Final Report Summary - FISHPOPTRACE (Fish population structure and traceability)
Executive summary:
Marine fish species, in common with all living organisms, are fragmented to varying degrees, into a series of locally interbreeding populations. The extent to which such populations differ biologically ('population diversity') and their distribution in time and space are among the most important drivers of species' survival and persistence in the face of environmental change. Fisheries biologists and managers have emphasised such thinking since the early 20th century, though there typically remains a mismatch between biological populations and fisheries management units. Thus, it remains difficult to devise strategies to relate population diversity to variable harvesting regimes and even more difficult to conserve overexploited stocks. Further uncertainty in striving for sustainability is the high level of illegal, unregulated and unreported fishing (IUU), estimated globally to cost the industry EUR?10-20 billion and prior to 2010, EUR?1.1 billion worth of illegal fish was imported into the European Union (EU) annually. FISHPOPTRACE, a 15 partner EU Seventh Framework Programme (FP7) project, aimed to address such challenges to sustainability within the context of the Common Fisheries Policy (CFP) to:
1. develop a range of cost-effective and reliable tools for identifying, monitoring and tracing marine fish populations in four representative European species (cod, herring, hake and common sole);
2. promote fisheries governance by ensuring that the most effective tools can be applied to forensic standards and thereby be legally supportive for prosecution and enforcement;
3. foster technology transfer of outputs in relation to enforcement and conservation policies of the EU CFP and associated socioeconomic consequences.
We have applied our primary traceability tools: deoxyribonucleic acid (DNA) single-nucleotide polymorphism (SNP) analysis (single genetic variants) and otolith ('calcareous ear bones') microchemistry and morphometrics (form, shape) to test their power for tracing fish from the four species back to population/area of origin. Such information provides a framework for a pan-European traceability database for fish and fish products to be applied for defining management areas, fighting IUU and for assuring safe, healthy and sustainable fish products for the European consumer. Our focus has been on the traceability of populations, but each test can also incorporate a species identity marker. We can now correctly assign fish to populations from more areas and with higher certainty than previously possible, reaching standards which can be used in a court of law. Based on use of the most highly distinct genes among populations we have already developed 'minimum assays with maximum power' with from 10 to 30 SNPs. These assays have been developed to target some of the most pertinent needs for traceability tools in European fisheries management. For example, we now have fast, efficient and forensically robust tools to discriminate between cod from Canada, North Sea, Baltic Sea and Northeast Arctic populations, between North Sea and North Atlantic herring, between sole from the Irish Sea and Thames and between hake from the Mediterranean and the Atlantic. The use of a marker system such as SNPs, which is essentially based on the presence or absence of large numbers of single genetic variants means that data can be compiled from sources in a much more reliable and high throughput way. The approach thereby enables the generation of baseline and ongoing additions for subsequent genetic monitoring. Moreover it is imperative that any such tools can be used in a legal context, necessitating forensic validation. This has been achieved for SNP markers within FISHPOPTRACE across a range of policy-driven IUU scenarios. FISHPOPTRACE outputs will thus contribute to efficient fishing activities within an economically viable and competitive fisheries industry and hence contribute to the CFP's aim of providing a fair standard of living for those who depend on fishing activities as well as taking into account the interests of consumers.
The aim and objectives of FISHPOPTRACE
From the establishment of FISHPOPTRACE, we have highlighted the central role that population diversity plays in sustainable utilisation and conservation of exploited stocks. FISHPOPTRACE was aimed at improving our ability to trace fish and fish products through enhanced understanding of the dynamics, temporal stability and distribution of major populations of four key exploited fish species: European hake, herring, sole and cod. Data and traceability tools have been integrated in a forensic framework to enable traceability within the context of enforcement and conservation policy. This in turn will facilitate implementation of effort limitation and regional allocation of quotas based on the relative abundance of respective populations or stocks, not only in the wild, but also through enforcement based on fish products. Traceability tools can also play a prominent role in consumer protection both at the species and population level, especially in the face of frequent mislabelling, fraud and to sustain 'ecolabelling' schemes of stocks certified in various ways.
The FISHPOPTRACE consortium consists of 15 partners with expertise in fish biology, population and conservation genetics, molecular biology, biochemistry and wildlife forensics. Partners were drawn from a range of representative bodies including the food industry, the European Commission (EC), universities, government laboratories and small and medium sized enterprises (SMEs). Collectively, the objectives encompass to:
1. integrate recent and on-going data from European fish species traceability projects and generate a single compatible database and tissue archive managed by the joint research centre (JRC) of the EC;
2. develop and apply a new type of genetic marker, applied for the first time to traceability and population structure of marine fishes: SNPs, together with tools developed from otolith microchemistry and shape variation in widely distributed populations of cod, hake, sole and herring, to discriminate biologically differentiated populations as a basis for traceability;
3. undertake validation of traceability tools to maximise robustness, transferability and reliance for subsequent use by end-users;
4. develop a population monitoring system based on otolith and genetic data that will assess the extent to which population traceability might change over time from various locations;
5. test the utility of additional novel traceability systems as potential independent tools in traceability;
6. foster technology transfer of FISHPOPTRACE outputs in relation to enforcement and conservation policies of the EU CFP and associated socioeconomic consequences.
Project results:
Before considering the outputs from FISHPOPTRACE in more detail, we first highlight three fundamental aspects of the design of the project: the choice of fish species as targets for testing; second, the rationale for choice of main traceability tools developed and finally, the role that fisheries forensics, the application of science to legal requirements in the fishery management context, has played in helping the consortium to address its objectives.
Why the choice of FISHPOPTRACE target species?
FISHPOPTRACE employed three primary criteria in the choice of target species, relating to conservation status, traceability issues and representation of marine fish life-styles: Atlantic cod (Gadus morhua L.), Atlantic herring (Clupea harengus L.), European hake (Merluccius merluccius L.) and common sole (Solea solea L.). The selected species are all economically important, relatively widespread on a European scale, known to exhibit population structure and fall within EC priority species for enforcement and/or conservation. All target species are vulnerable to a varying degree to overfishing. Finally, the four species have been selected to represent different life-styles ranging from a small pelagic (herring), coastal flatfish (sole), to benthopelagic (cod) and demersal 'deep sea' (hake) species. Such a range in ecology is associated with variability in the extent and pattern of population structuring, thereby providing diverse material as proof-of-concept for other marine fish species. The FISHPOPTRACE target species have also been chosen on the basis of their relative distribution, where cod and herring have a more northerly distribution than sole and hake. For the four species different levels of population genetic information are available, where cod has been studied genetically for decades, while relatively little is known about the population structure of hake. The multispecies approach encompasses different geographical scales for tracing individuals, representing a range of policy-led traceability scenarios.
Choice of primary traceability tools: genetics and otoliths
Several tools are available to understand the extent to which fish populations interbreed and to trace back the geographic origin of landed fish. These include physical external tags, natural tags and genetic markers linked to DNA, the genetic material that makes up genes. However, once a fish enters the food supply chain, several tools become less suitable. Tools for monitoring natural populations and application to fisheries enforcement should therefore meet stringent criteria: they should mirror population identity and stability over an ecological and evolutionary scale. Traceability tools should be available throughout the food supply chain from capture to a customer's plate and should be amenable to forensic validation for use in a court of law. Currently available genetic tools for traceability are of limited utility in an international and forensic context. An emerging class of genetic markers are SNPs. They are very abundant and widespread. Analyses of SNPs reach hitherto unprecedented levels of population identification, rendering them optimal tools in fundamental biology, conservation and traceability. Importantly, data from SNPs are especially amenable to archiving because they exhibit high reproducibility among different laboratories: newly collected data can then be readily compared with reference data with high reliability and ease. In addition, the identification of SNPs is not only responsive to changes unrelated to environmental differences, but also to natural selection, greatly improving the power to detect the distinct signatures of local and regional groupings.
On the other hand, phenotypic characteristics can also be robust indicators of the origin of fish, though they are not available for use throughout the food supply chain. Whole fish can be characterised by their appearance, such as body shape, counts of body parts, shape and composition of otoliths and composition of muscle fat. Since the early 1900s fisheries scientists have made regular collections of commercial fish, to monitor abundance, growth and condition. The otoliths collected for aging have frequently been archived, thereby yielding a valuable legacy to set a historical base line. These calcified structures grow in layers, similar to tree rings, throughout life. Seasonal and yearly increments are deposited, matching the age of the fish. But there is more. The width of each increment represents the growth history of each individual. The chemical composition of each increment reflects the local water mass, forming a permanent record, a kind of logbook, of where a fish was born and where it has lived prior to capture. Chemical methods, such as inductively-coupled mass-spectrometry (ICPMS), measure otolith chemical composition, in order to determine the origin and fishing ground. The approach is particularly interesting as a tool to validate catch records. In this way, fish from different areas can be distinguished from each other.
Otolith shape is another characteristic that distinguishes fish species and populations. Modern imaging techniques are used to photograph and analyse shape differences - such as the pattern of scalloped margins or notches. The example above compares the 'average' otolith shape between different hake populations caught in the North-East Atlantic Ocean and Mediterranean Sea. In combination, otolith shape and chemical composition are valuable tools to trace fish to their spawning and fishing ground. Interestingly it is possible to trace back in time the pattern of fishing and fish populations over past decades in the archived collection of otoliths housed at fisheries institutes world-wide. In many cases, small traces of fish tissue with its DNA remains attached to these otoliths. DNA can then be isolated and characterised genetically, similarly to recently collected tissue - a practice followed in FISHPOPTRACE.
Fisheries forensics: the policy relevance of a forensic framework
Forensics is a field of science dedicated to the methodical gathering and analysis of evidence to establish facts that can be presented in a legal proceeding. Evidence is sometimes required within the fisheries context, when regulations have been breached, such as illegal fishing activity, mislabelling of a fish product, or under-sized fish are captured. While there has been no shortage of policies and maritime laws to assist in the governance of our oceans, it has proven much more difficult to enforce maritime law and to execute prosecutions. In part, such constraints are related to having an appropriate set of tools to yield sufficiently robust evidence in a court of law. As previously stated, the global level of IUU fishing and supply chain fraud offer financial gains on a scale that attracts organised criminal groups as well as unscrupulous individuals; the tools required to detect and tackle such activities and enforce fishing regulations need to be equally sophisticated.
The outputs produced by FISHPOPTRACE help us to understand how fish populations are structured and allows us to look for ways of identifying where fish come from. These techniques can be developed into a series of applied tools for forensic analysis and tracing fish and fish products from ocean to fork. However, before this can be achieved, method validation must be performed to demonstrate that laboratory testing is accurate and reliable. The method validation stage focussed primarily on understanding the strengths and limitations of the DNA and otolith analysis techniques. Scientifically speaking, this involved assessing the level of variation among samples from the same population when tested under different conditions and comparing these data with the level of variation among populations.
The application of FISHPOPTRACE outputs - initial comments
It is also worth pointing out, prior to a more detailed consideration of FISHPOPTRACE results and dissemination, that a parallel aim of the consortium was to identify and promote mechanisms that would engage stakeholders and end-users in fisheries. While starting out as a research project, FISHPOPTRACE set the ambitious goal to move beyond and bridge the often prevailing gap between academic science, policy making and end-users. To this end, a technology transfer strategy was developed which included pro-active engagement with the CFP and communication with stakeholders.
Highlights of FISHPOPTRACE findings and outputs
Having provided some general comments on the overall aim, scope and design of FISHPOPTRACE, here we consider some of the key findings arising from our research, as well as examples of how these have been communicated to a diverse community. Complementary information is also available from the consortium website (http://fishpoptrace.jrc.ec.europa.eu/) which is updated regularly, as well as remaining 'active', with further development, after the conclusion of the project in May 2011. FISHPOPTRACE was organised into a series of work packages (WPs), details of which can be obtained from project documentation and the FISHPOPTRACE website. The WPs, which essentially follow a logical flow from establishing the state of art within the fish traceability field by accessing details of related European project, moves onto sample acquisition, traceability tool development and validation, to their eventual full deployment across the many geographic and spawning populations of our target fish species. The flow of work throughout has utilised core principles of technology transfer, dissemination of activities and outputs and management.
What was the state of art in European marine fish population structure and traceability at the outset of FISHPOPTRACE?
Tremendous progress has been made in the fields of marine fish genetic improvement, genetic stock identification and genomics in recent years. The continuing development of new molecular genetics tools has provided high-resolution markers for assessing genetic population structure, for estimating demographic parameters and for providing insights into fish stock management. However, while part of this information is available in the primary scientific outputs (scientific journals, etc.), there is another kind of information, sometimes scattered, in the way of technical information, collections and databases held by individual experts or institutions, which are still very useful.
Having that in mind, a primary objective of FISHPOPTRACE was to provide a summary review of existing information bases for marine fish genetic resources and to place the FISHPOPTRACE project within the context of recent and on-going fish traceability programmes. To achieve this, three main tasks were fully accomplished during the first year of the project. Firstly, the establishment of a common platform for accessing archived tissues for genetic analyses from past and current EU and national projects. By searching at http://fishpoptrace.jrc.ec.europa.eu/tools/projects it is possible to find the links to other existing projects related to genetic identification and traceability of marine fish species. Secondly, filtering of valuable data produced and stored in other fish traceability projects and compiled into the new FISHPOPTRACE database, the 'meta-crawler', accessible at http://fishpoptrace.jrc.ec.europa.eu/crawler. Finally, the establishment and the distribution of a tissue archive database of external samples that has been extensively used during the project, especially the historical samples. Links to collections of specific external samples with biogeographical information and source and distribution of samples external to the FISHPOPTRACE general sampling scheme can be accessed at http://fishpoptrace.jrc.ec.europa.eu/data-access. The latter resource is an especially distinctive and valuable contribution to the community of scientists interested in traceability and population structure of marine fishes. It provides an integrated platform for accessing 12 EU-related projects associated with fish population structure and/or traceability, bringing together under a single umbrella, the recent collective effort and outputs within the field. Moreover, it aims at addressing the obstacles and missed opportunities arising from the fragmented nature of many tissue archives. Finally, the design of the portal allows for ease of use and access, with minimal prior experience required in using the interfaces, thereby allowing a range of end-users to benefit from its use.
How did we ensure the collection of representative samples of cod, herring, hake and cod from European waters?
The ultimate goal of FISHPOPTRACE was to provide applications in support of fishery management, including control, enforcement and traceability. However to get there, FISHPOPTRACE started out as a research project and asked the question: 'Can we detect distinct populations of cod, hake, herring and sole in European waters?' Due to the apparent openness of the oceans, migration over long distances of many marine species and dispersal of eggs and larvae over vast areas, the notion that marine fish tend to breed randomly across their geographic distribution and do therefore not show population structure prevailed for a long time. FISHPOPTRACE aimed to identify distinct features of target fish populations. These markers should ultimately allow the assignment of fish to their population of origin. Fortunately, in recent years, research has shown that many marine fish are indeed subdivided, occasionally over small spatial scales. This sets a robust rationale for FISHPOPTRACE and our question can be refined to 'can the population structure of cod, hake, herring and sole be determined at a scale that is useful for fishery managers and control authorities'?
To answer this question scientifically, careful planning and consideration was required, starting with the collection of fish at sea before analysis in the laboratory. An elaborate sampling strategy must be developed to maximise the probability that all potential populations of the target species are included across the area under investigation. To this end, the FISHPOPTRACE sampling team designed species-specific sampling schemes covering the European seas, taking into account the species life-history traits, population dynamics and structure and environmental features. Standardisation procedures were developed from on-board species identification and tissue collection to documentation, storing, archiving and cataloguing in laboratories and in a central database system. In addition, sampling progress was monitored online on a publicly available sampling map. A specific achievement of FISHPOPTRACE was the species traceability of the common sole and of its cryptic species solea aegyptiaca. The fish collected during the sampling phase were analysed in the laboratory using state-of-the-art technologies, namely high throughput DNA sequencing, ICPMS and proteomic and gene expression profiling.
How can we develop a common set of traceability tools for identifying fish populations across different spatial scales, as well as tracing diverse natural and processed fish products?
A major objective of the consortium was to develop a set of traceability tools that could be applied comprehensively to key spawning populations of each of our target species. It was also important to develop tools that could be used at different stages of the food supply chain and that could also ideally reveal information about population structure and traceability at different levels. The genetic markers, SNPs and analysis of chemical signatures and shape variation in ear bones (otoliths) were chosen. As already mentioned, SNPs can be retrieved throughout the food supply chain. In addition, for many decades, fisheries and other laboratories have archived otoliths in connection with information on growth rates obtained from examining the otolith growth rings. Through our consortium extensive links, these archived and stored otoliths provided an additional set of samples from which DNA could be obtained, as well using them in the chemical and shape analyses.
Among the most striking scientific results is the provision of several hundred novel genetic markers in, hake, herring and sole. The genetic markers in cod were already largely available from another project ('Canadian Cod Genome' project), with which FISHPOPTRACE had an agreement. Access to new genetic methods, the so-called next generation sequencing, has changed the picture in a matter of just a few years. From a dozen genetic markers a few years ago, we now have knowledge about thousands of small genetic differences (genetic variation) at numerous genes, allowing the design of hundreds to thousands of new genetic markers. The unique combinations of the variation make it feasible to assign the fish to specific populations and in some conditions to identify unique individuals.
For SNP development, we first sequenced a high number of so-called expressed sequence tags (ESTs) from the muscle of sole, hake and herring. After assembly, a large number of SNPs were identified using a range of approaches. We first performed a visual inspection of 1 536 SNPs per species and along a pipe line of detailed analyses followed by validation, and then chose the best panel for traceability and forensic analyses to be used in WPs 5 and 6. A total of 1 200 to 1 300 samples per species were analysed during the validation-genotyping step and initial genetic analyses were performed to define the final analysis panel. It was important to ensure that any SNP markers chosen were easily scored using our high throughput methods, where many hundreds of SNPs can be typed simultaneously, as well as ensuring that they did represent accurately the underlying genetic variation in individuals; it is the cumulative sum of genetic variation across hundreds of SNPs from a population that ultimately determines to what extent we can distinguish it from other populations - a co-called 'population signature'. The following number of reliable SNPs was defined: 426 SNPs for sole, 281 for herring and 395 for hake. In conjunction with the previously-mentioned cod project, we were able to type 1 290 SNPs.
Procedures for the analysis of otolith microchemistry have been standardised and optimised to provide the most robust and clear signal for otolith core information and the narrowest otolith edge signal. Additionally, the methodology for otolith shape analysis was standardised to produce reliable digital photographs. Protocols for image analysis have been developed to automate the processing of the images and the statistical analysis of shape.
What about forensic validation of our primary traceability tools?
The remit for FISHPOPTRACE was to take the research and technological development (RTD) outputs across work packages and evaluate their suitability and readiness for application to fish traceability by performing a series of validation studies. These validation studies would generate the data necessary to assess the robustness, reliability and reproducibility of tools developed within FISHPOPTRACE for monitoring or enforcement purposes. Briefly we wanted to establish a set of criteria for the validation of traceability tools; undertake developmental validation of each identification technique; develop statistical methods for the identification of source populations using multiple markers; implement internal validation studies and inter-laboratory calibration of techniques and collate all validation data to enable the construction of standard operating procedures (SOPs). Work towards these objectives was conducted across multiple techniques and for several different aspects within a single technique. As such, the work conducted was broad in nature and required a high level of integration with other WPs, as well as the direct involvement of multiple partners with diverse areas of expertise.
A large amount of effort was put into the otolith validation work and many advances were made, as the basic technologies used for both the microchemistry and shape analysis were essentially specific to individual laboratories and were not routinely used for the forensic identification of biological material. The largest component of validation work however focussed on the validation of molecular markers. In order to deliver a validated assay, the marker validation work expanded to include a range of tests. In each case, a second laboratory was used to test the analytical protocols and ensure that comparable data were generated using the same method. For otolith analyses, control samples were prepared by using the matching otolith pair to ensure that testing was based on a like-for-like comparison. In addition, the otolith microchemistry analysis uses commercially available elemental calibration standards to ensure that the analytical baseline is identical in each laboratory. For the genetic markers, standard control samples representing each genotype for each SNP marker were prepared. These control samples are available alongside the standard protocols generated for transfer into testing laboratories beyond the lifetime of FISHPOPTRACE. For the otolith microchemistry and morphometric analyses, variation among laboratories was observed, as expected, though the magnitude of the measurement differences was larger than anticipated, particularly for microchemistry. These results were one of the main reasons why the otolith microchemistry analysis method is not yet considered to be suitable for routine use in the production of forensic evidence.
The results of from the forensic studies are documented in a series of validation protocols, reports and analytical methods. As the validation process is designed to be entirely objective, the fact that several techniques were not considered to have performed sufficiently well to pass through to formal application is not seen as a failure; on the contrary, the ability to critically evaluate each technique based on quantitative, comparative data demonstrates the success of outputs in terms of its intended function. In the final analysis, the SNP markers were the only technique demonstrably suitable for application within a forensic framework; the other markers, while often extremely informative, are not yet applied in a format that is likely to withstand the scrutiny of legal challenge.
How informative were the SNP and otolith traceability tools when applied to fish populations and processed products?
It was important for FISHPOPTRACE not only to provide informative data on the structuring and dynamics of main spawning populations of cod, herring, hake and sole, but also to demonstrate the feasibility of obtaining large quantities of robust traceability data in a timely and cost-effective manner. Thus, there was considerable discussion about the most appropriate method for genotyping the SNPs, that is, the routine determination of the genetic variants across many SNP loci in each population. A large number of genotyping methods and platforms were evaluated resulting in the selection of an Illumina 1536 golden gate array; as our general genotyping method when exploring many hundreds of SNP markers simultaneously. This array was used for a single step validation and genotyping approach for putative SNPs and all population samples of the targeted species. Analysis of otolith microchemistry and morphometrics were conducted in parallel for a large subset of the samples used for genetic analysis. Based on the SNP genotyping, genetic maps illustrating genetic relationships among population samples were created.
Separate maps of so-called 'neutral' and loci 'subject to selection' were created as well as maps including all loci. Distinct regional groups were apparent using both classes of markers, where Atlantic and Mediterranean basins were the most distinct. However, also on a smaller geographical scale distinct groups were obvious, such as minor basins within the Mediterranean and the Baltic Sea. On a very fine scale, markers under selection revealed previously unrecognised population structure, highlighting the concept of investigating variation in genes as very powerful for resolving fine scale population structure. The exact number of loci subject to selection varied according to species, geographical region studied and statistical method employed. Overall the outlier loci appeared to be associated with major climatic differences among populations. In total the high number of markers and those with highly elevated levels of genetic differentiation provided unprecedented high power for assignment back to population of origin. At large and medium geographical scales, minimum panels including only a subset of loci could provide close to 100 % correct assignment to population of origin. On a small geographical scale more loci were needed, however in most cases highly informative results could be provided.
The core approach for the genetic component of the work was the development of a 'SNP chip' for each of three species sole, hake and herring. Canadian researchers had already created a SNP chip for cod to assist in aquaculture research. These DNA-covered microchip-like devices enabled testing the identity of 1 536 possible SNPs for each group of individuals from a specific population. Once all individual fish were characterised using the chip, the frequency of each SNP variant was determined, creating a potentially diagnostic pattern, or 'population signature'.
Based on our analyses, as far as the new SNP markers are concerned, it has been possible by varying the numbers used on a SNP-chip, to assign individuals back to their source population across different geographic scales with high levels of certainty and reproducibility. Such outputs are especially significant since previous types of genetic markers either detect levels of population differences that are too low, or there are inherent difficulties in comparing data generated from different laboratories. The use of a marker system such as SNPs, which is essentially based on the presence or absence of large numbers of single genetic variants means that data can be compiled from sources in a much more reliable and high throughput way. The approach thereby enables the generation of baseline and ongoing additions for subsequent genetic monitoring. Moreover it is imperative that any such tools can be used in a legal context, necessitating forensic validation. This has been achieved for SNP markers within FISHPOPTRACE across a range of policy-driven IUU scenarios.
In addition to improved traceability of fish populations, the DNA code of thousands of genes now makes it possible to link the function of genes with the phenotype/external characteristics of each fish. The long-awaited link between natural selection and the genetic blueprint has now become possible in natural marine fish populations. Such an advance is significant because scientists are poised to move beyond the mere detection of genetic differences among marine fish populations, to the identification of how and why such differences relate to their survival in stressful environments, so-called local adaptation. Such information can inform managers on the ability of populations from specific localities to adapt to natural and man-made changes, including over-fishing, contaminants and climate change.
The results of the otolith analysis provided independent insights into the population structure and ecology of the targeted species. The information generally corroborated the genetic results, but also in some cases provided complementary information with higher resolution power, in particular on a small geographical scale.
The tools developed and tested within FISHPOPTRACE provide a giant leap forward in terms of tracing fish and fish products. Here we provide new, more powerful, fast and cost-effective tools as well as the baseline data on a pan-European scale for four economically and culturally highly valuable species. We are now able to trace the fish from the four species from ocean to fork in all types of products. In some areas, however, fish from different populations meet and mix. Establishing mixture signatures for these areas is a high priority. Likewise, although our sampling effort has been substantial there may be populations which have not been fully sampled. However, due to the flexibility of our approach, more data can be added to the database with ease, as there is little need for calibration of SNPs compared to previously applied genetic markers. Finally the implementation of the methods into practical fisheries management and control and enforcement should be a priority for the project members and for the EU.
In summary, we applied our primary traceability tools, i.e. DNA SNP analysis and otolith microchemistry and morphometrics to test their power for tracing fish from the four species back to population/area of origin. Such information provides a framework for a pan-European traceability database for fish and fish products to be applied for defining management areas, fighting IUU and for assuring safe, healthy and sustainable fish products for the European consumer. Our focus has been on the traceability of populations, but each test can also incorporate a species identity marker. The latter is important because over 60 % of fish products consumed within Europe derive from imports outside the EU. Any traceability system that allows simultaneous testing of species and provenance will have significant global application.
How stable are our traceability tools over time?
Any traceability system needs to take into consideration whether the method is resilient to changes over time that may alter the accuracy and precision with which fish and fish products can be traced. We therefore needed to establish the time scale over which the developed tools work, i.e. not just for the populations in specific areas as they are now, but also in the future. It is of interest, therefore, to establish whether or not temporal changes in populations can be assessed using the marker approaches developed in FISHPOPTRACE. Potential changes include long term changes in the sizes of local populations, e.g. in response to global change and exploitation. Another type of change may be geographic, if individual populations change spatial distributions, for example in response to environmental changes to spawning or feeding grounds, or if their relative contribution to mixed-stock fisheries change. All major changes need to be taken into consideration for sustainable management of resources and for conservation of biodiversity. FISHPOPTRACE has examined such issues. The approach is to assess the stability over time (temporal stability) of the developed population markers back in time based on analyses of a series of samples collected on different dates and to model which levels of future change can be assessed using the markers at hand.
Due to the biological differences among both species and populations within species, as well as the management and conservation issue arising, the scopes for monitoring temporal change will vary on a case-by-case basis and will operate on different time scales. We specifically targeted our work to reflect and represent different types of monitoring objectives.
The FISHPOPTRACE traceability tools have also proved highly informative for addressing potential changes in migratory behaviour, both in terms of annual changes in where populations feed and whether fisheries exploit different stocks disproportionately during different times of the year, as well as for assessing the persistence and dynamics, of local populations. For otoliths, the temporal studies indicated overall stability, but also some differences between samples from different dates within locations that should be taken into consideration when developing tools for traceability and determination of management units. The genetic marker analyses in all cases demonstrated large potential for genetic monitoring. Such tests over time for stability coupled with analysis of population structuring demonstrate a set of valuable tools for tracing the locations and movements of individuals in space and time. Analyses can now be targeted to real-time monitoring of seasonal or annual changes in population exploitation rates. The genetic analyses of longer time series conveyed novel information on other aspects of genetic monitoring. Such methodologies can also be targeted to monitoring population changes in adaptive responses to climate change. We also demonstrated the SNP-based method's power for monitoring population origin of individuals in data poor areas. Other types of markers, such as otoliths also exhibited temporal stability although temporal differences were also evident. Such differences underline the care that needs be taken if applying these types of tools in monitoring programmes.
In summary, we show that population-specific traits vary little over time across species, demonstrating that the developed tools are reliable for insights across shorter and longer time scales. The observed temporal stability and conversely lack of short term change in population structure, is an important step in validating the traceability approach as reliable in time and space. Needless to say, future changes in population structure and the spatial distribution of individual management units are difficult to predict. Nonetheless, FISHPOPTRACE has laid the basis for a monitoring framework that can be developed and extended in years to come.
In addition to SNPs and otoliths what other traceability tools might be developed for future application?
Besides developing applications based on genetics and otolith analysis for traceability of fish and fish products, FISHPOPTRACE also explored the value of 'novel tools'. Novel tools are new ways to analyse fish products that could complement more established methods and help to trace individual fish to the place where they have been fished. To this end, two major goals were set. The first one was to develop a DNA chip, a miniaturised tool with improved sensitivity, speed and costs, based on a sophisticated technology that uses a special glass slide with DNA 'sensors' spotted onto it. A pilot study on cod individuals selected from different geographic regions showed that this new technology is very promising but requires further optimisation. Two different versions of a prototype platform for SNP genotyping were developed and tested on a test panel of cod SNPs and samples. Both are based on a so-called 'padlock probe technology' and a second version, with a smaller number of targeted SNPs, showed very promising results, opening the possibility of routine genotyping of an informative SNP panel (10-100 loci) at a reduced cost.
The second goal was to find additional traceability markers. The idea is that environmental differences, such as water temperature, or the nutrients taken up by fish, might influence specific features locally, making it possible to trace where the animal was living. Firstly, we looked at the composition in fatty acids (FA) of fish flesh. FA analysis was performed on all four target species (334, 292, 132, 98 samples respectively for sole, hake, cod and herring), exploring spatial and temporal variation, demonstrating that geographic populations can be effectively differentiated according to FA composition. Statistical analysis of the FA composition in the tissues showed a clear differentiation of geographical populations, both within the Atlantic or the Mediterranean as well as between these two basins. Secondly, we examined which genes are switched on or off, that is expressed (active) or not, in the muscle tissue of hake and sole living in different areas. A DNA microarray was developed to analyse 14 898 genes in the muscle of hake collected from six different areas. We demonstrated that for certain areas it might be possible to trace back fish to their geographic origin measuring the expression of few marker genes. Thirdly, we looked at protein expression. Recent advances in molecular biology have greatly facilitated the development of powerful tools for proteomic analysis. A proteome scan approach based on two-dimensional fluorescence difference gel electrophoresis (DIGE) technology generated thousands of protein markers that allow the identification of different hake populations. Among the identified proteins, some that reliably assigned individuals to their population of origin have been selected to be used as biomarkers for traceability of hake fish products. Currently an antibody array specific for hake protein markers is under development, to make the assignment of individuals to the original stock feasible as a diagnostic routine.
Overall, our quest to test the potential of additional traceability tools yielded very exciting results and opened the way for the use of novel technologies to complement more established tools (genetic markers, otoliths shape and composition) in traceability studies. It should be noted, however, that this was an exploratory exercise in FISHPOPTRACE and these novel tools need to be validated fully before being used for control and enforcement purposes. Importantly, any traceability system needs to have sufficient flexibility such that constraints imposed by the specific requirements of any specific technique can be matched with available resources and opportunities. Additionally, where required, information on spawning population as revealed by genetic tools, might be usefully complemented by knowing what water body the fish was captured from, such as otolith microchemistry, lipids or gene expression. The genetic data typically generates population-level signatures across a longer-term evolutionary scale based on the average breeding behaviour of fish populations across generations, whereas the other marker types have potential to reveal provenance across a shorter-term, ecological time-scale.
Once all the traceability results have been generated, how can these be archived, retrieved and made accessible to others?
While large collections of information are created during the sampling and analysis phases of projects such as FISHPOPTRACE, usually only a small subset of this information is published as primary scientific outputs. The body of data generated during the lifetime of the project remains vastly inaccessible, as collections and databases are maintained by individual experts or institutions. Such data sources may have a high potential value for future questions, but might get lost or forgotten as new projects begin or objectives change. To avoid such a scenario, a flexible database structure was designed and implemented for FISHPOPTRACE. The database is housed at the JRC of the EC and was designed to grant access to data and further addition of data beyond the project duration.
The FISHPOPTRACE website was constructed and was operational from the early stages of the project acting as a source of information about the project and similar projects. A member section of the website was constructed as a file sharing resource. This store has accumulated a wealth of project derived data, articles, presentations, reports and related documents over the project lifetime.
The FISHPOPTRACE database was designed to be as data independent and flexible as possible through strong collaboration from all consortium partners, including a database-specific meeting hosted by the JRC. Following its construction it has been hosted and maintained by the JRC, including backups, versioning and modifications/bug fixes. An Excel template was agreed on and used for uploading the characteristics of collected specimens to the database. A web-based interface, deployed in the website, was developed to allow consortium members to query and view the contents of the database. Additional functionality was added to allow the uploading of samples to the database through the interface, using the Excel template. Furthermore, the results of user-specified queries through the interface can be downloaded and saved in the Excel template format. These Excel sheets can be edited and uploaded again through the web interface. The interface automatically handles updates to samples, deletion of samples and the insertion of new samples. A crawler tool was developed as part of our initiative to act as a data hub. This crawler automatically searches for news and articles related to FISHPOPTRACE by matching a set of key words. The results are collated and updated daily on our website. Information stored in the database structures can be accessed in two ways through the FISHPOPTRACE website, the information dissemination portal of the consortium. Selected genetic characteristic information can be publicly accessed through a map-based interface, designed to highlight biological characteristics of the target species in relation to their environment in a geographic context. Members of the consortium can also login and query the database, view the information online or download the information in a spreadsheet to apply changes 'at home' before uploading the edited data sets.
The purpose of the geographic visualisation platform is to disseminate information, also to non-experts, in a transparent and easily understood way for use as a management and regulation decision support tool. Data are provided in a geographic and environmental context. Aspects such as ocean currents, surface temperature and salinity value variations and their potential relationship with the development of features, such as genetic differences in the target species, can be visualised. This provides valuable support to research, but constitutes also a useful communication tool for population/stock structure information to stakeholders involved in fishery management and policy development.
Potential impact:
FISHPOPTRACE is designed to integrate information on the distribution, extent and dynamics of fish population structuring in four target species, as a basis for conservation in relation to sustainable utilisation of bioresources and in the provision of population markers for traceability of fish and fish products. As previously mentioned, a key strategy of FISHPOPTRACE was to bridge the often prevailing gap between academic science, policy making and end-users. To this end a technology transfer strategy was developed, which consisted of intensive and targeted networking and communication with policy institutions, authorities and stakeholders, engagement with the CFP and a cost benefit analysis (CBA). Here we present salient highlights, rather than an exhaustive account, of how we aimed to facilitate uptake of outputs by end-users.
FISHPOPTRACE website, documents, publications, consultations etc.
The approach followed by FISHPOPTRACE has generated a paradigm of a holistic framework for fisheries control and enforcement. Peer reviewed scientific publications have already and will continue to ensure that the results of FISHPOPTRACE are scrutinised and disseminated in the research area. Such dissemination of outputs, however, is unlikely to be sufficient to reach out to end-users, which is why FISHPOPTRACE stressed the importance of inter-laboratory trials of the protocols developed and used by the project. The consortium also stressed the central role of the FISHPOPTRACE database and interface. The experimental FISHPOPTRACE data has been validated by the consortium members and submitted to the JRC, which acts as the host of the databases. The consortium spent considerable effort on finding ways how best to use the FISHPOPTRACE website to generate maximum benefit for stakeholders and end-users. This is well reflected in the development of the web-based project crawler tool developed in the early stages of the project. Ultimately it was agreed that a geo-visualisation platform would generate a practical opportunity to support the combination of experimental data and results with environmental parameters, in a so-called 'seascape genetics' approach, on one side and to demonstrate experimental outcomes and their value for applications to stakeholders and end-users. Such environmental parameters will be enlarged progressively in collaboration with the JRC institute for environmental sustainability (IES). Importantly it is also possible to visualize and overlay fisheries management sections such as the International Council for the Exploration of the Sea (ICES) or the Food and Agriculture Organisation (FAO) subdivisions. Thus, the findings from FISHPOPTRACE open the door to various valuable applications for fisheries management, especially when displayed and accessible in the context of geographic information systems (GIS).
A key component of technology transfer is to ensure that strictly designed protocols of traceability can be generated for practical uptake; the so-called SOPs. SOPs are documents containing instructions that forensic scientists and laboratory staff follow to perform procedures that are routine, standardised and for which no ad hoc modification is acceptable. They are indispensible when supporting crime investigations scientifically as they help to ensure the quality and integrity of data and provide a basis for guidance, uniformity and accountability. Within the project SOPs for selected analytical scenarios have been developed.
A significant dissemination opportunity was provided by the EC Directorate General (DG) RTD as they invited the FISHPOPTRACE consortium to join the organising steering committee for the food chain integrity conference, a major dissemination event on traceability to be held in September 2011 in Brussels; a book chapter has resulted from the original invitation to participate, to be presented officially during the conference.
A notable highlight in terms of international exposure of FISHPOPTRACE was the publication of a two-page news focus article in Science, in December 2010, entitled: 'To fight illegal fishing, forensic DNA gets local', in which Erik Stokstad, environment and sustainability editor for Science, described how FISHPOPTRACE had produced a new generation of genetic tests that could provide authorities a much better idea of exactly where fish have been caught. Profiling of FISHPOPTRACE in Science not only acknowledges the state-of-art approach employed by the consortium, but importantly also provided detailed examples of how the SNP-based tools in particular had been employed to tackle various policy-led case European traceability case studies for each of the target fish species.
An amendment to the FISHPOPTRACE technical annex, associated with the granted three month no-cost extension, was the production of a stakeholder pamphlet, published in May 2011 in conjunction with the Brussels stakeholder workshop. The remit of the brochure was to present the context, key findings and policy relevance of FISHPOPTRACE activities and outputs, presented in an attractive, accessible and concise format. In addition to a contribution from Mr Timothy Hall, Head of unit 'E4: Agriculture, forests, fisheries and aquaculture DG RTD', the brochure presented details of consortium membership, the role of traceability and population structure in sustainable management of fish resources, the challenges of IUU and fish fraud, key methods and findings from the project, as well as the extent of exposure of the project in the media. The brochure concluded with a consideration of mechanisms and opportunities for securing a legacy to FISHPOPTRACE and a call for need to extend the impact of FP7 science projects beyond the natural life of the funded period. One-thousand copies of the brochure were produced and distributed throughout the consortium.
Meetings, conferences etc.
A major dissemination opportunity for FISHPOPTRACE was the participation of various partners and co-organisation of a major international symposium held in Seattle, United States of America (USA), in March 2010, 'SNP Symposium III: Applications of SNP genotyping in non-model organisms'. Members of FISHPOPTRACE also presented the application of forensics at the sea food summit (Paris, 2009) and the global fishery enforcement training workshop Maputo; 2011) organised by the international monitoring, control and surveillance network (iMCS), which is endorsed by the FAO. Through our consortium member, Robin Waples, FISHPOPTRACE has built a link to the USA federal agency National Oceanic and Atmospheric Administration (NOAA), a key scientific advisory body for fishery management, including control and enforcement. Several members of the FISHPOPTRACE consortium are also members of the ICES working group on applied genetics for fisheries and mariculture (WGAGFM), thereby building a permanent link between FISHPOPTRACE and ICES. Furthermore, primarily through the JRC, contributions from Jann Martinsohn have promoted ways to better include modern molecular technologies in fishery management decisions under the CFP remit. Particularly noteworthy is also that FISHPOPTRACE has been invited in April 2011 to present project and results at the plenary session of the Scientific, Technical, Economic Committee for Fisheries (STECF), which resulted in a highly positive assessment of the potential of genetics for fisheries management in the plenary report which will be submitted to the EU member states.
Martin Taylor and Gary Carvalho organised a DNA barcoding workshop at the University of Penang, Malaysia, in November 2009, with a focus on the application of barcoding technologies to fish biology and fisheries management. A key application of the barcoding pipeline was traceability in fisheries forensics and there was also the opportunity to present aspects of the FISHPOPTRACE project to workshop participants, as well as seminars in the Fisheries Research Institute, Department of Fisheries, Malaysia, the University of Malaya in Kuala Lumpur and the Centre of Research for Computational Science and Informatics in Biology, based at the University of Malaya and the Malaysian Fisheries Department in Penang. The format of the workshop included seminars and hands-on experience of the BOLD database, which also forms a component of the FISHPOPTRACE traceability framework.
A meeting was held at the FAO, Rome in December 2009 to discuss the use of forensic approaches in the context of MCS and IUU fishing. An invitation was received from Michelle Kuruc, who attended the FISHPOPTRACE Madeira workshop, for four members of FISHPOPTRACE to participate: Jann Martinsohn, Rob Ogden, Einar Nielsen and Gary Carvalho, as participants. The aim of the workshop was to identify the most appropriate traceability tools for use within a forensic framework and to generate a set of recommendations for taking the approach forward within a global context. The workshop included broad representation from a variety of nations and backgrounds, including forensic biologists, fisheries compliance officers, directorates for fish inspection quality control and various research organisations and universities with experience in the development and application of advanced tools, as well as their implementation to fisheries. The workshop provided a key opportunity for profiling the objectives and progress of FISHPOPTRACE, which was deemed a significant international effort in the quest to tackle IUU and promote consumer protection. An output from the workshop was to provide support for the establishment of an FAO expert group, in fisheries forensics.
Other modes of dissemination
In addition to extended interest from the popular press and media (including 'Science', 'New scientist' and 'the Economist'), FISHPOPTRACE has enjoyed enhanced profiling from within EC instruments. For example, FISHPOPTRACE was profiled on the CORDIS technology marketplace, as an example of cutting-edge technologies and applications to environmental management, 'So our fish do not disappear without trace', as well as being chosen as a 'success story' for the EC's CORDIS website, project success stories - A sustainable future for fishing and for inclusion in association with Euresin (ygaethofs@euresin.eu via e-mail), a new, on-line database that will cover all life science projects funded under the knowledge based bio-economy (KBBE) programme in FP7.
Bangor University submitted a Press Release on FISHPOPTRACE in the early spring which resulted in an article in Fishing News (see http://www.fishingnews.co.uk online), entitled 'Protecting the fishing industry: New £3.6m programme to safeguard fish stocks for industry and consumers'. The press release also stimulated a response from the United Kingdom (UK) body charged with the management and conservation of marine resources, the Department for Food and Rural Affairs (DEFRA): Tim Bostock, international fisheries policy adviser to DEFRA's sustainable fisheries team emphasises that the UK has taken an international lead in the fight against illegal fishing. 'As part of our work under the ministerial high seas task force, we have been working with like-minded countries in developing approaches intended to deter illegal operations by hitting them where it hurts- in their wallets. Forensic tools such as those being developed (in FISHPOPTRACE) will provide a powerful deterrent and help protect those who wish to operate legitimately. Tackling illegal fishing is an important facet of more effective fishery management and the monitoring, control and surveillance operations. A successful completion of this project will certainly be welcomed by agencies charged with the enforcement of conservation management' he said.
In addition, articles appeared in the local press and the Coordinator, Gary Carvalho, was interviewed about the project on the British Broadcasting Corporation (BBC) Radio Wales, for the science programme, 'Science Café'. The project has been publicised on the Bangor University homepage under 'Latest News'. FISHPOPTRACE also enjoyed additional exposure on the UK Natural Environmental Research Council, Planet Earth on-line, which profiles via podcasts, timely and significant studies in environmental sciences. The podcast, Cracking down on illegal fishing, presented an interview with Sarah Helyar and Gary Carvalho from Bangor University, in which they described the context and impact of IUU fishing at European and global levels and the use of molecular approaches, including those developed within FISHPOPTRACE, in tracing fish products for monitoring, control and enforcement.
FISHPOPTRACE fully acknowledged that in order to transfer results and tools to stakeholders beyond the scientific realm, major efforts have to be put into achieving maximal visibility, while not overselling potential benefits (impartial dissemination). Obviously the publication of articles in peer reviewed journals had to remain one major dissemination component. Dissemination of results has also been achieved through pro-active engagement with stakeholders during the reporting period. This includes invitation of various external 'observing' scientists such as Dr Petra Spaniol (DG Fisheries) to our kick-off meeting in April 2008; Dr M Kuruk, FAO, Fisheries Forensics in Monitoring, Control and Surveillance (workshop two); Dr Jan van Aken, EU Marine Programme, Pew Foundation (workshop two); Dr Harald Barne, DNV Research and Innovation / Biorisk Management (workshop two); Dr. K. Glover (Institute for Marine Research; Norway), a renowned fisheries geneticists working, among others, in the field of traceability of fish farm escapees and monitoring of genetic impacts on native fish populations, to the 5th FISHPOPTRACE consortium meeting.
Project website: http://fishpoptrace.jrc.ec.europa.eu/
The contact details of all partners are available at http://fishpoptrace.jrc.ec.europa.eu/contacts
Marine fish species, in common with all living organisms, are fragmented to varying degrees, into a series of locally interbreeding populations. The extent to which such populations differ biologically ('population diversity') and their distribution in time and space are among the most important drivers of species' survival and persistence in the face of environmental change. Fisheries biologists and managers have emphasised such thinking since the early 20th century, though there typically remains a mismatch between biological populations and fisheries management units. Thus, it remains difficult to devise strategies to relate population diversity to variable harvesting regimes and even more difficult to conserve overexploited stocks. Further uncertainty in striving for sustainability is the high level of illegal, unregulated and unreported fishing (IUU), estimated globally to cost the industry EUR?10-20 billion and prior to 2010, EUR?1.1 billion worth of illegal fish was imported into the European Union (EU) annually. FISHPOPTRACE, a 15 partner EU Seventh Framework Programme (FP7) project, aimed to address such challenges to sustainability within the context of the Common Fisheries Policy (CFP) to:
1. develop a range of cost-effective and reliable tools for identifying, monitoring and tracing marine fish populations in four representative European species (cod, herring, hake and common sole);
2. promote fisheries governance by ensuring that the most effective tools can be applied to forensic standards and thereby be legally supportive for prosecution and enforcement;
3. foster technology transfer of outputs in relation to enforcement and conservation policies of the EU CFP and associated socioeconomic consequences.
We have applied our primary traceability tools: deoxyribonucleic acid (DNA) single-nucleotide polymorphism (SNP) analysis (single genetic variants) and otolith ('calcareous ear bones') microchemistry and morphometrics (form, shape) to test their power for tracing fish from the four species back to population/area of origin. Such information provides a framework for a pan-European traceability database for fish and fish products to be applied for defining management areas, fighting IUU and for assuring safe, healthy and sustainable fish products for the European consumer. Our focus has been on the traceability of populations, but each test can also incorporate a species identity marker. We can now correctly assign fish to populations from more areas and with higher certainty than previously possible, reaching standards which can be used in a court of law. Based on use of the most highly distinct genes among populations we have already developed 'minimum assays with maximum power' with from 10 to 30 SNPs. These assays have been developed to target some of the most pertinent needs for traceability tools in European fisheries management. For example, we now have fast, efficient and forensically robust tools to discriminate between cod from Canada, North Sea, Baltic Sea and Northeast Arctic populations, between North Sea and North Atlantic herring, between sole from the Irish Sea and Thames and between hake from the Mediterranean and the Atlantic. The use of a marker system such as SNPs, which is essentially based on the presence or absence of large numbers of single genetic variants means that data can be compiled from sources in a much more reliable and high throughput way. The approach thereby enables the generation of baseline and ongoing additions for subsequent genetic monitoring. Moreover it is imperative that any such tools can be used in a legal context, necessitating forensic validation. This has been achieved for SNP markers within FISHPOPTRACE across a range of policy-driven IUU scenarios. FISHPOPTRACE outputs will thus contribute to efficient fishing activities within an economically viable and competitive fisheries industry and hence contribute to the CFP's aim of providing a fair standard of living for those who depend on fishing activities as well as taking into account the interests of consumers.
The aim and objectives of FISHPOPTRACE
From the establishment of FISHPOPTRACE, we have highlighted the central role that population diversity plays in sustainable utilisation and conservation of exploited stocks. FISHPOPTRACE was aimed at improving our ability to trace fish and fish products through enhanced understanding of the dynamics, temporal stability and distribution of major populations of four key exploited fish species: European hake, herring, sole and cod. Data and traceability tools have been integrated in a forensic framework to enable traceability within the context of enforcement and conservation policy. This in turn will facilitate implementation of effort limitation and regional allocation of quotas based on the relative abundance of respective populations or stocks, not only in the wild, but also through enforcement based on fish products. Traceability tools can also play a prominent role in consumer protection both at the species and population level, especially in the face of frequent mislabelling, fraud and to sustain 'ecolabelling' schemes of stocks certified in various ways.
The FISHPOPTRACE consortium consists of 15 partners with expertise in fish biology, population and conservation genetics, molecular biology, biochemistry and wildlife forensics. Partners were drawn from a range of representative bodies including the food industry, the European Commission (EC), universities, government laboratories and small and medium sized enterprises (SMEs). Collectively, the objectives encompass to:
1. integrate recent and on-going data from European fish species traceability projects and generate a single compatible database and tissue archive managed by the joint research centre (JRC) of the EC;
2. develop and apply a new type of genetic marker, applied for the first time to traceability and population structure of marine fishes: SNPs, together with tools developed from otolith microchemistry and shape variation in widely distributed populations of cod, hake, sole and herring, to discriminate biologically differentiated populations as a basis for traceability;
3. undertake validation of traceability tools to maximise robustness, transferability and reliance for subsequent use by end-users;
4. develop a population monitoring system based on otolith and genetic data that will assess the extent to which population traceability might change over time from various locations;
5. test the utility of additional novel traceability systems as potential independent tools in traceability;
6. foster technology transfer of FISHPOPTRACE outputs in relation to enforcement and conservation policies of the EU CFP and associated socioeconomic consequences.
Project results:
Before considering the outputs from FISHPOPTRACE in more detail, we first highlight three fundamental aspects of the design of the project: the choice of fish species as targets for testing; second, the rationale for choice of main traceability tools developed and finally, the role that fisheries forensics, the application of science to legal requirements in the fishery management context, has played in helping the consortium to address its objectives.
Why the choice of FISHPOPTRACE target species?
FISHPOPTRACE employed three primary criteria in the choice of target species, relating to conservation status, traceability issues and representation of marine fish life-styles: Atlantic cod (Gadus morhua L.), Atlantic herring (Clupea harengus L.), European hake (Merluccius merluccius L.) and common sole (Solea solea L.). The selected species are all economically important, relatively widespread on a European scale, known to exhibit population structure and fall within EC priority species for enforcement and/or conservation. All target species are vulnerable to a varying degree to overfishing. Finally, the four species have been selected to represent different life-styles ranging from a small pelagic (herring), coastal flatfish (sole), to benthopelagic (cod) and demersal 'deep sea' (hake) species. Such a range in ecology is associated with variability in the extent and pattern of population structuring, thereby providing diverse material as proof-of-concept for other marine fish species. The FISHPOPTRACE target species have also been chosen on the basis of their relative distribution, where cod and herring have a more northerly distribution than sole and hake. For the four species different levels of population genetic information are available, where cod has been studied genetically for decades, while relatively little is known about the population structure of hake. The multispecies approach encompasses different geographical scales for tracing individuals, representing a range of policy-led traceability scenarios.
Choice of primary traceability tools: genetics and otoliths
Several tools are available to understand the extent to which fish populations interbreed and to trace back the geographic origin of landed fish. These include physical external tags, natural tags and genetic markers linked to DNA, the genetic material that makes up genes. However, once a fish enters the food supply chain, several tools become less suitable. Tools for monitoring natural populations and application to fisheries enforcement should therefore meet stringent criteria: they should mirror population identity and stability over an ecological and evolutionary scale. Traceability tools should be available throughout the food supply chain from capture to a customer's plate and should be amenable to forensic validation for use in a court of law. Currently available genetic tools for traceability are of limited utility in an international and forensic context. An emerging class of genetic markers are SNPs. They are very abundant and widespread. Analyses of SNPs reach hitherto unprecedented levels of population identification, rendering them optimal tools in fundamental biology, conservation and traceability. Importantly, data from SNPs are especially amenable to archiving because they exhibit high reproducibility among different laboratories: newly collected data can then be readily compared with reference data with high reliability and ease. In addition, the identification of SNPs is not only responsive to changes unrelated to environmental differences, but also to natural selection, greatly improving the power to detect the distinct signatures of local and regional groupings.
On the other hand, phenotypic characteristics can also be robust indicators of the origin of fish, though they are not available for use throughout the food supply chain. Whole fish can be characterised by their appearance, such as body shape, counts of body parts, shape and composition of otoliths and composition of muscle fat. Since the early 1900s fisheries scientists have made regular collections of commercial fish, to monitor abundance, growth and condition. The otoliths collected for aging have frequently been archived, thereby yielding a valuable legacy to set a historical base line. These calcified structures grow in layers, similar to tree rings, throughout life. Seasonal and yearly increments are deposited, matching the age of the fish. But there is more. The width of each increment represents the growth history of each individual. The chemical composition of each increment reflects the local water mass, forming a permanent record, a kind of logbook, of where a fish was born and where it has lived prior to capture. Chemical methods, such as inductively-coupled mass-spectrometry (ICPMS), measure otolith chemical composition, in order to determine the origin and fishing ground. The approach is particularly interesting as a tool to validate catch records. In this way, fish from different areas can be distinguished from each other.
Otolith shape is another characteristic that distinguishes fish species and populations. Modern imaging techniques are used to photograph and analyse shape differences - such as the pattern of scalloped margins or notches. The example above compares the 'average' otolith shape between different hake populations caught in the North-East Atlantic Ocean and Mediterranean Sea. In combination, otolith shape and chemical composition are valuable tools to trace fish to their spawning and fishing ground. Interestingly it is possible to trace back in time the pattern of fishing and fish populations over past decades in the archived collection of otoliths housed at fisheries institutes world-wide. In many cases, small traces of fish tissue with its DNA remains attached to these otoliths. DNA can then be isolated and characterised genetically, similarly to recently collected tissue - a practice followed in FISHPOPTRACE.
Fisheries forensics: the policy relevance of a forensic framework
Forensics is a field of science dedicated to the methodical gathering and analysis of evidence to establish facts that can be presented in a legal proceeding. Evidence is sometimes required within the fisheries context, when regulations have been breached, such as illegal fishing activity, mislabelling of a fish product, or under-sized fish are captured. While there has been no shortage of policies and maritime laws to assist in the governance of our oceans, it has proven much more difficult to enforce maritime law and to execute prosecutions. In part, such constraints are related to having an appropriate set of tools to yield sufficiently robust evidence in a court of law. As previously stated, the global level of IUU fishing and supply chain fraud offer financial gains on a scale that attracts organised criminal groups as well as unscrupulous individuals; the tools required to detect and tackle such activities and enforce fishing regulations need to be equally sophisticated.
The outputs produced by FISHPOPTRACE help us to understand how fish populations are structured and allows us to look for ways of identifying where fish come from. These techniques can be developed into a series of applied tools for forensic analysis and tracing fish and fish products from ocean to fork. However, before this can be achieved, method validation must be performed to demonstrate that laboratory testing is accurate and reliable. The method validation stage focussed primarily on understanding the strengths and limitations of the DNA and otolith analysis techniques. Scientifically speaking, this involved assessing the level of variation among samples from the same population when tested under different conditions and comparing these data with the level of variation among populations.
The application of FISHPOPTRACE outputs - initial comments
It is also worth pointing out, prior to a more detailed consideration of FISHPOPTRACE results and dissemination, that a parallel aim of the consortium was to identify and promote mechanisms that would engage stakeholders and end-users in fisheries. While starting out as a research project, FISHPOPTRACE set the ambitious goal to move beyond and bridge the often prevailing gap between academic science, policy making and end-users. To this end, a technology transfer strategy was developed which included pro-active engagement with the CFP and communication with stakeholders.
Highlights of FISHPOPTRACE findings and outputs
Having provided some general comments on the overall aim, scope and design of FISHPOPTRACE, here we consider some of the key findings arising from our research, as well as examples of how these have been communicated to a diverse community. Complementary information is also available from the consortium website (http://fishpoptrace.jrc.ec.europa.eu/) which is updated regularly, as well as remaining 'active', with further development, after the conclusion of the project in May 2011. FISHPOPTRACE was organised into a series of work packages (WPs), details of which can be obtained from project documentation and the FISHPOPTRACE website. The WPs, which essentially follow a logical flow from establishing the state of art within the fish traceability field by accessing details of related European project, moves onto sample acquisition, traceability tool development and validation, to their eventual full deployment across the many geographic and spawning populations of our target fish species. The flow of work throughout has utilised core principles of technology transfer, dissemination of activities and outputs and management.
What was the state of art in European marine fish population structure and traceability at the outset of FISHPOPTRACE?
Tremendous progress has been made in the fields of marine fish genetic improvement, genetic stock identification and genomics in recent years. The continuing development of new molecular genetics tools has provided high-resolution markers for assessing genetic population structure, for estimating demographic parameters and for providing insights into fish stock management. However, while part of this information is available in the primary scientific outputs (scientific journals, etc.), there is another kind of information, sometimes scattered, in the way of technical information, collections and databases held by individual experts or institutions, which are still very useful.
Having that in mind, a primary objective of FISHPOPTRACE was to provide a summary review of existing information bases for marine fish genetic resources and to place the FISHPOPTRACE project within the context of recent and on-going fish traceability programmes. To achieve this, three main tasks were fully accomplished during the first year of the project. Firstly, the establishment of a common platform for accessing archived tissues for genetic analyses from past and current EU and national projects. By searching at http://fishpoptrace.jrc.ec.europa.eu/tools/projects it is possible to find the links to other existing projects related to genetic identification and traceability of marine fish species. Secondly, filtering of valuable data produced and stored in other fish traceability projects and compiled into the new FISHPOPTRACE database, the 'meta-crawler', accessible at http://fishpoptrace.jrc.ec.europa.eu/crawler. Finally, the establishment and the distribution of a tissue archive database of external samples that has been extensively used during the project, especially the historical samples. Links to collections of specific external samples with biogeographical information and source and distribution of samples external to the FISHPOPTRACE general sampling scheme can be accessed at http://fishpoptrace.jrc.ec.europa.eu/data-access. The latter resource is an especially distinctive and valuable contribution to the community of scientists interested in traceability and population structure of marine fishes. It provides an integrated platform for accessing 12 EU-related projects associated with fish population structure and/or traceability, bringing together under a single umbrella, the recent collective effort and outputs within the field. Moreover, it aims at addressing the obstacles and missed opportunities arising from the fragmented nature of many tissue archives. Finally, the design of the portal allows for ease of use and access, with minimal prior experience required in using the interfaces, thereby allowing a range of end-users to benefit from its use.
How did we ensure the collection of representative samples of cod, herring, hake and cod from European waters?
The ultimate goal of FISHPOPTRACE was to provide applications in support of fishery management, including control, enforcement and traceability. However to get there, FISHPOPTRACE started out as a research project and asked the question: 'Can we detect distinct populations of cod, hake, herring and sole in European waters?' Due to the apparent openness of the oceans, migration over long distances of many marine species and dispersal of eggs and larvae over vast areas, the notion that marine fish tend to breed randomly across their geographic distribution and do therefore not show population structure prevailed for a long time. FISHPOPTRACE aimed to identify distinct features of target fish populations. These markers should ultimately allow the assignment of fish to their population of origin. Fortunately, in recent years, research has shown that many marine fish are indeed subdivided, occasionally over small spatial scales. This sets a robust rationale for FISHPOPTRACE and our question can be refined to 'can the population structure of cod, hake, herring and sole be determined at a scale that is useful for fishery managers and control authorities'?
To answer this question scientifically, careful planning and consideration was required, starting with the collection of fish at sea before analysis in the laboratory. An elaborate sampling strategy must be developed to maximise the probability that all potential populations of the target species are included across the area under investigation. To this end, the FISHPOPTRACE sampling team designed species-specific sampling schemes covering the European seas, taking into account the species life-history traits, population dynamics and structure and environmental features. Standardisation procedures were developed from on-board species identification and tissue collection to documentation, storing, archiving and cataloguing in laboratories and in a central database system. In addition, sampling progress was monitored online on a publicly available sampling map. A specific achievement of FISHPOPTRACE was the species traceability of the common sole and of its cryptic species solea aegyptiaca. The fish collected during the sampling phase were analysed in the laboratory using state-of-the-art technologies, namely high throughput DNA sequencing, ICPMS and proteomic and gene expression profiling.
How can we develop a common set of traceability tools for identifying fish populations across different spatial scales, as well as tracing diverse natural and processed fish products?
A major objective of the consortium was to develop a set of traceability tools that could be applied comprehensively to key spawning populations of each of our target species. It was also important to develop tools that could be used at different stages of the food supply chain and that could also ideally reveal information about population structure and traceability at different levels. The genetic markers, SNPs and analysis of chemical signatures and shape variation in ear bones (otoliths) were chosen. As already mentioned, SNPs can be retrieved throughout the food supply chain. In addition, for many decades, fisheries and other laboratories have archived otoliths in connection with information on growth rates obtained from examining the otolith growth rings. Through our consortium extensive links, these archived and stored otoliths provided an additional set of samples from which DNA could be obtained, as well using them in the chemical and shape analyses.
Among the most striking scientific results is the provision of several hundred novel genetic markers in, hake, herring and sole. The genetic markers in cod were already largely available from another project ('Canadian Cod Genome' project), with which FISHPOPTRACE had an agreement. Access to new genetic methods, the so-called next generation sequencing, has changed the picture in a matter of just a few years. From a dozen genetic markers a few years ago, we now have knowledge about thousands of small genetic differences (genetic variation) at numerous genes, allowing the design of hundreds to thousands of new genetic markers. The unique combinations of the variation make it feasible to assign the fish to specific populations and in some conditions to identify unique individuals.
For SNP development, we first sequenced a high number of so-called expressed sequence tags (ESTs) from the muscle of sole, hake and herring. After assembly, a large number of SNPs were identified using a range of approaches. We first performed a visual inspection of 1 536 SNPs per species and along a pipe line of detailed analyses followed by validation, and then chose the best panel for traceability and forensic analyses to be used in WPs 5 and 6. A total of 1 200 to 1 300 samples per species were analysed during the validation-genotyping step and initial genetic analyses were performed to define the final analysis panel. It was important to ensure that any SNP markers chosen were easily scored using our high throughput methods, where many hundreds of SNPs can be typed simultaneously, as well as ensuring that they did represent accurately the underlying genetic variation in individuals; it is the cumulative sum of genetic variation across hundreds of SNPs from a population that ultimately determines to what extent we can distinguish it from other populations - a co-called 'population signature'. The following number of reliable SNPs was defined: 426 SNPs for sole, 281 for herring and 395 for hake. In conjunction with the previously-mentioned cod project, we were able to type 1 290 SNPs.
Procedures for the analysis of otolith microchemistry have been standardised and optimised to provide the most robust and clear signal for otolith core information and the narrowest otolith edge signal. Additionally, the methodology for otolith shape analysis was standardised to produce reliable digital photographs. Protocols for image analysis have been developed to automate the processing of the images and the statistical analysis of shape.
What about forensic validation of our primary traceability tools?
The remit for FISHPOPTRACE was to take the research and technological development (RTD) outputs across work packages and evaluate their suitability and readiness for application to fish traceability by performing a series of validation studies. These validation studies would generate the data necessary to assess the robustness, reliability and reproducibility of tools developed within FISHPOPTRACE for monitoring or enforcement purposes. Briefly we wanted to establish a set of criteria for the validation of traceability tools; undertake developmental validation of each identification technique; develop statistical methods for the identification of source populations using multiple markers; implement internal validation studies and inter-laboratory calibration of techniques and collate all validation data to enable the construction of standard operating procedures (SOPs). Work towards these objectives was conducted across multiple techniques and for several different aspects within a single technique. As such, the work conducted was broad in nature and required a high level of integration with other WPs, as well as the direct involvement of multiple partners with diverse areas of expertise.
A large amount of effort was put into the otolith validation work and many advances were made, as the basic technologies used for both the microchemistry and shape analysis were essentially specific to individual laboratories and were not routinely used for the forensic identification of biological material. The largest component of validation work however focussed on the validation of molecular markers. In order to deliver a validated assay, the marker validation work expanded to include a range of tests. In each case, a second laboratory was used to test the analytical protocols and ensure that comparable data were generated using the same method. For otolith analyses, control samples were prepared by using the matching otolith pair to ensure that testing was based on a like-for-like comparison. In addition, the otolith microchemistry analysis uses commercially available elemental calibration standards to ensure that the analytical baseline is identical in each laboratory. For the genetic markers, standard control samples representing each genotype for each SNP marker were prepared. These control samples are available alongside the standard protocols generated for transfer into testing laboratories beyond the lifetime of FISHPOPTRACE. For the otolith microchemistry and morphometric analyses, variation among laboratories was observed, as expected, though the magnitude of the measurement differences was larger than anticipated, particularly for microchemistry. These results were one of the main reasons why the otolith microchemistry analysis method is not yet considered to be suitable for routine use in the production of forensic evidence.
The results of from the forensic studies are documented in a series of validation protocols, reports and analytical methods. As the validation process is designed to be entirely objective, the fact that several techniques were not considered to have performed sufficiently well to pass through to formal application is not seen as a failure; on the contrary, the ability to critically evaluate each technique based on quantitative, comparative data demonstrates the success of outputs in terms of its intended function. In the final analysis, the SNP markers were the only technique demonstrably suitable for application within a forensic framework; the other markers, while often extremely informative, are not yet applied in a format that is likely to withstand the scrutiny of legal challenge.
How informative were the SNP and otolith traceability tools when applied to fish populations and processed products?
It was important for FISHPOPTRACE not only to provide informative data on the structuring and dynamics of main spawning populations of cod, herring, hake and sole, but also to demonstrate the feasibility of obtaining large quantities of robust traceability data in a timely and cost-effective manner. Thus, there was considerable discussion about the most appropriate method for genotyping the SNPs, that is, the routine determination of the genetic variants across many SNP loci in each population. A large number of genotyping methods and platforms were evaluated resulting in the selection of an Illumina 1536 golden gate array; as our general genotyping method when exploring many hundreds of SNP markers simultaneously. This array was used for a single step validation and genotyping approach for putative SNPs and all population samples of the targeted species. Analysis of otolith microchemistry and morphometrics were conducted in parallel for a large subset of the samples used for genetic analysis. Based on the SNP genotyping, genetic maps illustrating genetic relationships among population samples were created.
Separate maps of so-called 'neutral' and loci 'subject to selection' were created as well as maps including all loci. Distinct regional groups were apparent using both classes of markers, where Atlantic and Mediterranean basins were the most distinct. However, also on a smaller geographical scale distinct groups were obvious, such as minor basins within the Mediterranean and the Baltic Sea. On a very fine scale, markers under selection revealed previously unrecognised population structure, highlighting the concept of investigating variation in genes as very powerful for resolving fine scale population structure. The exact number of loci subject to selection varied according to species, geographical region studied and statistical method employed. Overall the outlier loci appeared to be associated with major climatic differences among populations. In total the high number of markers and those with highly elevated levels of genetic differentiation provided unprecedented high power for assignment back to population of origin. At large and medium geographical scales, minimum panels including only a subset of loci could provide close to 100 % correct assignment to population of origin. On a small geographical scale more loci were needed, however in most cases highly informative results could be provided.
The core approach for the genetic component of the work was the development of a 'SNP chip' for each of three species sole, hake and herring. Canadian researchers had already created a SNP chip for cod to assist in aquaculture research. These DNA-covered microchip-like devices enabled testing the identity of 1 536 possible SNPs for each group of individuals from a specific population. Once all individual fish were characterised using the chip, the frequency of each SNP variant was determined, creating a potentially diagnostic pattern, or 'population signature'.
Based on our analyses, as far as the new SNP markers are concerned, it has been possible by varying the numbers used on a SNP-chip, to assign individuals back to their source population across different geographic scales with high levels of certainty and reproducibility. Such outputs are especially significant since previous types of genetic markers either detect levels of population differences that are too low, or there are inherent difficulties in comparing data generated from different laboratories. The use of a marker system such as SNPs, which is essentially based on the presence or absence of large numbers of single genetic variants means that data can be compiled from sources in a much more reliable and high throughput way. The approach thereby enables the generation of baseline and ongoing additions for subsequent genetic monitoring. Moreover it is imperative that any such tools can be used in a legal context, necessitating forensic validation. This has been achieved for SNP markers within FISHPOPTRACE across a range of policy-driven IUU scenarios.
In addition to improved traceability of fish populations, the DNA code of thousands of genes now makes it possible to link the function of genes with the phenotype/external characteristics of each fish. The long-awaited link between natural selection and the genetic blueprint has now become possible in natural marine fish populations. Such an advance is significant because scientists are poised to move beyond the mere detection of genetic differences among marine fish populations, to the identification of how and why such differences relate to their survival in stressful environments, so-called local adaptation. Such information can inform managers on the ability of populations from specific localities to adapt to natural and man-made changes, including over-fishing, contaminants and climate change.
The results of the otolith analysis provided independent insights into the population structure and ecology of the targeted species. The information generally corroborated the genetic results, but also in some cases provided complementary information with higher resolution power, in particular on a small geographical scale.
The tools developed and tested within FISHPOPTRACE provide a giant leap forward in terms of tracing fish and fish products. Here we provide new, more powerful, fast and cost-effective tools as well as the baseline data on a pan-European scale for four economically and culturally highly valuable species. We are now able to trace the fish from the four species from ocean to fork in all types of products. In some areas, however, fish from different populations meet and mix. Establishing mixture signatures for these areas is a high priority. Likewise, although our sampling effort has been substantial there may be populations which have not been fully sampled. However, due to the flexibility of our approach, more data can be added to the database with ease, as there is little need for calibration of SNPs compared to previously applied genetic markers. Finally the implementation of the methods into practical fisheries management and control and enforcement should be a priority for the project members and for the EU.
In summary, we applied our primary traceability tools, i.e. DNA SNP analysis and otolith microchemistry and morphometrics to test their power for tracing fish from the four species back to population/area of origin. Such information provides a framework for a pan-European traceability database for fish and fish products to be applied for defining management areas, fighting IUU and for assuring safe, healthy and sustainable fish products for the European consumer. Our focus has been on the traceability of populations, but each test can also incorporate a species identity marker. The latter is important because over 60 % of fish products consumed within Europe derive from imports outside the EU. Any traceability system that allows simultaneous testing of species and provenance will have significant global application.
How stable are our traceability tools over time?
Any traceability system needs to take into consideration whether the method is resilient to changes over time that may alter the accuracy and precision with which fish and fish products can be traced. We therefore needed to establish the time scale over which the developed tools work, i.e. not just for the populations in specific areas as they are now, but also in the future. It is of interest, therefore, to establish whether or not temporal changes in populations can be assessed using the marker approaches developed in FISHPOPTRACE. Potential changes include long term changes in the sizes of local populations, e.g. in response to global change and exploitation. Another type of change may be geographic, if individual populations change spatial distributions, for example in response to environmental changes to spawning or feeding grounds, or if their relative contribution to mixed-stock fisheries change. All major changes need to be taken into consideration for sustainable management of resources and for conservation of biodiversity. FISHPOPTRACE has examined such issues. The approach is to assess the stability over time (temporal stability) of the developed population markers back in time based on analyses of a series of samples collected on different dates and to model which levels of future change can be assessed using the markers at hand.
Due to the biological differences among both species and populations within species, as well as the management and conservation issue arising, the scopes for monitoring temporal change will vary on a case-by-case basis and will operate on different time scales. We specifically targeted our work to reflect and represent different types of monitoring objectives.
The FISHPOPTRACE traceability tools have also proved highly informative for addressing potential changes in migratory behaviour, both in terms of annual changes in where populations feed and whether fisheries exploit different stocks disproportionately during different times of the year, as well as for assessing the persistence and dynamics, of local populations. For otoliths, the temporal studies indicated overall stability, but also some differences between samples from different dates within locations that should be taken into consideration when developing tools for traceability and determination of management units. The genetic marker analyses in all cases demonstrated large potential for genetic monitoring. Such tests over time for stability coupled with analysis of population structuring demonstrate a set of valuable tools for tracing the locations and movements of individuals in space and time. Analyses can now be targeted to real-time monitoring of seasonal or annual changes in population exploitation rates. The genetic analyses of longer time series conveyed novel information on other aspects of genetic monitoring. Such methodologies can also be targeted to monitoring population changes in adaptive responses to climate change. We also demonstrated the SNP-based method's power for monitoring population origin of individuals in data poor areas. Other types of markers, such as otoliths also exhibited temporal stability although temporal differences were also evident. Such differences underline the care that needs be taken if applying these types of tools in monitoring programmes.
In summary, we show that population-specific traits vary little over time across species, demonstrating that the developed tools are reliable for insights across shorter and longer time scales. The observed temporal stability and conversely lack of short term change in population structure, is an important step in validating the traceability approach as reliable in time and space. Needless to say, future changes in population structure and the spatial distribution of individual management units are difficult to predict. Nonetheless, FISHPOPTRACE has laid the basis for a monitoring framework that can be developed and extended in years to come.
In addition to SNPs and otoliths what other traceability tools might be developed for future application?
Besides developing applications based on genetics and otolith analysis for traceability of fish and fish products, FISHPOPTRACE also explored the value of 'novel tools'. Novel tools are new ways to analyse fish products that could complement more established methods and help to trace individual fish to the place where they have been fished. To this end, two major goals were set. The first one was to develop a DNA chip, a miniaturised tool with improved sensitivity, speed and costs, based on a sophisticated technology that uses a special glass slide with DNA 'sensors' spotted onto it. A pilot study on cod individuals selected from different geographic regions showed that this new technology is very promising but requires further optimisation. Two different versions of a prototype platform for SNP genotyping were developed and tested on a test panel of cod SNPs and samples. Both are based on a so-called 'padlock probe technology' and a second version, with a smaller number of targeted SNPs, showed very promising results, opening the possibility of routine genotyping of an informative SNP panel (10-100 loci) at a reduced cost.
The second goal was to find additional traceability markers. The idea is that environmental differences, such as water temperature, or the nutrients taken up by fish, might influence specific features locally, making it possible to trace where the animal was living. Firstly, we looked at the composition in fatty acids (FA) of fish flesh. FA analysis was performed on all four target species (334, 292, 132, 98 samples respectively for sole, hake, cod and herring), exploring spatial and temporal variation, demonstrating that geographic populations can be effectively differentiated according to FA composition. Statistical analysis of the FA composition in the tissues showed a clear differentiation of geographical populations, both within the Atlantic or the Mediterranean as well as between these two basins. Secondly, we examined which genes are switched on or off, that is expressed (active) or not, in the muscle tissue of hake and sole living in different areas. A DNA microarray was developed to analyse 14 898 genes in the muscle of hake collected from six different areas. We demonstrated that for certain areas it might be possible to trace back fish to their geographic origin measuring the expression of few marker genes. Thirdly, we looked at protein expression. Recent advances in molecular biology have greatly facilitated the development of powerful tools for proteomic analysis. A proteome scan approach based on two-dimensional fluorescence difference gel electrophoresis (DIGE) technology generated thousands of protein markers that allow the identification of different hake populations. Among the identified proteins, some that reliably assigned individuals to their population of origin have been selected to be used as biomarkers for traceability of hake fish products. Currently an antibody array specific for hake protein markers is under development, to make the assignment of individuals to the original stock feasible as a diagnostic routine.
Overall, our quest to test the potential of additional traceability tools yielded very exciting results and opened the way for the use of novel technologies to complement more established tools (genetic markers, otoliths shape and composition) in traceability studies. It should be noted, however, that this was an exploratory exercise in FISHPOPTRACE and these novel tools need to be validated fully before being used for control and enforcement purposes. Importantly, any traceability system needs to have sufficient flexibility such that constraints imposed by the specific requirements of any specific technique can be matched with available resources and opportunities. Additionally, where required, information on spawning population as revealed by genetic tools, might be usefully complemented by knowing what water body the fish was captured from, such as otolith microchemistry, lipids or gene expression. The genetic data typically generates population-level signatures across a longer-term evolutionary scale based on the average breeding behaviour of fish populations across generations, whereas the other marker types have potential to reveal provenance across a shorter-term, ecological time-scale.
Once all the traceability results have been generated, how can these be archived, retrieved and made accessible to others?
While large collections of information are created during the sampling and analysis phases of projects such as FISHPOPTRACE, usually only a small subset of this information is published as primary scientific outputs. The body of data generated during the lifetime of the project remains vastly inaccessible, as collections and databases are maintained by individual experts or institutions. Such data sources may have a high potential value for future questions, but might get lost or forgotten as new projects begin or objectives change. To avoid such a scenario, a flexible database structure was designed and implemented for FISHPOPTRACE. The database is housed at the JRC of the EC and was designed to grant access to data and further addition of data beyond the project duration.
The FISHPOPTRACE website was constructed and was operational from the early stages of the project acting as a source of information about the project and similar projects. A member section of the website was constructed as a file sharing resource. This store has accumulated a wealth of project derived data, articles, presentations, reports and related documents over the project lifetime.
The FISHPOPTRACE database was designed to be as data independent and flexible as possible through strong collaboration from all consortium partners, including a database-specific meeting hosted by the JRC. Following its construction it has been hosted and maintained by the JRC, including backups, versioning and modifications/bug fixes. An Excel template was agreed on and used for uploading the characteristics of collected specimens to the database. A web-based interface, deployed in the website, was developed to allow consortium members to query and view the contents of the database. Additional functionality was added to allow the uploading of samples to the database through the interface, using the Excel template. Furthermore, the results of user-specified queries through the interface can be downloaded and saved in the Excel template format. These Excel sheets can be edited and uploaded again through the web interface. The interface automatically handles updates to samples, deletion of samples and the insertion of new samples. A crawler tool was developed as part of our initiative to act as a data hub. This crawler automatically searches for news and articles related to FISHPOPTRACE by matching a set of key words. The results are collated and updated daily on our website. Information stored in the database structures can be accessed in two ways through the FISHPOPTRACE website, the information dissemination portal of the consortium. Selected genetic characteristic information can be publicly accessed through a map-based interface, designed to highlight biological characteristics of the target species in relation to their environment in a geographic context. Members of the consortium can also login and query the database, view the information online or download the information in a spreadsheet to apply changes 'at home' before uploading the edited data sets.
The purpose of the geographic visualisation platform is to disseminate information, also to non-experts, in a transparent and easily understood way for use as a management and regulation decision support tool. Data are provided in a geographic and environmental context. Aspects such as ocean currents, surface temperature and salinity value variations and their potential relationship with the development of features, such as genetic differences in the target species, can be visualised. This provides valuable support to research, but constitutes also a useful communication tool for population/stock structure information to stakeholders involved in fishery management and policy development.
Potential impact:
FISHPOPTRACE is designed to integrate information on the distribution, extent and dynamics of fish population structuring in four target species, as a basis for conservation in relation to sustainable utilisation of bioresources and in the provision of population markers for traceability of fish and fish products. As previously mentioned, a key strategy of FISHPOPTRACE was to bridge the often prevailing gap between academic science, policy making and end-users. To this end a technology transfer strategy was developed, which consisted of intensive and targeted networking and communication with policy institutions, authorities and stakeholders, engagement with the CFP and a cost benefit analysis (CBA). Here we present salient highlights, rather than an exhaustive account, of how we aimed to facilitate uptake of outputs by end-users.
FISHPOPTRACE website, documents, publications, consultations etc.
The approach followed by FISHPOPTRACE has generated a paradigm of a holistic framework for fisheries control and enforcement. Peer reviewed scientific publications have already and will continue to ensure that the results of FISHPOPTRACE are scrutinised and disseminated in the research area. Such dissemination of outputs, however, is unlikely to be sufficient to reach out to end-users, which is why FISHPOPTRACE stressed the importance of inter-laboratory trials of the protocols developed and used by the project. The consortium also stressed the central role of the FISHPOPTRACE database and interface. The experimental FISHPOPTRACE data has been validated by the consortium members and submitted to the JRC, which acts as the host of the databases. The consortium spent considerable effort on finding ways how best to use the FISHPOPTRACE website to generate maximum benefit for stakeholders and end-users. This is well reflected in the development of the web-based project crawler tool developed in the early stages of the project. Ultimately it was agreed that a geo-visualisation platform would generate a practical opportunity to support the combination of experimental data and results with environmental parameters, in a so-called 'seascape genetics' approach, on one side and to demonstrate experimental outcomes and their value for applications to stakeholders and end-users. Such environmental parameters will be enlarged progressively in collaboration with the JRC institute for environmental sustainability (IES). Importantly it is also possible to visualize and overlay fisheries management sections such as the International Council for the Exploration of the Sea (ICES) or the Food and Agriculture Organisation (FAO) subdivisions. Thus, the findings from FISHPOPTRACE open the door to various valuable applications for fisheries management, especially when displayed and accessible in the context of geographic information systems (GIS).
A key component of technology transfer is to ensure that strictly designed protocols of traceability can be generated for practical uptake; the so-called SOPs. SOPs are documents containing instructions that forensic scientists and laboratory staff follow to perform procedures that are routine, standardised and for which no ad hoc modification is acceptable. They are indispensible when supporting crime investigations scientifically as they help to ensure the quality and integrity of data and provide a basis for guidance, uniformity and accountability. Within the project SOPs for selected analytical scenarios have been developed.
A significant dissemination opportunity was provided by the EC Directorate General (DG) RTD as they invited the FISHPOPTRACE consortium to join the organising steering committee for the food chain integrity conference, a major dissemination event on traceability to be held in September 2011 in Brussels; a book chapter has resulted from the original invitation to participate, to be presented officially during the conference.
A notable highlight in terms of international exposure of FISHPOPTRACE was the publication of a two-page news focus article in Science, in December 2010, entitled: 'To fight illegal fishing, forensic DNA gets local', in which Erik Stokstad, environment and sustainability editor for Science, described how FISHPOPTRACE had produced a new generation of genetic tests that could provide authorities a much better idea of exactly where fish have been caught. Profiling of FISHPOPTRACE in Science not only acknowledges the state-of-art approach employed by the consortium, but importantly also provided detailed examples of how the SNP-based tools in particular had been employed to tackle various policy-led case European traceability case studies for each of the target fish species.
An amendment to the FISHPOPTRACE technical annex, associated with the granted three month no-cost extension, was the production of a stakeholder pamphlet, published in May 2011 in conjunction with the Brussels stakeholder workshop. The remit of the brochure was to present the context, key findings and policy relevance of FISHPOPTRACE activities and outputs, presented in an attractive, accessible and concise format. In addition to a contribution from Mr Timothy Hall, Head of unit 'E4: Agriculture, forests, fisheries and aquaculture DG RTD', the brochure presented details of consortium membership, the role of traceability and population structure in sustainable management of fish resources, the challenges of IUU and fish fraud, key methods and findings from the project, as well as the extent of exposure of the project in the media. The brochure concluded with a consideration of mechanisms and opportunities for securing a legacy to FISHPOPTRACE and a call for need to extend the impact of FP7 science projects beyond the natural life of the funded period. One-thousand copies of the brochure were produced and distributed throughout the consortium.
Meetings, conferences etc.
A major dissemination opportunity for FISHPOPTRACE was the participation of various partners and co-organisation of a major international symposium held in Seattle, United States of America (USA), in March 2010, 'SNP Symposium III: Applications of SNP genotyping in non-model organisms'. Members of FISHPOPTRACE also presented the application of forensics at the sea food summit (Paris, 2009) and the global fishery enforcement training workshop Maputo; 2011) organised by the international monitoring, control and surveillance network (iMCS), which is endorsed by the FAO. Through our consortium member, Robin Waples, FISHPOPTRACE has built a link to the USA federal agency National Oceanic and Atmospheric Administration (NOAA), a key scientific advisory body for fishery management, including control and enforcement. Several members of the FISHPOPTRACE consortium are also members of the ICES working group on applied genetics for fisheries and mariculture (WGAGFM), thereby building a permanent link between FISHPOPTRACE and ICES. Furthermore, primarily through the JRC, contributions from Jann Martinsohn have promoted ways to better include modern molecular technologies in fishery management decisions under the CFP remit. Particularly noteworthy is also that FISHPOPTRACE has been invited in April 2011 to present project and results at the plenary session of the Scientific, Technical, Economic Committee for Fisheries (STECF), which resulted in a highly positive assessment of the potential of genetics for fisheries management in the plenary report which will be submitted to the EU member states.
Martin Taylor and Gary Carvalho organised a DNA barcoding workshop at the University of Penang, Malaysia, in November 2009, with a focus on the application of barcoding technologies to fish biology and fisheries management. A key application of the barcoding pipeline was traceability in fisheries forensics and there was also the opportunity to present aspects of the FISHPOPTRACE project to workshop participants, as well as seminars in the Fisheries Research Institute, Department of Fisheries, Malaysia, the University of Malaya in Kuala Lumpur and the Centre of Research for Computational Science and Informatics in Biology, based at the University of Malaya and the Malaysian Fisheries Department in Penang. The format of the workshop included seminars and hands-on experience of the BOLD database, which also forms a component of the FISHPOPTRACE traceability framework.
A meeting was held at the FAO, Rome in December 2009 to discuss the use of forensic approaches in the context of MCS and IUU fishing. An invitation was received from Michelle Kuruc, who attended the FISHPOPTRACE Madeira workshop, for four members of FISHPOPTRACE to participate: Jann Martinsohn, Rob Ogden, Einar Nielsen and Gary Carvalho, as participants. The aim of the workshop was to identify the most appropriate traceability tools for use within a forensic framework and to generate a set of recommendations for taking the approach forward within a global context. The workshop included broad representation from a variety of nations and backgrounds, including forensic biologists, fisheries compliance officers, directorates for fish inspection quality control and various research organisations and universities with experience in the development and application of advanced tools, as well as their implementation to fisheries. The workshop provided a key opportunity for profiling the objectives and progress of FISHPOPTRACE, which was deemed a significant international effort in the quest to tackle IUU and promote consumer protection. An output from the workshop was to provide support for the establishment of an FAO expert group, in fisheries forensics.
Other modes of dissemination
In addition to extended interest from the popular press and media (including 'Science', 'New scientist' and 'the Economist'), FISHPOPTRACE has enjoyed enhanced profiling from within EC instruments. For example, FISHPOPTRACE was profiled on the CORDIS technology marketplace, as an example of cutting-edge technologies and applications to environmental management, 'So our fish do not disappear without trace', as well as being chosen as a 'success story' for the EC's CORDIS website, project success stories - A sustainable future for fishing and for inclusion in association with Euresin (ygaethofs@euresin.eu via e-mail), a new, on-line database that will cover all life science projects funded under the knowledge based bio-economy (KBBE) programme in FP7.
Bangor University submitted a Press Release on FISHPOPTRACE in the early spring which resulted in an article in Fishing News (see http://www.fishingnews.co.uk online), entitled 'Protecting the fishing industry: New £3.6m programme to safeguard fish stocks for industry and consumers'. The press release also stimulated a response from the United Kingdom (UK) body charged with the management and conservation of marine resources, the Department for Food and Rural Affairs (DEFRA): Tim Bostock, international fisheries policy adviser to DEFRA's sustainable fisheries team emphasises that the UK has taken an international lead in the fight against illegal fishing. 'As part of our work under the ministerial high seas task force, we have been working with like-minded countries in developing approaches intended to deter illegal operations by hitting them where it hurts- in their wallets. Forensic tools such as those being developed (in FISHPOPTRACE) will provide a powerful deterrent and help protect those who wish to operate legitimately. Tackling illegal fishing is an important facet of more effective fishery management and the monitoring, control and surveillance operations. A successful completion of this project will certainly be welcomed by agencies charged with the enforcement of conservation management' he said.
In addition, articles appeared in the local press and the Coordinator, Gary Carvalho, was interviewed about the project on the British Broadcasting Corporation (BBC) Radio Wales, for the science programme, 'Science Café'. The project has been publicised on the Bangor University homepage under 'Latest News'. FISHPOPTRACE also enjoyed additional exposure on the UK Natural Environmental Research Council, Planet Earth on-line, which profiles via podcasts, timely and significant studies in environmental sciences. The podcast, Cracking down on illegal fishing, presented an interview with Sarah Helyar and Gary Carvalho from Bangor University, in which they described the context and impact of IUU fishing at European and global levels and the use of molecular approaches, including those developed within FISHPOPTRACE, in tracing fish products for monitoring, control and enforcement.
FISHPOPTRACE fully acknowledged that in order to transfer results and tools to stakeholders beyond the scientific realm, major efforts have to be put into achieving maximal visibility, while not overselling potential benefits (impartial dissemination). Obviously the publication of articles in peer reviewed journals had to remain one major dissemination component. Dissemination of results has also been achieved through pro-active engagement with stakeholders during the reporting period. This includes invitation of various external 'observing' scientists such as Dr Petra Spaniol (DG Fisheries) to our kick-off meeting in April 2008; Dr M Kuruk, FAO, Fisheries Forensics in Monitoring, Control and Surveillance (workshop two); Dr Jan van Aken, EU Marine Programme, Pew Foundation (workshop two); Dr Harald Barne, DNV Research and Innovation / Biorisk Management (workshop two); Dr. K. Glover (Institute for Marine Research; Norway), a renowned fisheries geneticists working, among others, in the field of traceability of fish farm escapees and monitoring of genetic impacts on native fish populations, to the 5th FISHPOPTRACE consortium meeting.
Project website: http://fishpoptrace.jrc.ec.europa.eu/
The contact details of all partners are available at http://fishpoptrace.jrc.ec.europa.eu/contacts