|
||
In May 2003, the first Canadian-born beef cow tested positive for bovine spongiform encephalopathy (BSE), commonly known as mad cow disease. Eating diseased cow products is linked with variant Creutzfeldt-Jakob disease (vCJD) – a rare but fatal disease in humans. In both BSE and vCJD, normal proteins that are found on the surface of neurones start to change their shape so they are no longer functional (misfold). These abnormal infectious little proteins, called prions, cause misfolding in other proteins and then they begin to gather together. These aggregations eventually create a type of fibre that’s commonly seen in other neurological diseases such as Alzheimer’s, Huntington’s and Parkinson’s. Sometime between the first prions getting together and the formation of the fibre, the original neurone is killed. As the neurones die, holes form in the brain of the infected animal and ultimately lead to its death. Unlike bacteria and viruses, prions can’t be broken down with normal cooking. In 1986, when England made its first diagnosis of BSE, five million cattle were destroyed to stop the spread of BSE, but not before one million untested cattle made it into the food supply. The first person to develop symptoms of vCJD became ill in the UK in 1994; there is a long incubation time while the misfolded prions reach any symptomatic critical mass. Deaths peaked in 2000 when 28 people died of the disease. There have been around 220 cases of vCJD worldwide to date, but it is estimated that there are approximately 15,000 citizens in the UK who are potentially incubating the disease, so this story is not yet over.
What does mad cow disease have to do with food fraud, and why are we talking about beef at the beginning of a chapter on fish, you ask? In 2003, when that single Canadian-born cow was found to have BSE, a fish processor in Ucluelet, British Columbia (BC) was forced to shut down early in the season. The processing plant turned hake into surimi – the gelatinous paste that is flavoured, coloured and shaped into flakes, sticks and other wonderful shapes and used in fish cakes, fish burgers, fish balls and the imitation crab meat so commonly found in California rolls. Surimi is a clever way of converting a cheaper white fish into likenesses of higher-end products, such as crab, eel and lobster. As part of the process, beef blood plasma – the clear part of the blood that is left over when you remove all the cells and platelets – is added to the fish paste to help it form a gel (these days egg and something called transglutaminase, which we discuss in the next chapter, are more commonly used). Blood plasma products are widely used as gel enhancers in the food industry. The BC plant lost business because nobody wanted a beef product that was made in Canada, despite the fact that the Canadian processors were obtaining their beef plasma from the mad-cow-free US.
Just to be clear, we are not saying that this was in any way a form of food fraud. It was probably written right on the label, most likely as bovine blood plasma or perhaps fibrinogen or thrombin. It is, however, an example of how our food is crossing taxonomic boundaries – horse in our beef, beef in our fish and so on. This creates complications (and opportunities) in terms of food authenticity testing, but it also means that when a food safety issue such as BSE rears its ugly head, understanding its reach within our food supply is daunting even in a world where the food we buy is labelled correctly. It gives one little hope that we can track it down in a world of substitutions and scams – two things very familiar when it comes to seafood.
Like any other food, seafood – fin fish and shellfish – has been prone to fraud throughout history. Recall the painted gills in the night markets of London in Chapter 1? The most commonly talked about fraud within the industry today, however, is mislabelling – mostly in terms of the species name and/or where it was caught. Owing to a combination of factors, seafood, particularly fish, is more prone to mislabelling than any other protein source in our food supply and there are a number of reasons why this might be the case.
First, our desire for fish products is on the rise. Global fish consumption has been growing at a rate of 3.6 per cent per year since 1961 – faster than can be accounted for through population expansion alone. Global per capita fish consumption has grown from 9.9kg (21.8lb) per person in the 1960s to 19.2kg (42.3lb) per person in 2012. Campaigns on the health benefits of eating fish have been widespread, leading many consumers to incorporate more seafood-based protein into their diet. And with more sophisticated processing, better storage techniques and endless capacity to move food around the planet quickly, fish and other seafood products are now accessible to regions that would not have had easy access to these products historically. Fort McMurray, Canada, for example, is over 1,600 kilometres (1,000 miles) by road from the nearest ocean, yet the town of 77,000 people supports at least four sushi restaurants.
Second, there are supply challenges inherent to the industry. Fisheries represent the last remaining commercial-scale harvest of animals from the wild (though the balance is shifting now that half of global production is from aquaculture). In the past decade we have consistently captured about 90 million tonnes of fish from the ocean and inland waterways each year. Management of this resource is rather difficult though, as it’s really a matter of trying to count animals we can’t see in an environment we can’t control. As a result, many fish stocks have been depleted, mostly because (to use the words of Professor Daniel Pauly in the film End of the Line) ‘we ate them’.
When fisheries’ stocks start to dwindle, managers are obliged to implement strict harvesting restrictions in order to help the stocks recover. This, by default, creates a premium product as scarce species are generally worth more on the market. In the 1980s, US fisheries managers put in place a number of regulations to help recover stocks of red snapper, which the US FDA has stated is the legally accepted common name for the species Lutjanus campechanus. Among other restrictions, there are limits on the number of fish that can be caught each year, so those that are caught sell for a good price. In 2011, red snapper averaged a value of about US$7.04 (£4.50) per kg (US$3.20 per lb). That same year, Labrador or Acadian redfish, Sebastes fasciatus, which has been substituted for red snapper, was worth a measly US$0.56 (£0.36) per kg (US$0.25 per lb). It is difficult, if not impossible, to differentiate the two species once they are fillets and the economic incentives to mislabel are substantial.
As mentioned already in Chapter 1, climate change is unlikely to improve the reliability of fish stocks. The range of some species may shift. Some species may thrive while others fail. There have also been predictions that in the future severe storms may become more frequent. Fishermen already contend with tides and weather, so frequent storms and big waves could be an additional challenge the industry faces in the future.
Perhaps more than anything else, however, it is globalisation of the industry that has made it particularly vulnerable to fraud. The seafood trade expanded globally in the early 1990s and the US’s National Seafood Inspection Laboratory (NSIL) began routinely examining seafood products. They reported that between 1988 and 1997, their tests confirmed that 37 per cent of fish species and 13 per cent of other seafoods were mislabelled; as many as 80 per cent of red snappers tested were mislabelled.1 This was some of the first evidence of the prevalence of mislabelling.
In 2013, the US imported just under 2.5 million tonnes of edible fishery products worth approximately US$18 billion (£11.3 billion). This is nearly double what the US imported 20 years earlier, in 1993 (1.3 million tonnes). The UK, with about 20 per cent of the US’s population, imported approximately 739,000 tonnes, valued at £2.6 billion (US$3.9 billion). While importing such vast quantities of fish products helps meet demand and introduces new and exciting products into the market, it also creates opportunities for labelling mistakes and intentional fraud.
As we have seen over and over again, vulnerability to fraud is introduced with every link in the food chain; fish caught and sold in one place are being shipped around the globe for processing, introducing some very distant links. Fish and seafood caught in Alaska, for example, are sent to China for processing at one-fifth to one-tenth the cost of processing in the US. The fish are filleted and the crabs are de-shelled, then they’re packaged and given a ‘Product of China’ label before being shipped back for the US market. The Codex Alimentarius states that if a food has been processed in a way that changes the nature of the product, the origin of that food becomes the country in which it was processed. Russian sockeye salmon that are processed in BC, Canada, become a product of Canada and are labelled as BC salmon. It’s perfectly legal, but very misleading. About 90 per cent of the 104 million kilograms (230 million pounds) of squid caught in California each year is sent to China for processing before ending up back in the US for sale – a 19,000-kilometre (12,000-mile) round trip. Despite fuel costs, it is still cheaper to have seafood processed overseas, and cost is critical for an industry trying to keep afloat among cheaper imported and farmed seafood products.
Fish are so well travelled, in fact, that they are one of the most traded food commodities in the world. And, as anyone who has travelled knows, there are challenges associated with operating in different countries among unique cultures, foreign languages and unfamiliar regulations. Then add to this the complication of common names. Many species are given common names to try and simplify things – it’s undeniable that clownfish is far more memorable than Amphiprion ocellaris. However, a single species can accumulate many names over time, which can complicate matters. Let’s take Atlantic cod, species name Gadus morhua, as an example. It has nearly 200 known names. In the English language alone this species has at least 58 names, 56 of which are used in Canada. We couldn’t possibly provide all the names, but some of our favourites include bastard, blackberry fish, duffy, foxy, tom-cod, grog fish, hen, loader, old soaker, pea, snubby, split and swallow tail. It would be equivalent to you showing up at border control with 200 passports each bearing a different name. A private interrogation room would probably figure in your near future.
Some of the name confusion is generated by marketing campaigns aimed at making species names somewhat more palatable, or at least marketable. For example, Marks & Spencer received permission to rename Glyptocephalus cynoglossus from witch flounder to Torbay sole (a nice local-sounding name). Between 1973 and 1981 the US National Marine Fisheries Service (NMFS) spent half a million US dollars (£315,000) looking at underutilised species to determine which ones could benefit from a name change. It was during this period that slimeheads (Hoplostethus atlanticus) became orange roughy.
To try and overcome the challenges of multiple names, many countries have issued lists that provide the legally accepted market names for each species sold in that country. Canada’s fish list states that out of the 56 English names for cod used in Canada, only cod and Atlantic cod are acceptable market names. But differences are bound to exist between countries. Basa, for example, can refer to the catfish species Pangasius bocourti or Pangasius hypophthalmus in Canada, while the US FDA states that this common name can only refer to P. bocourti. In the UK, basa can refer to any of the 21 species within the genus Pangasius. There are over 60 species of fish on the FDA’s list that can be sold as grouper, while the UK simply states that any fish from the two genera Epinephelus and Mycteroperca can be labelled as grouper – a combined total of over 100 species.
Consumer demand for more non-fuss meals also encourages opportunities for mislabelling. Many of us don’t want to come home from work and start filleting fish, cleaning squid or digging crab meat out of the shell before we cook our evening meal. In general, consumers are looking for prepared portions of fish that we can cook up in a matter of minutes. In response, supermarkets place such specific requirements in terms of how the fish looks and what the portions weigh that they need to be hand processed. This is a round-trip ticket to Asia. And it is not only consumers that are looking for a prepared product – restaurants are too. Many regions have very rightly established regulations to minimise food waste and recycle it in composting facilities rather than having it end up in landfill. This often includes a levy or fee based on the weight of food waste produced by the establishment each week. It therefore doesn’t pay to bring in whole seafood products that kitchen staff have to take the time to prepare, and that also contribute significantly to their weekly waste. Commercial kitchens with savvy chefs are already doing what they can to reduce food waste as every scrap in the bin is money lost. Buying pre-prepared fillets and portions of fish is simply an extension of this efficiency. It has meant, unfortunately, that many chefs are losing some of the most basic skills of preparing seafood for cooking and that the entrees may be better travelled than many of the restaurant’s customers.
Should we then be surprised that as portions of white, boneless, skinless, nondescript fish travel around the world, their names (of which they have many) and where they were caught have a tendency to change, whether accidentally or intentionally? Perhaps not, but how often it happens may be surprising. And with heads, skin and most other recognisable features removed, we need to turn to DNA-based methods to determine whether the cod goes by any other name.
Dealing with diversity
Before we get into the nitty-gritty of DNA analyses, however, we feel obliged to lay out just how monumental a task it is to identify the species and origins of fish. There are over 32,000 known species of fish. They are, by far, the most diverse group of vertebrates on the planet, making up more than half of the 62,000 (give or take) identified vertebrate species. Estimates available on the number of species used for food globally range from 9002 to 20,000.3 The FDA’s seafood list contains over 1,800 records of seafood species encountered in US markets. This would include species that don’t have a targeted fishery, but are caught as by-catch and then sold. The point is, there are a lot of species.
Compare the more than 1,800 fish and seafood species listed in the US with the number of species of meat sold in the US – namely, beef, pork, chicken, turkey and, to a much lesser extent, lamb, mutton and goat. There are about 13 breeds of cattle reared for meat in the US and they have all been bred from one of three species: Bos taurus, B. indicus and bison or hybrids thereof. There are hundreds of breeds of pig, but all of them come from just one species – the Eurasian wild boar, Sus scrofa. There are 34 different breeds of chicken reared for meat, all of which belong to the single species Gallus gallus domesticus. There are eight different varieties of turkey recognised by the American Poultry Association, all belonging to the domesticated turkey species Meleagris gallopavo. When it comes to terrestrial protein, we’re only really dealing with a handful of species and this helps simplify testing.
To try and differentiate the thousands of fish species out there, protein-based techniques have traditionally been used. Each species has a unique protein profile and unknown species can be identified by comparing their profile with known species samples. Isoelectric focusing is the technique frequently used to do this. The principle is that a molecule, such as a protein, will lose its electric charge at a particular pH. This is known as the molecule’s isoelectric point and it can be used to separate out different proteins. An unknown sample is prepared by mixing tissue from the fresh or frozen fish with water. The water-soluble proteins dissolve out into the liquid and then it’s centrifuged to get rid of any bits that didn’t dissolve. The sample is added to one end of a gel (think very thick jelly in a thin layer on a plate) that has a pH gradient from one end to the other. A current is applied, which causes the charged proteins to travel through the gel until they reach a pH where they are no longer charged. Each type of protein will stop at a unique point, creating very distinct band patterns that are relatively unique to the species and can then be compared with a known sample for identification.
It’s not unlike taking an unknown, unlabelled bag of Lego pieces and running them through a series of sieves to separate the pieces by shape. You would then be able to compare what you’ve got with known Lego kits and could say that the unlabelled bag must be kit no. 31002, the Super Racer, because it contains 29 different types of Lego pieces, four single white pieces, two double yellow pieces, and so on.
Barcoding life
However, just as Lego builders are thwarted by the common vacuum, protein analysis is thwarted by the common oven. While protein-based techniques are quick, simple and cost effective, their utility in the food industry is somewhat limited as proteins are denatured during cooking and many other standard processing methods. They’re also limited because proteins are expressed differently in different tissues; the protein profile produced from a sample of Pacific halibut skin, for example, may not be the same as that produced from a sample of its muscle tissue. Some protein-based methods may have difficulty differentiating between closely related species as well. It is for these reasons that analysts are going straight to the source, to the blueprint that codes the proteins and, indeed, life itself – DNA. It is not as easily degraded by heat and it’s found in almost all cells of living organisms (an exception being red blood cells). DNA-based methods have also become far more affordable in recent years.
Many people within the scientific community point to a brief communication in 2004 in the journal Nature as the first use of DNA-based methods to reveal the prevalence of fish mislabelling. Peter Marko, then at the University of North Carolina and now at the University of Hawaii, and his team used DNA analysis and found three-quarters of fish being sold in the US as red snapper were species other than Lutjanus campechanus.4 Marko and his team analysed fish purchased from nine vendors across eight states and compared DNA sequences from these samples with those found in the open access sequence database GenBank. They found that 77 per cent were other species. With the margin of error associated with the technique, anywhere from 60 to 94 per cent of the samples could have been mislabelled. Five of the mislabelled species were other forms of Atlantic snapper and two of the samples were crimson snappers, which are from the Indo-West Pacific. A number of species couldn’t be identified because they were either from other regions of the world or were too rare to be included in the GenBank database.
It is possible that the mislabelling of the other Atlantic snapper species could have been genuine misidentification on the boat as these species are likely to have been harvested at the same time. This, unfortunately, can lead to inaccurate catch statistics – overestimating the harvest (and by default the abundance) of red snapper, while underestimating the misidentified species. Those species caught from other areas of the world, however, are likely to have been mislabelled at some point after they came off the boat because it seems unlikely that an Indonesian fisherman would misidentify a crimson snapper as an Atlantic species. Regardless of where along the supply chain the mislabelling happened, Marko’s conclusion was that this rampant mislabelling was distorting consumer perceptions about the availability of red snapper.
The year before Marko published his paper on red snapper, Professor Paul Hebert from the University of Guelph, Canada, was proposing DNA barcoding as a method of identifying species. You will recall this method from our honey example in Chapter 2. However, unlike for plants, the DNA segment Hebert and his colleagues at Guelph targeted for animal barcodes was about 650 base pairs long and codes the protein cytochrome c oxidase subunit 1 – more easily referred to as CO1. It is one of the subunits that make up the enzyme cytochrome c oxidase, which is found in the powerhouse of the cell – the mitochondrion. It plays a critical role in the cell’s energy production.
Targeting mitochondrial DNA has several advantages. First, there is more material to work with as mitochondrial DNA is far more abundant than nuclear DNA. Most cells have only a single nucleus, but they may have hundreds of mitochondria. In fact, a human liver cell may have as many as 2,000 mitochondria. In terms of sampling, this means that it’s easier to recover DNA from degraded material. Second, mitochondrial DNA has a higher mutation rate compared with nuclear DNA, which means there is a better chance of finding differences between species. Third, mitochondrial DNA is only inherited from one parent, which simplifies matters in terms of sequencing the DNA. As mammals, we have each received equal contributions of nuclear DNA from our mother’s egg and from our father’s sperm. We therefore have two complete sets of chromosomes – two different versions of a gene – which makes us diploid. However, the mitochondria that enter the egg from the sperm are destroyed early in development, leaving us with only the mitochondria we inherited from our mothers and therefore only one copy of mitochondrial DNA (haploid).
Hebert and his colleagues put this segment of DNA to the test and examined the CO1 sequences in GenBank of more than 26,000 animals from 11 broad taxonomic groupings (e.g. worms, crustaceans, beetles, flies, wasps and bees, butterflies, chordates, jellyfish and molluscs).5 They found that with the exception of the jellyfish and corals (cnidarians), they could clearly tell species apart. Cnidarians, like plants, seem to have a slow rate of evolution in mitochondrial DNA and therefore need to be coded using a different segment of DNA. Most relevant to this chapter, however, is that it was very useful in identifying fish and other seafood products.
To ensure a comprehensive reference database existed for fishes, Robert Hanner, Associate Professor at the University of Guelph and Associate Director for the Canadian Barcode of Life Network, began the Fish Barcode of Life (FISH-BOL) initiative in 2005. It was one of the first taxonomically focused barcode campaigns and there has been a global effort to contribute barcodes for expert-identified reference species from around the world. At the time of writing this book, more than 10,700 species of fish have been barcoded.
Taking DNA barcoding to the market
A couple of years into building the FISH-BOL database, Hanner and his graduate student, Eugene Wong, wanted to determine whether they had gathered enough reference samples to allow them to identify unknown samples down to the species level. So off to the markets they went. They analysed 91 samples of fish and seafood collected from commercial markets and restaurants in Canada and the US.6 They found two things. The first was that the database was mature enough to allow them to identify the samples to species level. The second was that 23 of the sequenced samples were mislabelled in some way. Like Marko and others, Wong and Hanner found that red snapper was the most commonly mislabelled fish in the study – seven of the nine samples collected in New York were species other than L. campechanus. But they found that mislabelling extended well beyond this one high-value species. Atlantic halibut was being sold as Pacific halibut, Mozambique tilapia (Oreochromis mossambicus) was sold as albacore or white tuna (Thunnus alalunga), capelin roe (Mallotus villosus) was sold as Tobiko/flying fish roe (Cheilopogon agoo) and spotted goatfish (Pseudupeneus maculatus) was sold as red mullet (Mullus sp.).
Their findings stimulated new studies, all of which showed the widespread mislabelling of seafood in North America. Determined to see whether his initial tests in Canada extended beyond the Toronto region, Hanner teamed up with investigative journalists from across the country and analysed a further 236 samples from the east coast to the west; they found that 41 per cent of the samples were mislabelled.
In the US, the international organisation Oceana started conducting its own research to determine the prevalence of fraud in areas such as Boston, South Florida, New York and Los Angeles. The numbers were alarming. They worked with Hanner and his team at Guelph to conduct their analyses and between 2010 and 2012 collected more than 1,200 seafood samples from 674 retail outlets in 21 states. It was one of the largest seafood fraud investigations in the world at the time. They found that 33 per cent of the samples analysed were mislabelled. Red snapper still had the highest mislabelling rate of any species – only seven of the 120 samples labelled as red snapper were truly red snapper. Nearly 10 years after Marko’s study, Oceana found little had changed; the mislabelling of red snapper is still rampant.
In 2012–13, the FDA stepped in and conducted its own study. They sampled products collected from the distribution chain across 14 states, prior to the fish arriving at the retailers (restaurants and supermarkets), targeting species that had a history of mislabelling. What the FDA found was that 15 per cent of the products they tested were mislabelled – a number quite a bit lower than the findings of Oceana, Hanner and others. Some groups, such as the National Fisheries Institute (a trade association), used these results to raise doubt about the other studies, suggesting the issue was not as widespread as they suggested. What they failed to point out, however, was that the studies targeted different areas of the supply chain. What the FDA’s study did highlight was the proportion of mislabelling happening prior to the seafood getting to retailers – evidence that mislabelling is happening along every step of the supply chain.
Of course, it isn’t just North Americans who are being fleeced on their fillets.7 In Australia, studies found that 41 per cent of red emperor (Lutjanus sebae) and 46 per cent of dhufish (Glaucosoma hebraicum) were mislabelled, and 13 per cent of barramundi (Lates calcarifer) were also mislabelled, substituted with cheaper species such as Nile perch (Lates niloticus) and King threadfin (Polydactylus macrochir). In New Zealand, 40 per cent of 200 sampled fillets labelled as lemon sharks were other shark species, including hammerheads and bronze whalers, which it is illegal to harvest. Eighty per cent of fish samples acquired in Brazilian markets were mislabelled. Up to 36 per cent of hake sold in Spain and Greece was cheaper African species that were labelled as American and European species. In the UK, 10 per cent of 380 samples of fish collected from catering establishments was mislabelled, with swaps between cod and haddock most common. Just over 5 per cent of white fish samples collected from six major supermarket chains across the UK were mislabelled, and while this may seem like a low percentage, in terms of volume this could translate to 200 million mislabelled products sold in the UK annually. There are mislabelling examples from Ireland, Turkey, Denmark, Egypt, the Philippines, South Africa ... the list goes on. Seafood substitution is a global issue.
It’s global and it’s in all manner of products. Barcoding is revealing substitutions in smoked fish, dried fish, boiled fish, fried fish, fresh fish and frozen fish. What barcoding can’t do, however, is trace fish back to particular regions or fish populations. Some substitutions are rather obvious – an Atlantic halibut was clearly not caught in the Pacific, for example. Finer resolution, however, such as whether an Atlantic cod was caught in the North Sea or the Baltic Sea, is beyond barcoding capability. To get to these origins, one needs to look at multiple genetic markers. This is exactly what the international EU funded project FishPopTrace set out to do. Focusing on commercial species that are susceptible to overfishing, the project began to identify minute mutations in the DNA sequence, known as Single Nucleotide Polymorphisms (SNPs), which could be used to distinguish different populations. Like the chemical fingerprints in oil that we described in Chapter 3, these are genetic signatures that become embedded in the DNA of discrete spawning populations of fish. These population markers – mutations that can be traced back to certain populations – provide a robust method for tracing fish products back to the source. But it can also be used in enforcement and as a deterrent for illegal, unreported and unregulated (IUU) fishing.
Another area where barcoding falls short is with tinned fish. The high pressure and temperature involved in the canning process clips the DNA into much smaller fragments than the 650bp CO1 sequence. The DNA is still there, but it’s far more degraded. The FDA was concerned that expensive salmon species, such as sockeye, were being adulterated with cheaper salmon, such as pink and chum. They contacted Hanner to help them develop a rapid and sensitive method for testing tinned salmon. Hanner and his colleagues turned to other DNA-based methods that exploit small regions of DNA that differ between the seven species of Pacific salmon and trout, and the one Atlantic salmon species.8 If the species is present in the sample, a chemical compound bound to the species-specific DNA segment will fluoresce. The amount of fluorescence can be measured to give an estimate of the quantity of the DNA and therefore the proportion of each fish in the sample – for example, 20 per cent pink and 80 per cent sockeye. The test is sensitive enough to detect if there’s as little as 1 per cent pink salmon in the tin of sockeye, but the FDA decided to provide a little room for error and say that anything over 5 per cent adulteration would be investigated as deliberate fraud.
Farmed fish passed off as wild
The adulteration of some sockeye with pink salmon seems relatively benign in the overall scheme of things. A far more common and potentially harmful substitution among salmon is the sale of farmed Atlantic salmon labelled as wild-caught Pacific species. Once again, there is considerable economic incentive to make this substitution. Wild-caught fish sell for three to four times the price of a farmed fish. Concerns over the sustainability of farmed fin fish have created a market preference (and premium) for wild-caught animals. Investigations have found that between 15 and 75 per cent of salmon labelled as wild are actually farmed. The New York Times tested salmon from eight stores around the city in 2005 and six of the stores were selling farmed salmon as wild-caught. At that time wild salmon was selling for as much as US$63 per kilo (£40/kg or US$29/lb) in the Big Apple. Farmed fish was selling for between US$11 and US$26 per kilo (£7–£16/kg or US$5–US$12/lb) – a potential profit of up to US$52 per kilo (£33/kg or US$24/lb)! Approximately 1.8 million tonnes (weighed with the head on, but the guts removed) of farmed salmon was produced in 2013. If we use the conservative value of 15 per cent mislabelling, that would suggest 276,000 tonnes of farmed salmon went out into the market labelled as wild-caught in just a single year.
And it’s not just salmon. More than 100 samples collected from retailers in the UK found that 11 per cent of sea bream and 10 per cent of sea bass were farmed rather than wild as the labels claimed. These are concerning values for consumers who are trying to avoid buying farmed fish.
More alarming is that all of these farmed fish have seeped in among wild fish, which are subject to different inspection regimes. Because farmed animals are reared in pens at high densities, pesticides and antibiotics are sometimes needed to control and cure parasites and disease. As a result, farmed species undergo random testing to ensure residues of these drugs don’t remain in the tissues. As with any farmed animal intended for consumption, farmed fish are subject to withdrawal periods. Before this conjures up images of anxious shaky fish circling the pens in search of a hit, this is the period of time where the fish must be drug-free before they can be killed and sold, to ensure any drugs are worked out of the system. The period of time differs greatly depending on the drug and the country.
Governments have introduced stricter regulations on many of these drugs to try and reduce their use on fish farms, yet there is some evidence that use is increasing. In 2012, a freedom of information request to the Scottish Environment Protection Agency (SEPA) revealed that the use of pesticides in the Scottish fish farming industry increased by 110 per cent over 2008 values, while salmon production had only increased by 22 per cent. It is suspected that more drugs are being used because the parasites and bacteria are becoming resistant to them. As we have learned in human health, the over-prescription of antibiotics can lead to the development of resistant strains. This has not been helped by the prophylactic use of antibiotics on many farms, particularly in developing countries. Fish are regularly dosed with antibiotics to encourage growth and prevent bacterial infections, particularly if the farm is a bit dodgy in terms of its hygiene. This creates the ideal conditions for resistant bacteria to thrive and susceptible bacteria to die off. The resistant bacteria can pass on those resistance genes and so begins perhaps one of the biggest threats we face to human health in our generation, as many of the drugs used on fish farms are also used in human health. Amoxicillin, for example, is a broad spectrum antibiotic used to treat fish for an infectious bacteria that causes, among other things, some very nasty lesions in the muscle of the animal. This drug is also one of the most commonly prescribed antibiotics for children for ailments such as ear, throat and urinary tract infections.
Farmed fish are also tested for other toxins, such as polychlorinated biphenyls (PCBs) and dioxins (by-products of some manufacturing processes), which they can accumulate from the environment but also from the fish feed. Malachite green – a fungicide – is another drug that is randomly tested for. It was once commonly used in the fish-farming industry to rid the salmon pens of parasites, but it is also a carcinogen in humans. For this reason, it was banned for use in fish farming in a number of countries, including the US, EU and Canada, by the early 2000s. Routine surveillance checks for the drug continue to be carried out and, sadly, there have been numerous accounts of the drug being found in farmed fish long after the ban.
So, between persistent organic pollutants, pesticides, antibiotics and a carcinogenic fungicide, it’s not surprising that farmed fish should be subjected to more rigorous testing than wild-caught. As of 2007, after a number of incidents where imported farmed products tested positive for banned drug residues, the FDA began detaining farm-raised catfish, basa, shrimp, dace and eel coming from China; they won’t release them until they’re proven to be drug-free. But what if the animals aren’t labelled as farmed? Farmed fish that evade testing may have unacceptable levels of toxins in them. Levels are likely to be low, but this is almost more dangerous, because eating them won’t cause any acute poisoning symptoms that will go noticed, and yet there may be unknown long-term health consequences.
While misrepresenting a farmed animal as wild is dishonest (and illegal), there may be some cultural differences that make such a swap seem more justified. While many North Americans and Europeans view farmed fish as inferior, some Asian cultures prefer them. Farmed is often considered superior for sushi as there is more control over parasites, in which case swapping a farmed fish for a wild fish might actually be seen as improving the product. It doesn’t justify the deception, but it does perhaps offer some insight into the minds of those committing the fraud. It’s not all black and white.
‘Swordfish’ with a side of anal leakage
Substitutions between farmed fish and wild fish are not the only risk to human health. In 2007, Hanner and his team at Guelph were once again at work with the FDA to try and resolve a case where members of a family were hospitalised after eating home-made fish soup made with two frozen monkfish purchased at an Asian market in Chicago. Hanner’s group used DNA barcoding to reveal that the monkfish weren’t monkfish, but actually puffer fish (Lagocephalus species).9 The FDA were surprised at the result as puffer fish is a highly regulated species in the US, but they tested the soup for the presence of a toxin associated with puffer fish, and there it was.
Puffer fish are not the fastest fish in the sea and have therefore developed some handy little defences. If they feel threatened by a predator they will take in water (or air if they are out of water) and swell in size, at the same time exposing a plethora of spikes all over their body. Most predators are deterred by the thought of trying to fit a spiky football into their mouth and give up, but those that don’t are rewarded by a rather unpalatable mouthful. Puffer fish contain tetrodotoxin – a neurotoxin – in their tissues, particularly in the liver, ovaries, intestines and skin. The toxin is thought to be produced by bacteria living in the gut as well as on the skin of the fish. While the toxin isn’t usually enough to kill the predator, it may deter them from ever eating puffer fish again. For some predators, the toxin is lethal, while others, such as tiger sharks, seem unaffected and happily munch on puffer fish. However, as little as two milligrams of the toxin is potentially enough to kill a human (though this varies depending on the person’s age, weight, health and sensitivity). Hence the strict regulations. Only certain parts of certain species are permitted into the US through the JFK International Airport in New York. These fish have to be prepared by authorised facilities, certified for safe consumption and sold only to restaurants belonging to the Torafugu Buyers Association.
Yet, here were two puffer fish that had escaped all of this under the guise of monkfish. And to make matters worse, there were unusually high levels of the toxin in the muscle of the fish, which isn’t usually the case. Luckily, medical staff and numerous agencies responded quickly and managed not only to treat the family appropriately but also to track down the source of the fish. The supplier was placed on the FDA’s Import Alert list for misbranding. One of the family members, who had consumed three milligrams of the toxin (a potentially lethal dose), suffered from chest pains, vomiting, numbness and profound weakness in her lower extremities. She required three weeks of inpatient therapy before being discharged to a long-term care facility for further rehabilitation. In summary, if you experience numbness or tingly lips after eating so-called monkfish (or any other fish for that matter), seek medical advice immediately.
Seafood substitution also presents a health risk to people with seafood allergies. People rarely have allergies to all seafood, just particular groups of seafood. Someone with an allergy to fin fish, for example, would no doubt be very unhappy to buy a crab product, only to find out that it is actually a surimi-based product made with hake. Even among fin fish, people may have an allergy to salmon but not tuna. It turns purchasing fish into a rather risky game of allergen roulette.
Somewhat less lethal but nonetheless unpleasant is the substitution of oilfish, snake mackerel or escolar for white tuna, swordfish and even Atlantic cod. The oilfishes contain high levels of wax esters, compounds made up of a fatty acid and a fatty alcohol that are largely indigestible. So, after a lovely romantic fish dinner, you may be in for a night of keriorrhea. Put bluntly, this is uncontrollable anal leakage of an orange to greenish oil – maybe not what you had in mind for those fancy new third-date underpants. Add a little vomiting and abdominal cramps to this and it makes for a lovely evening all around. In 2007, Hong Kong consumers who were excited to be paying four to five times less for ‘Canadian Atlantic cod’ had first-hand experience of the effects of oilfish, with over 600 people falling ill after eating the fish. Japan and Italy have banned the sale of escolar, while countries like Australia, Canada, the UK and the US have opted for more of a public education/fact sheet approach to protecting consumers, while still permitting its sale. Might be worth familiarising yourself with oilfish and escolar!
Fish laundering is bad for the environment
Seafood fraud is hurting our wallets, our health and sadly also our environment. Just as shell companies can help to clean drug money, species substitution helps to launder the products of IUU (illegal, unreported and unregulated) fishing. These practices undermine conservation measures put in place to help conserve species by catching more fish than is allowed, fishing in closed areas or out of season, using gear that is prohibited, or catching outside of the allowable size limits. In 2004, a US company, Neptune Fisheries Inc., was convicted of conspiring to import 86,182kg (190,000lb) of undersized frozen spiny lobster worth more than two million US dollars (over £1.2 million). Two years later, a husband-and-wife team operating Anchor Seafood Inc. were convicted for smuggling 7,500kg (16,500lb) of undersized spiny lobster into the US from Jamaica. They organised 40 illegal shipments between January 2000 and January 2001, valued at approximately US$229,000 (£144,630), violating harvest restrictions in both Jamaica and Florida. Both cases were uncovered by the US’s National Oceanic and Atmospheric Administration (NOAA) Fisheries Office for Law Enforcement. Special Agent Scott Doyle, an investigator for NOAA for over 27 years, said in an interview to the Baltimore Sun, ‘If I was going to be a criminal, I would be in the fish and wildlife smuggling business. Nobody has any idea what’s going on. They just buy fish.’10
IUU is estimated to be worth €10 to 20 billion (£7.1–14.2 billion, US$11.4–22.8 billion) per year and is equal to at least 15 per cent of the reported global catch. In some parts of the ocean it is thought that as much as 37 per cent of the catch is illegal and for some high-value species, such as tuna and swordfish in the Mediterranean and sharks across Europe, IUU fishing may be as high as 50 to 75 per cent. It is a major contributor to the depletion of fish stocks, and connections have been made with drug trafficking, human trafficking and other illegal activities. It is the biggest global threat to the sustainable management of our fisheries. The EU has banned imports from fishing vessels registered to countries that aren’t taking sufficient action against IUU, including Belize, Cambodia and Guinea. Curaçao, Fiji, Ghana, Korea, Panama, Togo and Vanuatu all received formal warnings of a potential ban in 2012. Since then, Belize has had its ban lifted owing to the steps the country has taken to help address illegal fishing. Fiji, Panama, Togo and Vanuatu have also been de-listed by the EU for their positive actions.
While seafood fraud is helping to launder dirty fish, it’s also undermining any efforts consumers are taking to make sustainable seafood choices. Many groups around the world, such as the Marine Conservation Society (UK), the Australian Marine Conservation Society, World Wide Fund for Nature (WWF), the Monterey Bay Aquarium (US) and the David Suzuki Foundation (Canada), have developed pocket guides to help consumers make better seafood choices to support sustainable fisheries. Yet, these efforts are undermined if the species or the location where it was caught is falsely labelled. For example, Atlantic halibut, Hippoglossus hippoglossus, was heavily overfished in the nineteenth and twentieth centuries and has never recovered. It is listed as Endangered on the International Union for Conservation of Nature (IUCN) Red List. Today there is no directed fishery for this species within US federal waters with the exception of some small-scale harvests off the coast of Maine. The Monterey Bay Aquarium’s Seafood Watch guide recommends avoiding Atlantic halibut, yet the species is caught as by-catch in US and Canadian ground fisheries. As there is little market for these endangered species, they are often labelled as the closely related Pacific halibut, Hippoglossus stenolepis. Pacific halibut fisheries have been better managed; a number of them have been independently certified by the Marine Stewardship Council (MSC) and are listed as good alternatives or better choices in seafood guides. To confuse matters, the Canadian Atlantic halibut fishery achieved MSC certification in 2013, which introduces the potential for non-sustainable US-caught populations to be passed off as an MSC-certified fishery.
It is incredibly disempowering to think that the efforts we, as consumers, go to in order to inform ourselves and feel able to interrogate our fishmongers about the fish we’re buying are futile at least once in every three purchases. Would we be so tolerant of this if mountain gorillas or Ganges River dolphins were being ground up and making their way illegally onto our plates?
Fish out of water
Of course seafood fraud isn’t only about species substitution. There are many other ways to cheat consumers. Perhaps the most common form of seafood fraud, though less talked about, is short-weighting. This is when processors increase the weight of the product by adding a little more breadcrumbs or batter to the fish fingers or more ice to the prawns and include this in the net weight of the product. In 2012, the Boston Globe surveyed 43 samples of seafood collected from supermarkets across the state of Mass-achusetts. They found that 20 per cent of the samples glazed in ice weighed less than the packaging claimed – a practice known as over-glazing. Another study of 240 shrimp samples imported into Europe from South-east Asia found that half of them were short-weighted, with some samples weighing as much as 28 per cent less than the weight declared on the label.
Another method of short-weighting is to over-soak the seafood products in sodium tripolyphosphate (STPP). This is a preservative commonly used in the food industry (labelled as E451 in Europe) and it helps keep seafood firmer and smoother by helping retain moisture in the flesh of the fish. STPP and other similar additives are a legitimate way of keeping seafood from drying out in the freezer. However, over-soaking in STPP causes the flesh to absorb and retain an unnatural level of water, which the customer ends up paying for. Scallops are particularly prone to this practice as it bloats them up into tempting plump portions. As a result, regulators have put in place specific regulations on the water content of scallops. Canada doesn’t permit the use of phosphate additives in scallops at all and considers any scallops with a moisture content above 81 per cent to be adulterated with water; the average water content of a scallop harvested fresh from the ocean is about 75 per cent. The US and EU require that phosphate additives and any added water be declared on the label. This includes any water inadvertently added during storage by the uptake of melted ice water. Leaving scallops sitting on ice for 10 days in a commercial storage facility can increase moisture content by 2 per cent.
Added water in our seafood may not pose any health risks and the cost to the individual consumer can be relatively low. It may be a matter of a few pennies here and there, but one US investigation in 2010 across 17 states found consumers paying as much as US$50 per kilo (£31.66/kg or US$23/lb) for unwanted ice. Multiply this through the industry as a whole and these costs are extraordinary – millions of US dollars each year. While this clearly constitutes economic fraud, the priority for seafood inspections will continue to be around food safety. Also, because glazing and soaking are legitimate processing techniques, it can be difficult to detect when the line has been crossed from providing customers with a better product to providing them with a whole lot of water.
As consumers it’s virtually impossible for us to determine whether we’re paying for excess water at the point of purchase. This water doesn’t get released until cooking, when you watch in horror as your sautéed fish dish shrivels and falls apart before your eyes as your dinner guests sip their wine. A fish fillet over-soaked in STPP will shed the excess water during cooking and release a milky white liquid in the pan. This liquid can wreak havoc on any sauce you might be using, and when the liquid is released, the fillet becomes very fragile, often breaking apart despite gentle handling. Over-soaked scallops will behave similarly and may not sear properly because of the added water. These revelations (if we even have them) may frustrate us, but will it be enough to motivate us to complain to the retailer or to the processor, or to report it to the responsible legal authorities? Let’s face it, most of us will probably have forgotten about it by the time dessert rolls around. You’re unlikely to get too emotional over a few pence-worth of water added to your fish fillet, just as you are unlikely to cross the street to pick up a couple of wayward pennies on the side of the road. But would you be willing to cross the road with a thousand people to get your share of a couple of million pounds? In 2010, over 29 per cent of fish for human consumption was sold as frozen product – worth just shy of US$55 billion (£35 billion). Those pennies add up.
Closing the net on seafood fraud
There’s no denying that seafood fraud is rampant, so the next question is what is being done about it. It would seem that some governments are starting to wake up to the issue of seafood fraud and its risks to consumers, the extent of economic fraud and its role in supporting IUU fishing. As of December 2014, EU rules changed so that it became mandatory to provide the scientific name as well as the commercial name for species on all unprocessed and some processed fishery products (mainly smoked and salted products). Canned, composite and breaded products do not need to provide a species name, but they do have to provide a quantity on the ingredients label. A tin of mackerel, for example, must have ‘mackerel (75%)’ in the ingredients list. In July 2014, President Obama established a national task force to combat seafood fraud and IUU fishing. The recommendations of the task force, released in December of the same year, included better systems for gathering, sharing and analysing information about the seafood products entering the US as well as an effective seafood traceability programme.
More random unannounced testing of seafood products along every link of the production chain would increase the risk for fraudsters, and stiff penalties when caught would probably help tip the scales on seafood fraud. DNA barcoding has provided a global standard for identifying mislabelled species, but if there are no resources available to do the testing, it’s useless in battling food fraud. When it comes to inspections, most countries are just trying to manage their risk. How they choose to do that may vary, but it really is about targeting items that present the highest risk to the public. Every country has particular products, importers and countries that are flagged due to a previous incident or known conditions in that country. These shipments will always be inspected. For example, the CFIA in Canada inspects 100 per cent of fresh or frozen tuna (all species) imported from India by Moon Fishery India Pvt. Limited – but they’re only looking for Salmonella. Less attention is generally given to imports from countries where there are existing agreements. Fishery products move relatively unrestricted between EU countries, for example. Some countries, such as Canada and New Zealand, try to maintain standards by specifying that seafood importers be licensed, while others, such as the US, look for certification in the foreign exporters. Australia inspects 100 per cent of risky foods and then reduces this rate to 5 per cent once there is a demonstrated history of compliance. The FDA in the US inspects 1 to 2 per cent of fish imports and less than 1 per cent of these will have any analyses done. An inspection can entail anything from going over paperwork, to climbing into the shipment (quite literally) and pulling out samples to make sure they match the paperwork, to opening the product up to conduct testing.
In fact, it is very difficult to gauge just how much testing is being conducted in a lot of countries. The FDA inspection rate seems low, but US Customs and Border Protection and NOAA are also doing inspections. The industry itself is conducting internal inspections; academics and independent watchdog organisations, journalists and conservation groups are doing testing. At the time of writing this book, US Senator David Vitter had introduced the Imported Seafood Safety Standards Act. If the bill is passed, it will mean more inspections and testing of imported seafood and increased penalties for mislabelling. Bills such as this can make authenticity testing of equal importance to pathogen testing.
Governments and stakeholder consortiums are also taking steps to standardise and harmonise testing around the world to better cope with the global movement of seafood. In 2011, the FDA released its own DNA barcode database. The FDA is part of the Barcode of Life consortium and has adopted its methods to target the CO1 gene sequence, but they wanted to develop an extremely rigorous database so that any evidence gathered could stand up in court. They worked with the Smithsonian Institution’s Laboratories for Analytical Biology and the Division of Fishes to ensure the database of 250 commercially important species were all identified by experts from the Smithsonian. While the FDA’s database may be more rigorous in terms of identification than BOLD, it isn’t as comprehensive yet, which may be limiting. More than 20 FDA analysts are now trained to do DNA barcoding. Europe is also considering adopting the Barcode of Life protocols. Labelfish is a multi-national project that was set up to develop a standardised approach to testing seafood across Europe. They are recommending that barcoding has consistent and straightforward protocols that can be used by most authenticity labs across the EU, and as countries such as Canada and the US are using these methods, it provides a common international approach to combating fraud.
With governments limited by resources and millions of tonnes of fish moving about the world, some groups within the industry are taking matters into their own hands. Fishermen are seeking ways to make their product traceable. In the US, fishermen in the Gulf of Mexico have developed the Gulf Wild brand of fish, which are responsibly caught and safety tested. In the interests of fraud prevention, Gulf Wild fish are tagged with a unique number that enables consumers to go online and use that number to find out who caught the fish and where.
In 2009, fishermen on the west coast of Canada approached the non-profit organisation Ecotrust Canada about setting up a system of traceability for their product. Ecotrust Canada launched ThisFish in October 2010. It’s an extensive database that, like Gulf Wild, uses a unique code associated with each fishery product to allow consumers to learn more about the product – who caught it and where, as well as what equipment they used. The code might be found on the label of a can of salmon or on a tag around a lobster claw. The programme is voluntary and fishing vessels simply have to register. Processors and distributors can also register and upload information to the system so that consumers can get information regarding every link along the seafood supply chain. There’s even an option for consumers to leave feedback for the harvesters and processors. It’s a sophisticated database and a simple idea that helps reconnect fishermen with consumers and share the story behind each seafood product. And the stories behind these Canadian products have an international reach. Shortly after ThisFish launched, they were seeing Canadian lobsters being tracked by consumers in Belgium, the Netherlands, Norway, France and Spain. ThisFish has expanded beyond Canada, working with fisheries in Indonesia, Iceland, the Netherlands and the US.
As well as improving traceability, the industry is conducting authenticity testing as part of its protocols. A seafood importer based in Victoria, Canada started incorporating DNA testing into its business practices in 2011. Tradex Foods Inc. started DNA testing seafood that it was bringing in from China for its Sinbad house label. The samples are taken by its China office and flown to a US-based testing facility while the fish itself is making its way across the Pacific by ship. It costs the business between CA$700 and CA$2,100 (between £360 and £1,078) each month to do the testing, but it’s part of their business model not only to have a rigorous quality control process, but also to help eliminate fraud.
Tradex uses a private laboratory in Illinois called ACGT, Inc., which is making a niche for itself with seafood importers, wholesalers and retailers who view DNA testing as an investment in their businesses. As consumers become more infuriated by fraud and begin to look for proof of authenticity or traceability in their products, this kind of initiative could help give forward-thinking businesses a market advantage.
Just as we complete this chapter, researchers at the University of South Florida have come up with a hand-held device known as Grouper Check, which uses DNA-based methods to check whether your grouper is really a grouper. The instrument sells for about US$2,000 (£1,268). We are likely to see more of these technologies hitting the market over the next few years, which could make it more cost effective for industry to conduct more internal testing.
As consumers, there are steps we can take as well to help reduce our exposure to seafood fraud, whether it’s species substitution or short-weighting. First and foremost, know your product. If you are out to buy monkfish, search for some pictures of the species before you go and, most importantly, buy fish that have some identifiable features, such as skin and maybe even a tail fin. If you can, buy fresh to be sure you aren’t paying for any additional water. Second, know your seller (and any others who have touched your fish along the way). If you are lucky enough to live in a place where you are able to buy direct from the people catching the fish, then do so. Go back to those you like and that provide a good product for a fair price because building a relationship with the people who sell you food will probably be more valuable in understanding your food than any label could ever be. If you can’t buy directly, then look for tracing organisations where you can check provenance, such as ThisFish or Marine Stewardship Council. Third, know your season. Just like fruit and vegetables, fish have seasons too. The substitution of farmed salmon for wild is far more common when wild salmon aren’t in season (approximately October to April).
There is no doubt that things become more complicated when dining out. The Oceana study found that 74 per cent of sushi bars, 38 per cent of restaurants and 18 per cent of grocery stores they took samples from were selling mislabelled fish. Technologies such as Grouper Check may be a prelude to consumer-targeted kits; a dipstick test much like a pregnancy test is possible, for example. But instead of peeing on it you hold it to your swordfish steak: two blue lines confirms you’ve been served swordfish, while one blue line means it’s something else. Of course, just because such a test is possible doesn’t mean there would ever be a market for such a thing. Ten quid for a one-use test that tells you your fish is what the menu says isn’t likely to be a big seller, and what would it say about the state of our food industry if it was?
Despite the rampant mislabelling in sushi bars and restaurants, there are establishments that are reputable and care not only about providing quality food to their customers, but also about making a commitment to the health of the ocean on which they depend. Moshi Moshi in the UK comes to mind immediately, as founder Caroline Bennett has been at the forefront of campaigns to protect fish stocks. She works alongside ethically minded local fishermen to provide seasonal quality products to her London sushi restaurants. She knows her seafood. And if you care about the environment, your body and your wallet, this is the best advice: get to know your seafood.