Hake DNA testing: How we did it

Introduction

The most commonly consumed fish in Spain is hake. Based on allegations of fraud in the hake market, ICIJ carried out a DNA study on hake in the Spanish market.

Last year a team of Spanish and Greek researchers at the University of Oviedo and Aristotle University of Thessaloniki published a study on high levels of apparently intentional mislabeling of hake imports in their respective countries.

Southern African hake species were mainly being marketed as European or South American Hake. European and South American hakes are worth double the amount of southern African hakes, researchers noted. Following publication, the Spanish government requested a copy of the study, but the official report did not include company names. The lead researcher Eva García Vázquez told ICIJ she would have provided the names associated with the mislabeled samples had officials asked.

The researchers also experienced complaints from industry. So when ICIJ requested the company identities, García declined to share that information. ICIJ decided to undertake its own snapshot study in Madrid –Spain’s capital– to determine if mislabeling was still occurring. ICIJ commissioned García Vázquez and her team at the University of Oviedo to conduct a second study to determine the extent of mislabeling in the fresh and frozen hake markets.

DNA experts told ICIJ our methodology was sound and simple. Geneticist Einar Neisen from the National Institute of Aquatic Resources at the Technical University of Denmark called the work “a walk in the park” as it was easy to identify the different species. The case might have been different if ICIJ were trying to determine the geographical locations among samples of the same species.

Between June 9 and 14, ICIJ reporters Marcos García Rey and Mar Cabra purchased 100 frozen samples and 50 fresh samples in the Madrid region of products labeled as: Merluccius capensis (Shallow water cape hake), M. paradoxus (Deep-water Cape Hake), Merluccius polli (Benguela Hake), Merluccius senegalensis (Senegalese Hake), M. merluccius (European Hake), M. australis (Southern Hake), or M. hubbsi (Argentine Hake). Because of time and logistical constraints, ICIJ reporters were unable to sample over an extended time period or outside the Madrid region.

Reporters purchased the frozen samples at the top supermarkets in terms of sales. These are also the markets that carry brands from some of the country’s largest importers of frozen seafood. ICIJ also selected samples from companies that sell bulk fish. According to a study by the Ministry of Environment, Agriculture and Fisheries on European hake, Spaniards buy half of their fresh hake in supermarkets and half in traditional fish markets. Fresh samples were purchased from top chain stores as well as local fishmongers selected randomly within the city of Madrid.

The shopping process was documented in spreadsheets, which contained the following information: sample number, date of buy, name of the shop, address, scientific name indicated, reported origin, frozen/fresh, presentation (whole piece, slice, filet, tail), commercial brand, distributor, ship owner, price per kilo and a field for other notes. This information was later typed into Excel.

The purchase and sampling process was captured on video. Each sample was placed in a sterilized plastic cup filled with 100 ml of ethanol for delivery to Oviedo. Following the recommendations of the researches, the cups had inside a penciled piece of paper with the sample number, which was also indicated with a sticker outside. This way, scientists could do a blind analysis of the fish.

ICIJ paid the University of Oviedo, Department of Biology of Organisms and Systems €1,500 to test the samples and provide a written analysis.

For reference, six hake samples of known origin and species were placed as positive controls in each reaction, as well as a negative control containing only water and PCR reaction mixtures, to exclude any possible contamination of vials and materials.

The extraction of DNA was performed employing a protocol based on resin Chelex. A fragment of each sample was introduced into an Eppendorf tube containing a solution of Chelex100 with proteinase K. The tubes were incubated at 55ºC for 1.5 hours. Finally, the samples were kept at 100ºC for 20 minutes for deactivating the proteinase K. The DNA remains diluted in the supernatant, which is employed for further reactions.

The species-specific DNA marker employed was the cytochrome oxidase subunit I gene.

Polymerase chain reactions (PCR) for amplifying the marker from DNA samples were performed in a total volume of 40 μl, employing the Barcode fish primers described by Ward et al. (2005). The PCR program was: initial DNA denaturing at 95 ºC for 5 minutes; 35 cycles of: denaturing at 95 ºC for 20 seconds, annealing at 57 ºC for 20 seconds, extension at 72 ºC for 30 seconds; final extension at 72 ºC for 10 minutes. The four products obtained after the PCR, which are many copies of the DNA marker, were purified and sequenced by Macrogen Holland using an Automatic sequencer 3730XL under BigDye Terminator cycling conditions. All the laboratory process was repeated employing a new bit of tissue taken from each sample. The results were identical for the two aliquots of each sample and ensure repeatability of the analysis.

To determine the species of a sample, the sequence obtained from the sample was compared with those contained in international databases, including the laboratory’s reference sequences for all Merluccius species in the GenBank, employing the program BLAST within NCBI (http://www.ncbi.nlm.nih.gov/). Species assignation was made based on > 99% sequence similarity with GenBank voucher specimens Sequence comparison was made independently by two different researchers to ensure reliability of the species determination.

The genetic results were recorded, containing the number of each sample and the corresponding species as authenticated from DNA. Up to then, the researchers had not received any information on the brands or species identified when the fish were purchased. This data was exchanged by email on June 28 at 5 pm. With the complete results, the University of Oviedo Researchers wrote a report analyzing the findings from a scientific perspective.

The mislabeled samples were double checked in a five-step process comparing notes, videos and receipts of the sales.