Environ. Sci. Technol. 2006, 40, 4031-4036
Plasma Proteome Analysis Reveals the Geographical Origin and Liver Tumor Status of Dab (Limanda limanda) from UK Marine Waters DOUGLAS G. WARD,† WENBIN WEI,† YAPING CHENG,† LUCINDA J. BILLINGHAM,† ASHLEY MARTIN,† PHILIP J. JOHNSON,† BRETT P. LYONS,‡ STEPHEN W. FEIST,‡ AND G R A N T D . S T E N T I F O R D * ,‡ Cancer Research UK Institute for Cancer Studies, The University of Birmingham, Edgbaston, Birmingham B15 2TT, U.K., and CEFAS Weymouth Laboratory, Barrack Road, Weymouth, Dorset DT4 8UB, U.K.
The flatfish species dab (Limanda limanda) is the sentinel for offshore marine monitoring in the United Kingdom National Marine Monitoring Programme (NMMP). At certain sites in the North and Irish Seas, the prevalence of macroscopic liver tumors can exceed 10%. The plasma proteome of these fish potentially contains reporter proteins or “biomarkers” that may enable development of diagnostic tests for liver cancer and further our understanding of the disease. Following selection of sample groups by qualityassured histopathology (“phenotype anchoring”), we used surface-enhanced laser desorption/ionization (SELDI) time-of-flight mass spectrometry to produce proteomic profiles of plasma from 213 dab collected during the 2004 UK NMMP. The resulting protein profiles were compared between fish from the North and Irish Seas and between fish with liver neoplasia or nondiseased liver. Significant differences were found between the plasma proteomes of dab from the North Sea and Irish Sea, which in conjunction with artificial neural networks can correctly determine from which sea dab were captured in 85% of the cases. In addition, the presence of liver tumors is associated with significant changes in the plasma proteome. We conclude that SELDI-based plasma profiling is potentially of use in nonlethal marine monitoring using wild sentinels such as dab. Furthermore, accurate selection of sample groups is critical for avoiding effects of confounding factors such as age, gender, and geographic origin of samples.
Introduction The presence of liver tumors in wild fish populations is assessed as a top-level indicator of exposure to anthropogenic contaminants in environmental monitoring programs (15). At some offshore sites in the North Sea, macroscopic liver tumor prevalence in wild flatfish exceeds 10% (6, 7) while * Corresponding author phone: +44 (0)1305 206722; fax: +44 (0)1305 206601; e-mail:
[email protected]. † The University of Birmingham. ‡ CEFAS Weymouth Laboratory. 10.1021/es052436q CCC: $33.50 Published on Web 05/12/2006
2006 American Chemical Society
prevalence in estuarine species can be even higher (4, 8). In addition to the assessment of grossly visible tumors, histopathological assessment of liver samples from flatfish populations collected under the UK National Marine Monitoring Programme (NMMP) allows for the detection of microscopic preneoplastic lesions. Currently, 30 categories of liver lesions are classified under the Biological Effects Quality Assurance in Monitoring (BEQUALM) program. These lesions range from nonspecific inflammatory pathologies through preneoplastic lesions to benign and malignant tumors. The diagnosis of these lesion types in the dab follows guidelines recently set out by Feist et al. (7). In addition to histopathological data, a range of external disease conditions are also recorded from these fish at point of capture. This thorough characterization meets criteria for quality assurance in sample grouping and allows for stringent “phenotypic anchoring” of samples to specific disease states in “-omic” studies (9). Surface-enhanced laser desorption/ionization (SELDI) time-of-flight mass spectrometry combines retentate chromatography and mass spectrometry in a high-throughput platform. A biofluid such as plasma is applied to a planar chromatographic surface (known as a “ProteinChip Array”); certain proteins will bind depending on their biophysical properties and salts and nonbinding proteins are removed by washing. The bound proteins are vaporized using a laser and analyzed by time-of-flight mass spectrometry. The peak intensities in the resulting mass spectra are related to the abundance of proteins and peptides in the plasma (10). To date SELDI has been used in only a few studies of aquatic organisms. These include an analysis of the proteome of the digestive gland and plasma of blue mussel from clean and polluted environments (11, 12), analysis of changes in the proteome of rainbow trout gills following exposure to zinc (13), and a study by ourselves on proteomic (SELDI) and metabolomic profile differences between tumorous and nontumorous liver tissue from the flatfish dab (Limanda limanda) (9). In all four studies SELDI detected proteomic changes resulting from toxicant exposure or disease status of the tissue being assayed. There is also evidence that SELDI can detect effects of environmental toxicants on the human serum proteome (14). We have used SELDI to analyze the plasma of 213 dab collected from 18 NMMP sites in the UK marine environment (see Figure 4, Supporting Information). These individuals were selected for analysis because of their clearly defined liver histology: no abnormality detected (NAD), foci of cellular alteration (FCA) or tumor (51 adenomas and 10 carcinomas). Information relating to the number of dab in each category, their age, and gender is summarized in Table 1. All samples were collected, prepared, stored, and analyzed in a quality-assured and identical manner as part of the UK NMMP. The plasma were assayed on Cu2+-loaded IMAC30 ProteinChip Arrays and the data analyzed for significant proteomic differences associated with geographical location and the presence of liver tumors.
Materials and Methods Plasma Collection, Preparation, and Selection. Dab (Limanda limanda) were captured at United Kingdom NMMP sites during June and July 2004 using 30-min tows of a standard Granton trawl. Fish were immediately placed into flowthrough tanks containing aerated seawater. The sex, length, weight, and presence of external signs of disease were recorded. Two milliliters of blood was drawn from the caudal vein into heparinized syringes within 2 h of capture. The VOL. 40, NO. 12, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4031
TABLE 1. Dab Plasma Samples Analyzed (One Sample per Fish)a no. of dab
male/female
mean age (SD)
Irish Sea North Sea
111 102
41/70 36/66
4.74 (1.64) 4.83 (1.69)
NAD FCA tumor
137 15 61
53/84 6/9 18/43
4.19 (1.32) 4.73 (1.22) 6.15 (1.67)
75 32 62 29
29/46 11/21 24/38 7/22
4.27 (1.38) 5.88 (1.74) 4.08 (1.25) 6.45 (1.56)
Irish Sea NAD Irish Sea tumor North Sea NAD North Sea tumor
a The table provides data on the number, gender, liver disease status, and sea of origin of the dab that provided the plasma samples used in this study. The mean (SD) age of each group is also shown.
blood was centrifuged for 5 min at 15000g and the plasma removed, immediately frozen in liquid nitrogen, and stored at -80 °C until analysis. To prevent the appearance of post mortem artifacts, only live fish were sampled. Following blood sampling, fish were sacrificed by a blow to the head followed immediately by severing of the spinal chord and a standard section through the liver was resected (7). Where visible liver tumors were present, a section was made through the tumor so as to contain tumor and surrounding nontumor tissue. Resected liver samples were fixed for 24 h in 10% neutral buffered formalin before transfer to 70% ethanol for subsequent histological confirmation of a range of histological lesion types classified under the Biological Effects Quality Assurance in Monitoring (BEQUALM) program (7). Lesions relevant to the current study include the preneoplastic foci of cellular alteration (FCA) and the neoplastic hepatocellular adenoma (HCA) and carcinoma (HCC). Fish showing none of the BEQUALM liver lesion categories were recorded as “No Abnormality Detected” (NAD). Otoliths were collected from each fish and age was analyzed following sectioning and staining using standard protocols. SELDI Analysis. Plasma samples were assayed using Cu2+loaded IMAC30 ProteinChip Arrays as previously described for human serum by Ward et al. (15). Briefly, plasma samples were diluted 5-fold with 9 M urea, 2% CHAPS, 50 mM Tris/ HCl, pH 9.0, followed by a 10-fold dilution in 500 mM NaCl, 100 mM NaH2PO4/NaOH, pH 7.0 (binding buffer). The 50fold diluted plasmas were loaded onto the ProteinChip Arrays (100 µL loaded per spot) using a 96-well bioprocessor and incubated at room temperature for 30 min. The ProteinChip Arrays were then washed with binding buffer, rinsed with water, and air-dried prior to addition of 2 × 1 µL of 50% saturated sinapinic acid in 50% acetonitrile, 0.5% trifluoroacetic acid. Spectra were collected over 0-20000 mass-tocharge ratio (m/z) and 0-200000 m/z ranges (372 laser shots) using laser intensities of 160 and 195, respectively, in a PBSIIc ProteinChip Reader (Ciphergen Biosystems Inc., Freemont, CA). Spectra were externally calibrated using all-in-one peptide standard (Ciphergen) and cytochrome C, chymotrypsinogen, bovine serum albumin, and phosphorylase b (Sigma-Aldrich). Spectra were normalized using the total ion current. Peaks were selected and clustered using Biomarker Wizard software (Ciphergen) with the signal-to-noise ratio >5 for the first pass and >2 for the second, a cluster mass window of 0.2%, and a requirement for peaks to be present in >5% of the spectra. All samples were assayed in duplicate and the peak intensities of the duplicate spectra averaged prior to further data analysis. All samples were assayed once in random order and then assayed again in random order. Data Analysis. Significant differences between the peak heights in the SELDI spectra of plasma from fish in various groups were determined by two-sample unpaired t-tests 4032
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 12, 2006
assuming unequal variance. Multiple regression analysis was performed using the lm function of R (http://www.rproject.org). For each proteomic feature a linear model was fitted where intensity was explained by tumor status, age (dichotomized using the median), gender, and Irish/North Sea and by interactions between tumor status and the other variables. Principal components analysis (PCA) and partial least-squares regression (PLS) were performed using PLS_Toolbox (Version 3.5, Eigenvector Research, Manson WA) running in Matlab (Version 7.1, The MathWorks, Natick, MA). Artificial neural networks (ANNs) were used to classify samples: data sets were split randomly and two-thirds of the data were used to train ANNs and one-third used to test the performance of the ANNs. The feed-forward neural networks consisted of an input layer, a hidden layer, and an output layer. The number of input nodes was determined by the number of proteomic features from which the ANNs were trained. The hidden layer controlled the complexity and performance of the neural networks. The output layer consisted of a single node whose output was used to classify sample status. The ANNs were trained using the backpropagation algorithm and assessed by 10-fold cross validation. During the training procedure, different models were built using varying numbers of the most significantly different proteomic features (as determined by t-tests on the training set). The best models were then applied to the test data sets.
Results Liver Histopathology. Liver pathology was classified according to the diagnostic criteria set out under the BEQUALM program (7). Grossly visible liver tumors were observed as masses of g2 mm in diameter (mean 7.8 ( 0.6 mm) on the upper or lower surfaces of the dissected liver (Figure 5, Supporting Information). Such lesions were generally shown to be well-vascularized. Tumors were also diagnosed from fish using microscopic assessment of histological samples. These tumors were not recorded in the field and were either 0.05). The mean intensities of these proteomic features for four groups of dab representing a progression in liver disease severity from NAD through FCA to HCA and HCC are plotted in Figure 2 and show a common behavior: FCA and NAD intensities are similar and tumor-associated changes are seen in HCA and HCC but are more pronounced in HCC than in HCA for 9 of the 10 features. PLS of the proteomic data with respect to tumor/NAD and with age-dependent features excluded is shown for the Irish and North Sea in Figures 3a and 3b, respectively. The different distributions of the tumor/NAD data points confirm the t-test and multiple regression results that liver tumors influence the plasma proteome. The proteomic features that differed most significantly between dab with NAD and dab with liver tumor and were independent of age were also used to train ANNs to classify dab as “tumor” or “NAD” in the North and Irish Seas. The best ANNs correctly predicted 7 out of 10 fish with liver tumor (70% sensitivity) and 17 out of 22 NAD (77% specificity) in the North Sea and 7 out of 11 tumors (64% sensitivity) and 18 out of 27 NAD (67% specificity) in the Irish Sea.
Discussion
FIGURE 3. PLS analysis of the plasma proteome profiles of dab with no abnormality detected or liver tumor in the Irish Sea (Figure 3a) and North Sea (Figure 3b) (solid triangles ) tumor; empty triangles ) NAD). Eleven age-associated proteomic features were withheld from this analysis. regression analysis to identify those proteomic features significantly associated with tumor status after adjusting for age, gender, and geographical location. This analysis showed that 27 proteomic features are significantly associated with tumor status (Table 4 in Supporting Information), of which 4034
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 12, 2006
The results presented here demonstrate that plasma proteome profiles obtained by SELDI contain information that reflects the geographic location (North Sea/Irish Sea) and liver tumor status in dab populations captured in UK waters. One explanation of the effect of geographic location is that dab from the North and Irish Seas may represent two genetically distinct populations that undergo little mixing, but this seems an unlikely explanation as fish collected from Burbo Bight and Morecombe Bay have plasma profiles resembling that of North Sea fish (Figure 7, Supporting Information). This assumption is further supported by evidence collated on the distribution of eggs and larvae across the species geographic range, which suggests appreciable gene flow supporting the maintenance of a single metapopulation (16). More likely, the plasma proteome reflects localized environmental conditions (such as feeding habits). Regardless of the underlying cause, the plasma proteome appears to contain information on where the dab were captured and therefore plasma proteome profiles may show significant potential for distinguishing commercial fish stocks, with the caveat that blood must be sampled from live fish within a few hours of capture.
Technologies that enable assessment of the genome, proteome, and metabolome of wild aquatic animals are increasingly being suggested as a means to generate biomarkers for the assessment of acute and chronic exposure to contaminants (9, 11-13). However, in this study, and in a previous study (9) we have demonstrated the importance of “phenotypically anchoring” sample groups to best available data for variables such as gender, age, season, geographic location, and disease status. This highlights the importance of employing an “integrated” approach to marine monitoring whereby all samples are collected in a quality-assured manner from defined numbers of the same specimens and that all life history data relating to these specimens are recorded at the individual level. Only when such confounding factors are included in downstream analyses of biomarker data can we expect to generate true markers of disease state or historical exposure. This general approach further highlights the critical nature of host population biology data when interpreting data relating to the biological effects of contaminants. A number of laboratories using SELDI have found that many human cancers are associated with changes in the serum proteome profile (15, 17-25). While these studies demonstrate great potential, many have used small numbers of samples, unrealistic ratios of cancer patients to noncancer controls, and control samples that may not have been collected and processed in a manner identical to the cancer samples. The current study represents an excellent model system to test whether liver tumors affect the plasma proteome in a manner that can be detected by SELDI profiling. Although the dab will have been stressed by the sampling procedure (26), all individuals were treated in an identical manner and the plasma was taken and stored before it was known whether the fish had liver tumors, preventing operator bias (and all downstream steps were randomized). In addition, to the best of our ability, the sample groups are all well-characterized with control (NAD) fish not showing any gross or histological liver tumor and tumor fish exhibiting a histologically confirmed tumor according to internationally agreed pathology criteria (7). Our ANNs suggest that it may be possible to identify dab with liver tumors from their plasma proteomic profile; however, larger studies are required to validate this finding and also to more rigorously exclude any confounding effects of age. Identification of the proteins that alter in the plasma proteome may further our understanding of the aetiology of liver tumors in wild fish: a driving force behind marine monitoring programs such as the UK NMMP. Further development of these markers to antibody-based diagnostics relating to specific cancer-related proteins may help to augment current approaches based upon histopathology and SELDI and may also allow for discrimination of specific tumor types that relate to exposure to specific contaminants or mixtures. As the histopathology of liver tumors in dab and humans show many similarities (Dr. Philip Taniere, personal communication) and fish have many orthologs of human oncogenes and tumor suppressor genes (27-29), one can hypothesize that any cancer biomarkers discovered in dab may also have human orthologs. Therefore, in addition to showing potential for marine monitoring and fish stock identification, plasma profile characterization in dab using SELDI analysis may eventually benefit human healthcare and help to assess the potential for a common environmental aetiology of certain tumor types across species.
Acknowledgments D.G.W. and W.W. contributed equally to this work. The authors acknowledge Defra (Contract AE003) and Cefas (Seedcorn Contracts DP180 and DP195) for support to G.D.S., S.W.F., and B.P.L. to undertake this work. We wish to thank
the crew of the RV Cefas Endeavour and colleagues from Cefas Burnham and Lowestoft laboratories for their assistance with sample collection. We also thank Mr. Michael Easey from Cefas for carrying out otolith sectioning and aging of dab used in this study.
Supporting Information Available Further details of the multiple regression analysis, collection sites, histopathology, SELDI spectra, and PCA analysis of the “average proteome” at each NMMP site. This information is free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Myers, M. S.; Rhodes, L. D.; McCain, B. B. Pathologic anatomy and patterns of occurrence of hepatic neoplasms, putative preneoplastic lesions, and other idiopathic hepatic conditions in English sole (Parophrys vetulus) from Puget Sound, Washington. J. Natl. Cancer Inst. 1987, 78, 333-363. (2) Reichert, W. L.; Myers, M. S.; Peck-Miller, K.; French, B.; Anulacion, B. F.; Collier, T. K.; Stein, J. E.; Varanasi, U. Molecular epizootiology of genotoxic events in marine fish: linking contaminant exposure, DNA damage, and tissue-level alterations. Mutat. Res. 1998, 411, 215-225. (3) Stein, J. E.; Reichert, W. L.; Nishimoto, M.; Varanasi, U. Overview of studies on liver carcinogenesis in English sole from Puget Sound; evidence for a xenobiotic chemical etiology. II: Biochemical studies. Sci. Total Environ. 1990, 1, 51-69. (4) Stentiford, G. D.; Longshaw, M.; Lyons, B. P.; Jones, G.; Green, M.; Feist, S. W. Histopathological biomarkers in estuarine fish species for the assessment of biological effects of contaminants. Mar. Environ. Res. 2003, 55, 137-159. (5) Vethaak, A. D.; Jol, J. G.; Meijboom, A.; Eggens, M. L.; Rheinallt, T.; Wester, P. W.; van de Zande, T.; Bergman, A.; Dankers, N.; Ariese, F.; Baan, R. A.; Everts, J. M.; Opperhuizen, A.; Marquenie, J. M. Skin and liver diseases induced in flounder (Platichthys flesus) after long-term exposure to contaminated sediments in large-scale mesocosms. Environ. Health Perspect. 1996, 104, 1218-1829. (6) Anon, UK national marine monitoring programme; Second report (1999-2001); Marine environment monitoring group, CEFAS, UK, 2004. (7) Feist, S. W.; Lang, T.; Stentiford, G. D.; Koehler, A. Use of liver pathology of the European flatfish dab (Limanda limanda) and flounder (Platichthys flesus) for monitoring. ICES Tech. Marine Environ. Sci. 2004, 38, 1-42. (8) Koehler, A. The gender specific risk to liver toxicity and cancer of flounder (Platichthys flesus) at the German Wadden sea coast. Aquat. Toxicol. 2004, 70, 257-276. (9) Stentiford, G. D.; Viant, M. R.; Ward, D. G.; Johnson, P. J.; Martin, A.; Wei, W.; Cooper, H. J.; Lyons, B. P.; Feist, S. W. Liver tumours in wild flatfish: a histopathological, proteomic, and metabolomic study. OMICS 2005, 9, 281-299. (10) Vorderwulbecke, S.; Cleverley, S.; Weinberger, S. R.; Weisner, A., Protein quantitation by the SELDI-TOF-MS-based proteinchip system. Nat. Methods 2005, 2, 393-395. (11) Bjornstad, A.; Larsen, B.; Skadsheim, A.; Jones, M. B.; Andersen, O. K., The potential of ecotoxicoproteomics in environmental monitoring: biomarker profiling in mussel plasma using proteinchip array technology. J. Toxicol. Environ. Health A 2006, 69, 77-96. (12) Knigge, T.; Monsinjon, T.; Andersen, O.-K. Surface enhanced laser desorption/ionition-time of flight-mass spectrometry approach to biomarker discovery in blue mussels (Mytilus edulis) exposed to polyaromatic hydrocarbons and heavy metals under field conditions. Proteomics 2004, 4, 2722-2727. (13) Hogstrand, C.; Balesaria, S.; Glover, C. N. Application of genomics and proteomics for study of the integrated response to zinc exposure in non-model fish species, the rainbow trout. Comp. Biochem. Physiol. 2002, 133, 523-535. (14) Vermeulen, R.; Lan, Q.; Zhang, L.; Gunn, L.; McCarthy, D.; Woodbury, R. L.; McGuire, M.; Podust, V. N.; Li, G.; Chatterjee, N.; Mu, R.; Yin, S.; Rothman, N.; Smith, M. T. Decreased levels of CXC-chemokines in serum of benzene-exposed workers identified by array-based proteomics. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17041-17046. (15) Ward, D. G.; Chen, Y. C.; N′Kontchou, G.; Thar, T. T.; Barget, N.; Wei, W.; Billingham, L. J.; Martin, A.; Beaugrand, M.; Johnson, VOL. 40, NO. 12, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4035
(16) (17) (18)
(19)
(20)
(21)
(22)
P. HCC induced changes in the serum proteome of hepatitis C infected chronic liver disease patients. Br. J. Cancer 2006, 94, 287-292. Henderson, P. On the variation in Limanda limanda recruitment: a zoogeographic study. J. Sea Res. 1998, 40, 131-142. Guang, Z.; Chun-Fang, G.; Guo-Ying, S.; Dong-Hui, L.; Xiu-Li, W. Identification of colorectal cancer using proteomic patters in serum. Chin. J. Cancer 2004, 23, 614-618. Kozak, K. R.; Amneua, M. W.; Pusey, S. M.; Su, F.; Luong, M. N.; Luong, S.; Reddy, S. T.; Farias-Eisner, R. Identification of biomarkers for ovarian cancer using strong anion-exchange proteinchips: potential use in diagnosis and prognosis. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 12343-12348. Li, J.; Rosenzweig, J. M.; Wang, Y. Y.; Chan, D. W. Proteomics and bioinformatics approaches for identification of serum biomarkers to detect breast cancer. Clin. Chem. 2002, 48, 12961304. Paradis, V.; Degos, F.; Dargere, D.; Pham, N.; Belghiti, J.; Degott, C.; Janeau, J.-L.; Bezeaud, A.; Delforge, D.; Cubizolles, M.; Laurendeau, I.; Bedossa, P. Identification of a new marker of hepatocellular carcinoma by serum protein profiling of patients with chronic liver disease. Hepatology 2005, 41, 40-47. Petricoin, E. F.; Ornstein, D. K.; Pawaleletz, C. P.; Ardekani, A.; Hackett, P.; Hitt, B. A.; Velassco, A.; Trucco, C.; Wiegand, L.; Wood, K.; Simone, C. B.; Levine, P. J.; Linehan, W. M.; EmmertBuck, M. R.; Steinberg, S. M.; Kohn, E. C.; Liotta, L. A. Serum proteomic patterns for detection of prostate cancer. J. Natl. Cancer Inst. 2002, 94, 1576-1578. Poon, T. C. W.; Yip, T.-T.; Chan, A. T. C.; Yip, C.; Yip, V.; Mok, T. S. K.; Lee, C. C. Y.; Leung, T. W. T.; Ho, S. K. W.; Johnson, P. J. Comprehensive proteomic profiling identifies serum proteomic signatures for detection of hepatocellular carcinoma and its subtypes. Clin. Chem. 2003, 49, 752-760.
4036
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 12, 2006
(23) Schwegler, E. E.; Cazares, L.; Steel, L. F.; Adam, B.-L.; Johnson, D. A.; Semmes, O. J.; Block, T. M.; Marrero, J. A.; Drake, R. R. SELDI-TOF MS profiling of serum for detection of the progression of chronic hepatitis C to hepatocellular carcinoma. Hepatology 2005, 41, 634-642. (24) Won, Y.; Song, H.-J.; Kang, T. W.; Kim, J.-J.; Han, B.-D.; Lee, S.-W. Pattern analysis of serum proteome distinguishes renal cell carcinoma from other urologic diseases and healthy persons. Proteomics 2003, 3, 2310-2316. (25) Yang, S.-Y.; Xiao, X.-Y.; Zhang, W.-G.; Zhang, L.-J.; Zhang, W.; Chen, G.; He, D.-C. Application of serum SELDI proteomic patterns in diagnosis of lung cancer. BMC Cancer 2005, 5. (26) Waring, C. P.; Stagg, R. M.; Poxton, M. G. The effects of handling on flounder (Platichthys flesus L.) and atlantic salmon (Salmo salar L.). J. Fish Biol. 1991, 41, 131-144. (27) Du Corbier, F. A.; Stentiford, G. D.; Lyons, B. P.; Rotchell, J. M. Isolation of the Retinoblastoma cDNA from the marine flatfish dab (Limanda limanda) and evidence of mutational alterations in liver tumors. Environ. Sci. Technol. 2005, in press. (28) Langheinrich, U.; Hennen, E.; Stott, G.; Vacun, G. Zebrafish as a model organism for the identification and characterisation of drugs and genes affecting p53 signalling. Curr. Biol. 2002, 12, 2023-2028. (29) Liu, T. X.; G, H. N.; Deng, M.; Langenau, D. M.; Hsu, K.; Rhodes, J.; Kanki, J. P.; D’Andrea, A. D.; Look, A. T. Knockdown of zebrafish Fancd2 causes developmental abnormalities via p53dependent apoptosis. Dev. Cell 2003, 5, 903-914.
Received for review December 5, 2005. Revised manuscript received April 3, 2006. Accepted April 5, 2006. ES052436Q
