stabilizing processes and might have facilitated cell damage and infection. Furthermore, both breeding lines used different strategies for energy provision, cellular defense, and cell death/survival pathways. As a main finding, the genes involved in energy provision showed generally higher transcript levels at both temperatures in BORN trout compared to imported trout, indicating adjusted metabolic rates under local environmental conditions. Altogether, this study provides a general overview of stress-induced transcriptional patterns in rainbow trout trunk kidney, in addition to identifying genes and networks that contribute to the robustness of the BORN strain. Our analyses suggest SERPINH1 and CIRBP as general marker genes for heat stress and cold stress in trout, respectively. Keywords Robustness . Rainbow trout . Temperature challenge . Kidney . Stress marker . Transcriptome analysis
Introduction Institute for Genome Biology, Leibniz Institute for Farm Animal Biology (FBN), Wilhelm-Stahl-Allee 2, 18196 Dummerstorf, Germany
2
Norwegian Institute of Food, Fisheries and Aquaculture Research (Nofima), Osloveien 1, NO-1430 Ås, Norway
3
Institute of Immunology, Friedrich-Loeffler-Institute (FLI), Federal Research Institute for Animal Health, Südufer 10, 17493 Greifswald, Insel Riems, Germany
4
Institute for Fishery, State Research Centre for Agriculture and Fishery (LFA-MV), Fischerweg 408, 18096 Rostock, Germany
5
Present address: Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce street, Philadelphia, PA, USA
Seasonal fluctuations of biotic and abiotic factors, such as changes in water temperature or quality, inevitably challenge farmed fish in semi-open aquaculture. Fish are poikilothermic animals, and changing temperatures may have fundamental effects on their physiology. Therefore, water temperature is the Babiotic master factor^ for fish (Brett 1971). Several studies proved that water temperatures deviating from their thermal optimum trigger stress-induced modifications of diverse cellular processes (Fry 1971; Bly and Clem 1992; Clarke and Fraser 2004). Local selection and breeding of adapted fish strains that are characterized by low stress susceptibility are
Mar Biotechnol
suitable possibilities that can be used to minimize the influence of stressful conditions on fish welfare. The German rainbow trout line BORN represents a robust breeding strain that is readily adapted to the challenges of the Baltic Sea’s coastal brackish water (Anders 1986; Rebl et al. 2014). Higher survival rates after pathogen-induced stress (Korytář et al. 2009), faster growth, and weight gain are economical key traits of this regional strain. Our observations document that fast rising water temperatures in spring, reaching unusually high values of 22 up to 27 °C, resulted in imported trout refusing food intake and demonstrating a reduced weight gain compared to the level of unaffected ingestion, while the BORN trout still fed and grew. The optimal growth temperature for rainbow trout ranges from 8 to 15°C, while temperatures above 18 °C impair feeding. Values above 25 °C are mostly lethal (Woynarovich et al. 2011). Regarding the different impact of high summer-like temperatures observed on both breeding strains, we assume a different character of thermoregulation, based on divergent transcription patterns. The kidney is the central organ of the primary stress response in higher vertebrates (Reid et al. 1998). The underlying endocrine patterns are evolutionarily conserved (Tort and Teles 2011). The renal steroidogenic cells of teleost fish release the corticosteroid cortisol through the hypothalamicpituitary-interrenal (HPI) axis into the blood circulation system. This leads to an adjusted stress response, metabolic regulation, and immune responses, as well as the expression of heat shock proteins (HSPs) (Pickering 1981; Barton 2002; Leatherland et al. 2010). The kidney of bony fish, along with the gills and intestine, is generally responsible for osmoregulation and excretion (Marshall and Grosell 2006; Engelund and Madsen 2011). Moreover, it is comprised of hematopoietic and lymphoid tissue, exerting immunological functions (Zapata et al. 2006). The latest research on teleost kidney mainly focuses on the clinical aspects of infections, such as the bacterial kidney disease (Schmidt-Posthaus et al. 2013), investigations on the expression of immune relevant genes in head kidney leukocytes (Villarroel et al. 2013) and the regulation of cortisol release (Conde-Sieira et al. 2013). Little is known about renal transcriptome profiles in fish, especially on the impact of environmental stressors, like temperature fluctuations, on functional gene regulation and stress adaptation. To the best of our knowledge, there has only been one report on rainbow trout kidney transcriptome up to now (Krasnov et al. 2005). This study focuses on general transcriptome profiles of rainbow trout trunk kidney after mild heat and cold stress (8 and 23 °C). Microarray analyses provide insights into the molecular mechanisms underlying the thermal stress response of fish. Our previous analysis identified thermal stress responses specific for trout gill, which is directly exposed to the environment (Rebl et al. 2013). In contrast to the gill, the fish kidney is part of the HPI axis
and therefore a crucial tissue directly involved in stress response. We intended to classify characteristic expression profiles in the trunk kidney as well as variations in the strainspecific transcription patterns of the robust regional rainbow trout strain BORN, in addition to an imported strain. The goal of the investigations was to identify kidney-specific, as well as general, markers and expression patterns for genetically determined thermal adaptation of the local breeding line.
Material and Methods Fish origin and Temperature Challenge Preliminary temperature experiments and subsequent transcriptome analyses used eggs of the local rainbow trout strain BORN (Born, Germany), in addition to an imported strain (Steelhead II-WA; Troutlodge, Tacoma, USA). The eggs were transferred into temperature-controlled, flow-through rearing tanks that provided identical rearing environments. The fish were simultaneously grown in freshwater conditions to 7–8month-old trout fingerlings, followed by an adaptation to 800L freshwater glass tanks. Subsequently, the 10-month-old rainbow trout of both strains were transferred into two separate 300-L freshwater tanks: ten animals per strain were mixed together in each tank, and the BORN trout were marked by a dorsal fin cut. The fish used in this study were identical to those used for Rebl et al.’s (2013) gill transcriptome analysis. At the start of the experiment, fish were 21±1 cm in length and 106±14 g in weight. Following an adaptation to 15 °C for 2 weeks, the water temperature was adjusted by 1 °C/day to the final temperatures of 8 and 23 °C. Both temperatures were maintained for another week. Finally, the tissues (trunk kidney, muscle, spleen, gills, and brain) were obtained from all fish of both strains and stored in RNAlater (25 mM Na3C6H5O7; 9.9 mM EDTA; 5.3 M (NH4)2SO4) at −80 °C until RNA extraction. This included tissues from ten control fish adapted to 15 °C to test the robustness of our predicted marker genes. RNA Preparation The tissue samples were homogenized in 1 ml TRIzol (Invitrogen, Karlsruhe, Germany), and the total RNA was extracted and purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany), following the manufacturer’s protocol. The isolated RNA was checked for (i) integrity using agarose gel electrophoresis and (ii) quality using the RNA integrity number (RIN) with the Agilent 2100 Bioanalyzer system. The Agilent 2100 used a Eukaryote total RNA Nano Series II assay (Agilent Technologies, Santa Clara, CA, USA). In addition, the sex of the animals used in this study was
Mar Biotechnol
determined using procedures that were described elsewhere (Rebl et al. 2013). cRNA Synthesis, Array Hybridization, and Data Analysis A comparative one-color microarray analysis of the renal transcriptome in BORN and imported trout, at 8 and 23 °C, was performed using 4×44K Salmon Gene Expression Microarray slides (Agilent Technologies) containing ∼43,800 salmonid 60-mer oligonucleotides. Agilent Service partner Atlas Biolabs (Berlin, Germany) conducted the study’s next steps. Probe preparation and slide hybridization were done following the MIAME guidelines (Wagner et al. 2007), using the protocol for a One-Color Microarray-Based Gene Expression Analysis, Low Input Quick Amp Labeling, v6.5 (http://genomics.agilent.com/). Therefore, pools (8 °C: each pool four individuals per strain; 23 °C: each pool three individuals per strain) of equal RNA amounts (100 ng) isolated from individuals of each treatment and rainbow trout line were transcribed into Cy3-labeled complementary RNA (cRNA) using a Quick Amp Labeling Kit, One Color (Agilent), followed by purification (Quiagen) and fragmentation, according to the manufacturer’s protocol. Finally, 15 ng/ μl of each fragmented cRNA pool (four pools in total) were hybridized to microarrays (65 °C, 10 rpm, 17 h). Each hybridization reaction was repeated as a biological replica (Online Resource 1). After washing in Agilent gene expression buffers 1 and 2 to remove any unbound dye, the slides were scanned using an Agilent Scanner (Model G2505C) and Agilent Scan Software (vA.8.3.1). This was followed by the primary data analysis using Agilent’s Feature Extraction Software (v10.7.3.1). The raw data were normalized using a quantile algorithm before subsequently being log2 transformed. Bad signals and signals below a predetermined threshold were excluded from the data: after the removal of outlier pixels, background correction, correction of spatial effects, etc. Primary tests, including an M versus A analysis, indicated a chip-to-chip comparison of the hybridizations and the variances of the ratios between the chips before and after normalization. Performing a boxplot analysis validated the data quality. Agilent Service partner Atlas Biolabs (Berlin) performed further statistical tests, including a Bonferroni correction of the p value to reduce false positives and a principal component analysis (PCA) of normalized data, as well as the determination of gene expression levels and calculation of fold change values. Annotation was performed using the STARS bioinformatics package and the annotation file BAgilent44kOligo_11_08_2009_FUSED.xls^ from the GRASP (Genomic Research on All Salmon Project) Web site (http://web.uvic.ca/grasp/microarray/). Only differently expressed genes with an absolute fold change of ≥3.0 and a corrected p value of
