- Email: [email protected]
Identification of transcriptional effects of ethynyl oestradiol in male plaice (Pleuronectes platessa) by suppression subtractive hybridisation and a nylon macroarray
MARINE ENVIRONMENTAL RESEARCH Marine Environmental Research 58 (2004) 559–563 www.elsevier.com/locate/marenvrev
Identification of transcriptional effects of ethynyl oestradiol in male plaice (Pleuronectes platessa) by suppression subtractive hybridisation and a nylon macroarray Margaret Brown a,*, Ian M. Davies b, Colin F. Moffat b, Craig Robinson b, John Redshaw c, John A. Craft a a
Biological and Biomedical Sciences, Glasgow Caledonian University, Cowcaddens Rd., Glasgow, Scotland, UK b Fisheries Research Services, Marine Laboratory, Aberdeen, Scotland, UK c Scottish Environment Protection Agency, Peel Park, East Kilbride, Scotland, UK
Abstract Suppression subtractive hybridisation (SSH) was used to generate cDNA libraries representing genes differentially expressed in response to ethynyl oestradiol (EE2) exposure in liver from male plaice (Pleuronectes platessa) previously analysed for vitellogenin (VTG) induction. Characterisation of the cDNA clones identified many as VTG (2 genes) and zona radiata proteins (ZRP) (3 genes), but 40 encoded other proteins, with more than half cryptic. Further analysis identified 85 non-redundant clones suitable for array on nylon membrane. Radiolabelled cDNAs were prepared from hepatic mRNA from EE2 treated plaice (0 and 21 days) and hybridised with the arrayed clones. Analysis of the data showed that 11/17 novel, 21/22 VTG, 13/14 ZRP, 2/2 liver aspartic proteinase (LAP) and 8/10 other mRNAs were up-regulated by EE2 exposure. Crown copyright Ó 2004 Published by Elsevier Ltd. All rights reserved. Keywords: Pleuronectes platessa; Suppression subtractive hybridisation; Ethynyl oestradiol; Endocrine disruption; Transcription profiling
*
Corresponding author. Tel.: +44-141-331-3222; fax: +44-141-331-3208. E-mail address: [email protected] (M. Brown).
0141-1136/$ - see front matter. Crown copyright Ó 2004 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.marenvres.2004.03.045
560
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
1. Introduction Array technologies offer great opportunities for discovery of the mechanistic basis of toxic responses in biota (Neumann & Galvez, 2002). However, they are currently unavailable to many laboratories because relevant genomic, or economic, resources are limited. As an alternative strategy, suppressive subtractive hybridisation (SSH) (described by Diatchenko et al., 1996) can generate cDNA libraries greatly enriched for genes differentially expressed in response to pollutants. This methodology is independent of sequence information and greatly reduces the complexity of clones an array must contain to be informative for pollutant-specific applications. In this context, cDNAs can be arrayed on nylon membrane as macroarrays (Vrana, Freeman, & Aschner, 2002), as opposed to glass slide microarrays, and can be manually loaded to eliminate requirement for automated printing. The aim of this study was to assess the potential of such an approach to develop an economical method for isolation of pollutant- and species-specific biomarker clones useful for environmental applications. Liver from male plaice, untreated or treated with ethynyl oestradiol (EE2) (20 ng/l flow through for 21 days) was used. Isolation of mRNA was with QuickPrep Micro mRNA Purification Kit (Amersham-Pharmacia Biotech, UK) and cDNA synthesis with SMARTTM PCR cDNA Synthesis Kit (BD Biosciences UK). Subtractions (forward and reverse) were carried out with a PCR-Select cDNA Subtraction Kit (BD Biosciences UK) and the resulting cDNA libraries ligated into a pT7 Blue 3 vector for cloning, using a Perfectly Blunt Cloning Kit (Novagen, CN Biosciences (UK) Ltd). PCR screening of colonies for recombinants was done with insert-specific primers in parallel with generation of archived, 96 well-format, bacterial stock cultures. Colonies containing inserts were identified by non-denaturing polyacrylamide gel electrophoresis (PAGE) of the PCR products using a MadgeBio Gel Kit. Following the method described in the PCR-Select Differential Screening Kit (BD Biosciences UK) protocol, cDNAs were assessed for confirmation of differential expression. Selected clones were sequenced and identity established where possible by BLAST searches. Non-redundant clones were selected for construction of a macroarray and the probes were generated by PCR amplification of the inserts directly from bacterial stocks and spotted on to replicate nylon membranes (see Fig. 1 for details). For analysis of altered gene expression mRNA was isolated from the livers of individual fish and then pooled ðn ¼ 5Þ. First-strand cDNA was synthesised (2.5 lg pooled mRNAs as starting material) using SuperscriptTM III Reverse Transcriptase (Invitrogen, Inchinnan, Renfrew). This was with oligo(dT)20 and random hexamer primers and addition of 0.4 lCi/ll [a-32 P]dCTP to generate radiolabelled single-stranded cDNA. Following the protocol described for non-formamide conditions of high temperature (68 °C) with salmon sperm ssDNA (Invitrogen, Inchinnan, UK), the macroarrays were incubated (3 h) in prehybridisation buffer with addition of a blocking primer mix containing DNA complementary to the adaptor sequences (MWG Biotech) (2.5 pmol/ml). The target cDNAs were then hybridised overnight with the macroarrays in hybridisation buffer/blocking primer mix under these same stringent conditions of high temperature (68 °C).
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
561
Fig. 1. Macroarray analysis. Insert specific primers were used to PCR amplify selected clones. The reaction products were denatured by addition of equal volume 0.6 M NaOH (Sigma). Each reaction product (1.5 ll) was spotted directly onto replicate nylon membranes (HyBond+, Amersham Biosciences, UK) neutralised in 0.5 M Tris–HCl (Sigma), air-dried, UV cross-linked (700 kJ) and stored at 4 °C. Hybridisation of the arrays was with [a-32 P]dCTP radiolabelled target cDNA populations, reverse transcribed from hepatic mRNA (2.5 lg) of pooled samples ðn ¼ 5Þ from EE2-treated male plaice for each time point. Results were analysed by autoradiography and the images digitised. Intensity of each spot was determined and normalised to the reference plaice genomic DNA (gDNA, highlighted in image) to allow comparison between filters.
Three low stringency (2 SSPE, 0.1% SDS) (20 min, 68 °C) and two high stringency (0.2 SSPE, 0.1% SDS) (10 min, 68 °C) washes were done and the results visualised by autoradiography. Semi-quantitative analysis of relative gene expression values was by ScanAnalysis (BioSoft, Cambridge, UK) of the images. Two subtracted libraries were constructed using either the cDNA from untreated plaice as driver (forward subtraction) or cDNA from EE2-treated fish as driver (reverse subtracted). More than 200 clones were identified with differential
562
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
expression of which 140 were sequenced and these were classified into groups. Clones representing up-regulated genes included (in parenthesis numbers of clones in each group): ZRP (35), VTG (26), cryptic, novel gene products (22), DNA sequence-like those in zebrafish (Danio rerio) but of unknown function (10), liver aspartic proteinases (LAP) (3), ubiquitin-like proteins and UDP-glucuronosyltransferases (UDP-GT). More extensive analysis selected clones for array on basis of showing no overlap or further redundancy and 63 forward- (up-) and 13 reverse-subtracted (down-regulated) were arrayed along with 4 reference genes and plaice genomic DNA (gDNA) as a standard. GENE vitellogenin ‘A’
vitellogenin ‘B’
zona radiata protein ‘A’
zona radiata protein ‘B’ zona radiata protein ‘C’ novel (forward-subtracted)
D. rerio DNA homology liver aspartic proteinase ubiquitin-like
UDP-GT novel (reverse-subtracted)
reference plaice gDNA
CLONE REFERENCE s2Ce2, s2Ed3, s2Dc2 S2Ed2, s2d4, s2Ab1, s2Cc3, s2Cc4, s2Cd2, s2c7, s2Ca2 s2Dc5 s2d2 s2h3 s2Ah5, s2Aa3 s2Ec10 s2f11, s2g10, s2Cb11, s2Ag4 s2Aa4, s2Aa10, s2Ca12 s2Cd4 s2h6, s2Ec4, s2Eb6 s2Cc8 s2Ag11, s2h7, s2Cd7 s2Ea4, s2Db12, s2Cg3 s2Ac9, s2c11, s2Ab3, s2d5, s2Ae12, s2Ah6 s2Ag9, s2f12, s2Ad4 s2Ah11 s2Ca7, s2Ac7 s2Ag3, s2Cb4, s2c8, s2Eb11, s2e3 s2Ec7 s2g4, s2Ag5 s2Ag8 s2Aa12 s1Bd11 s2Ae11 s2c10 s2Ab3, s2Dc1 s2Ac2 s1Bh5, s1Bh7, s1b6, s1d12 s1Bc4 s1a8, s1Bg5 s1a7, s1b2, s1Bh9, s1Bg9, s1Bf9 0.3 µg, 0.6 µg, 1.2 µg
EXPRESSION + ++ +++ ++++ No detectable signal ++ +++ ++++ + ++ +++ ++++ +++ ++++ No detectable signal + ++ +++ ++++ ++ ++++ ++ ++++ ++ ++++ No detectable signal ++++ No detectable signal + = =
Fig. 2. Relative gene expression Semi-quantitation of altered gene expression in liver of EE2-exposed male plaice by ScanAnalysis (BioSoft, Cambridge, UK) of autoradiograph.
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
563
Nylon macroarrays were used to assess altered gene expression in the EE2-treated male plaice and the autoradiograph of zero-day and 21-day exposed fish are shown in Fig. 1. These were scanned and the semi-quantitative data are shown in Fig. 2. The majority of the VTG and ZRP clones were up-regulated with the exception of a single clone for each group. The LAPs, zebrafish DNA-like, two of the ubiquitin-like and 11/17 putative up-regulated novel clones also showed increased expression levels of varying degrees. In contrast the reverse subtracted clones gave less clear results, generally reductions in signal intensity were marginal and not conclusive. The application of SSH generated cDNA libraries containing numerous ESTs representative of individual genes whose expression was altered in response to EE2 exposure. The many VTG clones appear to represent at least two genes while the many ZRP clones are derived from at least three genes. The number of VTG and ZRP genes in plaice has not been established but on-going analysis of the data and experimentation should provide a clearer picture. Investigation of gene expression profiles with hepatic mRNA from untreated and EE2-treated male plaice with the arrayed ESTs confirmed the potential of the macroarray format for future studies in the field.
Acknowledgements Funded by Defra, UK and SEPA, UK.
References Diatchenko, L., Lau, C. Y.-F., Campbell, A. P., Chenchik, A., Moqadam, F., Huang, B., Lukyanov, S., Lukyanov, K., Gurskaya, N., Sverdlov, E. D., & Siebert, P. D. (1996). Proceedings of the National Academic Sciences of the United States of America, 93, 6025–6030. Neumann, N. F., & Galvez, G. (2002). Biotechnology Advances, 20, 391–419. Vrana, K. E., Freeman, W. M., & Aschner, M. (2002). Neurotoxicology, 186, 1–12.
Identification of transcriptional effects of ethynyl oestradiol in male plaice (Pleuronectes platessa) by suppression subtractive hybridisation and a nylon macroarray Margaret Brown a,*, Ian M. Davies b, Colin F. Moffat b, Craig Robinson b, John Redshaw c, John A. Craft a a
Biological and Biomedical Sciences, Glasgow Caledonian University, Cowcaddens Rd., Glasgow, Scotland, UK b Fisheries Research Services, Marine Laboratory, Aberdeen, Scotland, UK c Scottish Environment Protection Agency, Peel Park, East Kilbride, Scotland, UK
Abstract Suppression subtractive hybridisation (SSH) was used to generate cDNA libraries representing genes differentially expressed in response to ethynyl oestradiol (EE2) exposure in liver from male plaice (Pleuronectes platessa) previously analysed for vitellogenin (VTG) induction. Characterisation of the cDNA clones identified many as VTG (2 genes) and zona radiata proteins (ZRP) (3 genes), but 40 encoded other proteins, with more than half cryptic. Further analysis identified 85 non-redundant clones suitable for array on nylon membrane. Radiolabelled cDNAs were prepared from hepatic mRNA from EE2 treated plaice (0 and 21 days) and hybridised with the arrayed clones. Analysis of the data showed that 11/17 novel, 21/22 VTG, 13/14 ZRP, 2/2 liver aspartic proteinase (LAP) and 8/10 other mRNAs were up-regulated by EE2 exposure. Crown copyright Ó 2004 Published by Elsevier Ltd. All rights reserved. Keywords: Pleuronectes platessa; Suppression subtractive hybridisation; Ethynyl oestradiol; Endocrine disruption; Transcription profiling
*
Corresponding author. Tel.: +44-141-331-3222; fax: +44-141-331-3208. E-mail address: [email protected] (M. Brown).
0141-1136/$ - see front matter. Crown copyright Ó 2004 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.marenvres.2004.03.045
560
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
1. Introduction Array technologies offer great opportunities for discovery of the mechanistic basis of toxic responses in biota (Neumann & Galvez, 2002). However, they are currently unavailable to many laboratories because relevant genomic, or economic, resources are limited. As an alternative strategy, suppressive subtractive hybridisation (SSH) (described by Diatchenko et al., 1996) can generate cDNA libraries greatly enriched for genes differentially expressed in response to pollutants. This methodology is independent of sequence information and greatly reduces the complexity of clones an array must contain to be informative for pollutant-specific applications. In this context, cDNAs can be arrayed on nylon membrane as macroarrays (Vrana, Freeman, & Aschner, 2002), as opposed to glass slide microarrays, and can be manually loaded to eliminate requirement for automated printing. The aim of this study was to assess the potential of such an approach to develop an economical method for isolation of pollutant- and species-specific biomarker clones useful for environmental applications. Liver from male plaice, untreated or treated with ethynyl oestradiol (EE2) (20 ng/l flow through for 21 days) was used. Isolation of mRNA was with QuickPrep Micro mRNA Purification Kit (Amersham-Pharmacia Biotech, UK) and cDNA synthesis with SMARTTM PCR cDNA Synthesis Kit (BD Biosciences UK). Subtractions (forward and reverse) were carried out with a PCR-Select cDNA Subtraction Kit (BD Biosciences UK) and the resulting cDNA libraries ligated into a pT7 Blue 3 vector for cloning, using a Perfectly Blunt Cloning Kit (Novagen, CN Biosciences (UK) Ltd). PCR screening of colonies for recombinants was done with insert-specific primers in parallel with generation of archived, 96 well-format, bacterial stock cultures. Colonies containing inserts were identified by non-denaturing polyacrylamide gel electrophoresis (PAGE) of the PCR products using a MadgeBio Gel Kit. Following the method described in the PCR-Select Differential Screening Kit (BD Biosciences UK) protocol, cDNAs were assessed for confirmation of differential expression. Selected clones were sequenced and identity established where possible by BLAST searches. Non-redundant clones were selected for construction of a macroarray and the probes were generated by PCR amplification of the inserts directly from bacterial stocks and spotted on to replicate nylon membranes (see Fig. 1 for details). For analysis of altered gene expression mRNA was isolated from the livers of individual fish and then pooled ðn ¼ 5Þ. First-strand cDNA was synthesised (2.5 lg pooled mRNAs as starting material) using SuperscriptTM III Reverse Transcriptase (Invitrogen, Inchinnan, Renfrew). This was with oligo(dT)20 and random hexamer primers and addition of 0.4 lCi/ll [a-32 P]dCTP to generate radiolabelled single-stranded cDNA. Following the protocol described for non-formamide conditions of high temperature (68 °C) with salmon sperm ssDNA (Invitrogen, Inchinnan, UK), the macroarrays were incubated (3 h) in prehybridisation buffer with addition of a blocking primer mix containing DNA complementary to the adaptor sequences (MWG Biotech) (2.5 pmol/ml). The target cDNAs were then hybridised overnight with the macroarrays in hybridisation buffer/blocking primer mix under these same stringent conditions of high temperature (68 °C).
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
561
Fig. 1. Macroarray analysis. Insert specific primers were used to PCR amplify selected clones. The reaction products were denatured by addition of equal volume 0.6 M NaOH (Sigma). Each reaction product (1.5 ll) was spotted directly onto replicate nylon membranes (HyBond+, Amersham Biosciences, UK) neutralised in 0.5 M Tris–HCl (Sigma), air-dried, UV cross-linked (700 kJ) and stored at 4 °C. Hybridisation of the arrays was with [a-32 P]dCTP radiolabelled target cDNA populations, reverse transcribed from hepatic mRNA (2.5 lg) of pooled samples ðn ¼ 5Þ from EE2-treated male plaice for each time point. Results were analysed by autoradiography and the images digitised. Intensity of each spot was determined and normalised to the reference plaice genomic DNA (gDNA, highlighted in image) to allow comparison between filters.
Three low stringency (2 SSPE, 0.1% SDS) (20 min, 68 °C) and two high stringency (0.2 SSPE, 0.1% SDS) (10 min, 68 °C) washes were done and the results visualised by autoradiography. Semi-quantitative analysis of relative gene expression values was by ScanAnalysis (BioSoft, Cambridge, UK) of the images. Two subtracted libraries were constructed using either the cDNA from untreated plaice as driver (forward subtraction) or cDNA from EE2-treated fish as driver (reverse subtracted). More than 200 clones were identified with differential
562
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
expression of which 140 were sequenced and these were classified into groups. Clones representing up-regulated genes included (in parenthesis numbers of clones in each group): ZRP (35), VTG (26), cryptic, novel gene products (22), DNA sequence-like those in zebrafish (Danio rerio) but of unknown function (10), liver aspartic proteinases (LAP) (3), ubiquitin-like proteins and UDP-glucuronosyltransferases (UDP-GT). More extensive analysis selected clones for array on basis of showing no overlap or further redundancy and 63 forward- (up-) and 13 reverse-subtracted (down-regulated) were arrayed along with 4 reference genes and plaice genomic DNA (gDNA) as a standard. GENE vitellogenin ‘A’
vitellogenin ‘B’
zona radiata protein ‘A’
zona radiata protein ‘B’ zona radiata protein ‘C’ novel (forward-subtracted)
D. rerio DNA homology liver aspartic proteinase ubiquitin-like
UDP-GT novel (reverse-subtracted)
reference plaice gDNA
CLONE REFERENCE s2Ce2, s2Ed3, s2Dc2 S2Ed2, s2d4, s2Ab1, s2Cc3, s2Cc4, s2Cd2, s2c7, s2Ca2 s2Dc5 s2d2 s2h3 s2Ah5, s2Aa3 s2Ec10 s2f11, s2g10, s2Cb11, s2Ag4 s2Aa4, s2Aa10, s2Ca12 s2Cd4 s2h6, s2Ec4, s2Eb6 s2Cc8 s2Ag11, s2h7, s2Cd7 s2Ea4, s2Db12, s2Cg3 s2Ac9, s2c11, s2Ab3, s2d5, s2Ae12, s2Ah6 s2Ag9, s2f12, s2Ad4 s2Ah11 s2Ca7, s2Ac7 s2Ag3, s2Cb4, s2c8, s2Eb11, s2e3 s2Ec7 s2g4, s2Ag5 s2Ag8 s2Aa12 s1Bd11 s2Ae11 s2c10 s2Ab3, s2Dc1 s2Ac2 s1Bh5, s1Bh7, s1b6, s1d12 s1Bc4 s1a8, s1Bg5 s1a7, s1b2, s1Bh9, s1Bg9, s1Bf9 0.3 µg, 0.6 µg, 1.2 µg
EXPRESSION + ++ +++ ++++ No detectable signal ++ +++ ++++ + ++ +++ ++++ +++ ++++ No detectable signal + ++ +++ ++++ ++ ++++ ++ ++++ ++ ++++ No detectable signal ++++ No detectable signal + = =
Fig. 2. Relative gene expression Semi-quantitation of altered gene expression in liver of EE2-exposed male plaice by ScanAnalysis (BioSoft, Cambridge, UK) of autoradiograph.
M. Brown et al. / Marine Environmental Research 58 (2004) 559–563
563
Nylon macroarrays were used to assess altered gene expression in the EE2-treated male plaice and the autoradiograph of zero-day and 21-day exposed fish are shown in Fig. 1. These were scanned and the semi-quantitative data are shown in Fig. 2. The majority of the VTG and ZRP clones were up-regulated with the exception of a single clone for each group. The LAPs, zebrafish DNA-like, two of the ubiquitin-like and 11/17 putative up-regulated novel clones also showed increased expression levels of varying degrees. In contrast the reverse subtracted clones gave less clear results, generally reductions in signal intensity were marginal and not conclusive. The application of SSH generated cDNA libraries containing numerous ESTs representative of individual genes whose expression was altered in response to EE2 exposure. The many VTG clones appear to represent at least two genes while the many ZRP clones are derived from at least three genes. The number of VTG and ZRP genes in plaice has not been established but on-going analysis of the data and experimentation should provide a clearer picture. Investigation of gene expression profiles with hepatic mRNA from untreated and EE2-treated male plaice with the arrayed ESTs confirmed the potential of the macroarray format for future studies in the field.
Acknowledgements Funded by Defra, UK and SEPA, UK.
References Diatchenko, L., Lau, C. Y.-F., Campbell, A. P., Chenchik, A., Moqadam, F., Huang, B., Lukyanov, S., Lukyanov, K., Gurskaya, N., Sverdlov, E. D., & Siebert, P. D. (1996). Proceedings of the National Academic Sciences of the United States of America, 93, 6025–6030. Neumann, N. F., & Galvez, G. (2002). Biotechnology Advances, 20, 391–419. Vrana, K. E., Freeman, W. M., & Aschner, M. (2002). Neurotoxicology, 186, 1–12.