Meat Science 90 (2012) 686–689

Contents lists available at SciVerse ScienceDirect

Meat Science

j ourna l homepage: www.e lsev ie r .com/ locate /meatsc i

Detection of porcine DNA in gelatine and gelatine-containing processed foodproducts—Halal/Kosher authentication

Yasemin Demirhan a, Pelin Ulca a, Hamide Z. Senyuva b,⁎a A&T Food Laboratory, Mega Center No. 29 34045, Istanbul, Turkeyb FoodLife International, ODTU Teknokent Ikizler Binasi No. ara-1, 06531 Ankara, Turkey

⁎ Corresponding author. Tel./fax: +90 312 2101060.E-mail address: hamide.senyuva@foodlifeint.com (H

0309-1740/$ – see front matter © 2011 Elsevier Ltd. Alldoi:10.1016/j.meatsci.2011.10.014

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 July 2011Received in revised form 24 October 2011Accepted 26 October 2011

Keywords:GelatineHalal authenticationKosher authenticationReal-time PCR

A commercially available real-time PCR, based on a multi-copy target cytochrome b (cyt b) using porcine spe-cific primers, has been validated for the Halal/Kosher authentication of gelatine. Extraction and purification ofDNA from gelatine were successfully achieved using the SureFood® PREP Animal system, and real-time PCRwas carried out using SureFood® Animal ID Pork Sens kit. The minimum level of adulteration that could bedetected was 1.0% w/w for marshmallows and gum drops. A small survey was undertaken of processedfood products such as gum drops, marshmallows and Turkish delight, believed to contain gelatine. Of four-teen food products from Germany, two samples were found to contain porcine gelatine, whereas oftwenty-nine samples from Turkey twenty-eight were negative. However, one product from Turkey containedporcine DNA and thus was not Halal, and neither was the use of porcine gelatine indicated on the productlabel.

© 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Gelatine is a highly processed protein, which is widely used as agelling and thickening agent (E441) in a variety of foodstuffs suchconfectionary products, water-based desserts and in the pharmaceu-tical industry e.g. in gel capsules for medicines. Gelatine is obtainedby hydrolysis of collagen, which is extracted from materials such asbones, hides and skins from animal slaughterhouses (Karim andBhat, 2008). Gelatine production involves controlled acidic or basichydrolysis of connective tissue raw material, high temperature ex-traction with water, sterilisation, and drying. These processes arenot standardised and have effects on the properties of the final gela-tine product. In the final gelatine product, both proteins and nucleicacids are highly degraded (Boran and Regenstein, 2010). Additionally,the amount of DNA in gelatine is very low and differs from material-to-material.

In Europe, about 80% of edible gelatine is produced from pigskin,but vegetarian, Halal and Kosher gelatine, prepared from seaweed,fish bones or non-porcine sources, is also available (Boran andRegenstein, 2010). Although gelatine must be labelled appropriately,once it has been manufactured, purified and in commercial trade, it isdifficult to ensure its provenance or whether it has been inadvertent-ly mixed at any point in the food chain. It is therefore important tohave methods available whereby pure gelatine can be checked to en-sure its authenticity and that it is free from cross-contamination with

.Z. Senyuva).

rights reserved.

porcine gelatine. Equally the ability to test processed food productsfor the presence of porcine gelatine is an essential requirement forfood control in Muslim or Jewish communities (Riaz and Chaudry,2004).

Most published methods have focussed on meat species identifi-cation rather than identification of gelatine. Polymerase chain reac-tion (PCR)-based methods have been the most successful in termsof both specificity and sensitivity of species detection. A review ofPCR-based methods applied to the authentication of meat productscites some twenty-nine publications (Mafra, Ferreira, and Oliveira,2008). Extraction of good quality DNA is an important pre-requisitefor PCR-based analysis and this can be a potential problem if therehas been extensive heat processing. For example, only poor qualitygenomic DNA was extractable from bread and biscuits, although it isnot clear if this was because of high temperature degradation duringcooking or because lard, containing only small amounts of DNA, wasthe target source of DNA (Aida, Che Man, Raha, and Son, 2007).DNA has been isolated from meat and cheese using a standard CTABprotocol and from milk using a PromegaWizard Magnetic kit and pu-rified by Qiagen silicon spin columns (Zhang, Fowler, Scott, Lawson,and Slater, 2007). With gelatine, despite both extensive heat andchemical treatment, it has been demonstrated that it is possiblewith nucleic acid binding columns or standard ethanol precipitationto obtain template DNA. Analysis of the extracted DNA on agarosegels was used to demonstrate that it had remained essentially intact(Tasara, Schumacher, and Stephan, 2005).

A number of PCR approaches have been used to detect porcineDNA in meat and meat products (e.g. Binke, Spiegel, and Schwägele,2007). Using restriction fragment length polymorphism (RFLP)

Table 1Composition of mixtures of gum drops/marshmallows mixed with various levels ofporcine gelatin.

Level of addition (%) Wt of food (mg) Wt of bovine gelatine (mg)

1.0 495.0 5.03.0 388.0 12.05.0 380.0 20.010.0 450.0 50.0

687Y. Demirhan et al. / Meat Science 90 (2012) 686–689

analysis of a conserved region of the mt cytb DNA extracted from sau-sages, clear PCR products were produced on amplification (Aida et al.,2007). For the same food products, using species-specific PCR, detec-tion of a conserved region in the mt 12S ribosomal RNA (rRNA) genewas employed as an alternative method. The extracted DNA was am-plified by PCR targeting specific regions of the 12S rRNA gene of 387base pairs (bp) from pork species. The species-specific PCR wasused for successful identification of pork DNA, but the performanceof the method in terms of sensitivity was not reported (Che Man,Aida, Raha, and Son, 2007).

A similar method utilising PCR–RFLP was reported for beef, pork,buffalo, quail, chicken, goat, and rabbit species identification andHalal authentication. PCR products of 359-bp were successfullyobtained from the cytb gene of these meats, and five different specificenzymes were identified as potential restriction endonucleases fordifferentiation purposes (Murugaiah et al., 2009). Specific PCR ampli-fication of a repetitive DNA sequence has been used for the identifica-tion of pork in processed and unprocessed food. A level of addition of1% pork was detectable with 20 PCR amplification cycles and 0.005%pork with 30 PCR amplification cycles (Calvo, Zaragoza, and Osta,2001). A species-specific duplex PCR assay has been used for the si-multaneous detection of pork and poultry meat species, again usingthe mt cytb as target gene for pork. By amplification of DNA frommeat mixtures of two species, linear calibration was obtained usingfluorescence intensities of PCR products for pork (149-bp) in therange of 1–75%, with a sensitivity of 0.1% addition. In-house valida-tion, using samples with known amounts of pork, gave a coefficientof variation from 4.1 to 7.6% (Soares, Amaral, Isabel Mafra, Oliveira,and Beatriz, 2010). Real-time PCR has also been used for the identifi-cation of beef, pork, horse, mutton, chicken and turkey in processedmeat down to a level of 0.01–0.05% (Jonker, Tilburg, Gele, and DeBoer, 2008).

TaqMan real-time PCR using a bovine-specific primer pair for themt cyt b gene and a FAM-labelled mammalian-specific cyt b probecould quantitatively detect as little as 35 pg bovine DNA and showedno cross-reaction with ovine, caprine or porcine DNA. The systemwasused to measure bovine DNA in fresh and processed meat, milk andcheese (Zhang et al., 2007). Specific primers and TaqMan probeshave been designed for the mt ND2, ND5 and ATP 6–8 genes for don-key, pork and horse, respectively. Only one cross-reaction was ob-served between the horse species specific primer-probe set and100 ng pork DNA at the cycle threshold (Ct) value of 33.01 (corre-sponding to 0.01 ng horse DNA). The assay enabled the detection of0.0001 ng of template DNA from pure meat for each species investi-gated (Kesmen, Gulluce, Sahin, and Yetim, 2009).

Several species-specific PCR methods have been published to de-termine the origin of raw materials used in gelatine manufacture. Abovine species-specific PCR primer set targeting the ATPase 8 sub-unit gene in bovine mt DNA was demonstrated to be suitable for de-tection of bovine material in gelatine. This PCR primer set was opti-mised using conventional and real-time PCR approaches. Theinclusion of bovine gelatine in pork or fish gelatine could be detectedat levels of 0.1% by conventional PCR and 0.001% by light cycler PCRafter DNase I decontamination (Tasara et al., 2005). The viability oftesting pure gelatine by PCR was demonstrated, although the methodwas not taken any further in terms of analysis of commercial gelatine-containing food products.

In this paper we have deliberately adopted the approach of usingcommercial test kits both for DNA extraction and for the real-timePCR analysis. Although, there have been many very successfulmethods published for detection of porcine DNA, there is a realneed for food control laboratories to apply these methods routinelyusing commercially available kits. We have focussed on gelatine be-cause this product seems to have been overlooked in terms of testingmethodology, and yet has a high potential for inadvertent adultera-tion with porcine material or mislabelling.

2. Materials and methods

2.1. Sample preparation

Twelve gelatine samples of known origin (bovine, porcine or sea-weed) were obtained in powder or sheet form, and employed as ref-erence standards. Pure gelatine mixtures were prepared by extractingand purifying the DNA from 500 mg of porcine gelatine and dilutingthe resulting DNA solution with bovine DNA solution to obtain 10%,1% and 0.1% mixtures. Mixtures of food products containing porcinegelatine were individually prepared by grinding gum drops or marsh-mallow together with porcine gelatine, and mixing to a fine powder.The composition of these mixtures is shown in Table 1.

Forty-three samples of the soft and fruity chew confectionery(gum drops), Turkish delight, jelly and marshmallows/cakes contain-ing gelatine were obtained from markets in Turkey and Germany.Samples were stored at −20 °C. Approximately, 300 g of each samplewas blended in the frozen state using a Waring blender (Torrington,USA) to produce a powder, which was thoroughly mixed. Sub-samples (400 mg) were taken for DNA extraction.

Spiking of the above retail food products with 5% porcine gelatinewas carried out by weighing a 380 mg amount of the composite prod-uct into a 1.5 ml reaction tube together with 20 mg of porcine gela-tine. The whole mixture (400 mg) was then taken for DNA extraction.

2.2. Extraction of DNA

DNA was extracted from pure gelatine or from food products con-taining gelatine (400 mg) using the Sure Food® Prep Animal X kit(CONGEN, R-Biopharm, Germany). Lysis buffer (1000 μl) and Protein-ase K (40 μl) were added to 400 mg of sample andmixed by vortexing(Fisherbrand ZX Wizard). The mixture was incubated at 65 °C for1 hour in a thermomixer (Eppendorf, comfort) under continuousshaking. At the end of the incubation, the solution was centrifugedat 24,150g for 2 min (Eppendorf 5430). After centrifuging, a spin filterwas placed in a receiver tube. The solution was transferred into spinfilter and centrifuged at 24,150g for 2 min. The spin filter was dis-carded. Binding buffer (200 μl) was added to the filtrate, which wasvortexed thoroughly. The filtrate was transferred to a new spin filterplaced in a new receiver tube and centrifuged again at 24,150g for2 min. After the filtrate was discarded, 550 μl of pre-wash bufferwas added into the spin filter and centrifuged at 24,150g for 1 min.This step was repeated twice. After discarding the filter, it was centri-fuged for 2 min at 24,150g to remove wash buffer completely. A newspin filter was placed in a new 1.5 ml receiver tube; 50 μl of pre-heated elution buffer was pipetted directly onto the spin filter and in-cubated at room temperature for 3 min. Finally, it was centrifuged for2 min at 16,770g and the purified DNA solution (50 μl) was stored at4 °C.

2.3. PCR amplification

A pork reaction mixture containing specific primers and Taq-Polymerase are supplied as part of the commercial test kit. The reac-tion mix, Taq-Polymerase (SureFood® Animal ID Pork SENS Plus Vkit) and extracted DNA were mixed in the ratio 9.95: 0.05: 2.5 for

Table 2Ct values of 100%, 10.0% and 1.0% pork gelatin. Results are shown for 6 replicate assays.

Ct values 100%pork gelatine

Ct values 10.0%pork gelatine

Ct values 1.0%pork gelatine

Ct values 0.1%pork gelatine

28.07 32.09 33.55 –

28.57 32.29 41.89 –

28.44 32.48 37.04 –

28.34 32.21 36.14 –

28.87 33.92 41.5928.43 31.63 37.32 –

Ct—cycle threshold value.– Indicates that porcine gelatine was not detectable at the % level of addition.

688 Y. Demirhan et al. / Meat Science 90 (2012) 686–689

dispensing for replicate analysis. For each reaction a total of 25 μl ofthe above mixture was dispensed (containing 19.9 μl pork reactionmix, 0.1 μl Taq-Polymerase (SureFood® Animal ID Pork SENS Plus Vkit) and 5 μl extracted DNA). Amplification was performed with areal-time PCR (Eppendorf, Mastercycler ep Realplex2). The thermalcycler followed the programme of initial denaturation at 95 °C for5 min to denature the DNA template completely, followed by 45 cy-cles of denaturation at 95 °C for 15 s, and annealing and extensionat 55 °C for 30 s. The proprietary computer software is designed to ex-amine the fluorescent intensity at 522 nm (FAM) and 553 nm (VIC). Anegative control was included in every run. A sample was deemed tobe pork positive, if the sample DNA showed amplification in the FAMchannel. A sample was deemed to be negative, if the sample DNAshowed no amplification in the detection system and the internal am-plification control (inhibition control) of the sample was positive.

2.4. Determination of the sensitivity of the assay

The sensitivity of the assay was measured in terms of both detec-tion of porcine DNA in pure gelatine and detection of porcine DNA infood products containing gelatine. Replicate real-time PCR measure-ments were made of extracted DNA from gelatine spiked with 0.1,1, and 10% and the limit of detection was taken as being the lowestamount that could be amplified with a reproducible Ct value. A simi-lar approach was adopted to determine the sensitivity for the detec-tion of porcine gelatine spiked into samples of gum drops andmarshmallows, which had been found to be negative by real-timePCR.

2.5. Determination of the reliability of the assay

The reliability of the assay was tested by spiking samples of gumdrops and marshmallows that were established as not containingany porcine DNA. For gum drops, spiking was carried out at 0.0%,0.5%, 3.0% and 5.0% addition of porcine gelatine and for marshmal-lows, addition was made at levels of 0.0%, 0.1% 0.5%. 3.0% and 5.0%.All samples, including unspiked gum drops and marshmallows wereanalysed as five blind replicates making a total of 45 determinations.

2.6. Analysis of survey samples

Details of all survey samples from Germany and Turkey wererecorded and the packaging was retained for reference. All sampleswere analysed in replicate and also spiked with 5% porcine gelatine,following the finalised protocol set out above.

Fig. 1. PCR amplification plots for samples of gelatin

3. Results and discussion

3.1. Detection of porcine DNA in pure gelatin

In the first set of experiments, two samples of gelatine, one desig-nated as bovine and another as Halal, were tested and both confirmedto be negative for pork by real-time PCR. In contrast, two samplessupplied as being porcine gelatine were both found to be clearly pos-itive, and a mixture of 10.0% porcine in bovine gelatine was also clear-ly positive. Samples of porcine gelatin gave Ct values from 28.1 to28.8, whilst samples of gelatine containing 10.0% porcine gelatinegave Ct values from 31.6 to 33.9. These experiments carried out assix replicates clearly indicated that good quality DNA could be reliablyextracted from gelatine, and that the pork characteristic cytb-sequence could be sensitively amplified and detected.

Experiments were then conducted taking various mixtures of bo-vine DNA with additions of 10.0%, 1.0% and 0.1% porcine DNA (DNAfrom gelatine used in this study). These results are shown in Table 2where it can be seen that samples of bovine DNA containing 0.1% por-cine DNA were all found to be negative. However, at the 1.0% level ofaddition of porcine DNA, Ct values between 33.5 and 41.8 wereobtained. Although within the six replicates there was some variabil-ity, a 1.0% addition (w/w) could, nevertheless, be reliably detectedand this was thus taken to be the effective limit of detection. TypicalCt curves for detection of porcine DNA at different levels, by real-time PCR are shown in Fig. 1. Thus, for any sample of gelatine desig-nated as Halal, if any cross-contamination of rawmaterials such as in-corporation of some pig skins had occurred during manufacture, thetest would be sufficiently sensitive to detect 1.0% adulteration. Al-though this commercial test kit appears to be less sensitive than thebeef method reported by Tasara et al. (2005), this could be due tothe different gelatine types used and the differences in the extractionmethods. In reality high sensitivity is not really a necessity for Halal/Kosher testing and a 1.0% cut-off provides a pragmatic limit fortesting.

e containing 100%, 10.0% and 1.0% pork gelatin.

Table 3Ct values for different amounts of porcine gelatine spiked into marshmallows and gum drops. Results are shown for 5 replicate assays.

Addition porcine gelatine (%) Ct values porcine gelatine in marshmallow Addition porcine gelatine (%) Ct values porcine gelatine in gum drops

0 – – – – – 0 – – – – –

0.1 – – – – – 0.1 – – – – –

0.5 36.86 37.63 36.13 38.92 37.76 0.5 – – – – –

3.0 32.62 33.39 33.59 34.08 31.26 3.0 35.78 34.41 37.74 37.18 37.745.0 24.08 28.16 29.12 29.24 29.27 5.0 – 29.10 31.89 34.41 32.44

Ct—cycle threshold value.– Indicates that porcine gelatine was not detectable at the % level of addition.

Table 4Results of survey of samples obtained from Turkey and Germany shown as numbers ofpositive and negative samples.

Products Total no. ofsamples

Porcine negative Porcine positive

Turkey Germany Turkey Germany Turkey Germany

Marshmallow/cake 17 – 16 – 1 –

Gum-drops 11 11 11 9 – 2Jelly 3 – 3 – – –

Turkish delight 1 – 1 – – –

Total 32 11 31 9 1 2

689Y. Demirhan et al. / Meat Science 90 (2012) 686–689

3.2. Detection of porcine DNA in gelatin-containing food products

The sensitivity of the assay was also established for the detectionof porcine gelatine spiked into retail samples of gum drops andmarshmallows shown to be negative by real-time PCR. In both foodproducts, a 1.0% addition of gelatine could be detected, consistentwith the results for spiked gelatine. The reliability of the assay wasestablished from blind replicate analysis of negative gum drops andmarshmallows and blind analysis of replicate samples spiked at levelsabove and below the detection limits. No false positives weredetected, but one false negative sample was obtained giving a falsenegative rate of 2.0% for the assay. The results of these tests to estab-lish the reliability of the assay in real food products are shown inTable 3.

3.3. Survey of retail food products from Germany and Turkey

The results of a small survey of retail products are shown inTable 4. Of 11 retail products purchased in Germany, 2 sampleswere found to contain porcine gelatine, with the remaining 9 samplesbeing negative. The 2 positive samples had Ct values of 30.04 and43.00. Of a total of 32 samples from Turkey, 31 samples were foundto be negative. However, one product (cake covered with gelatine)from the Turkish retail market was found to have a Ct value of 36.3clearing indicating that this product contained porcine DNA. Thisproduct was therefore not Halal and the labelling failed to indicatethe use of porcine gelatine. Spiking of all samples with 5.0% porcinegelatine confirmed that the test was functioning adequately, althoughone false negative was detected for a sample that remained negativeeven after spiking. Whilst this survey was very limited in scope, theclear discrimination between positive and negative retail samples ofdiffering compositions shows its robustness. The detection of a sam-ple in Turkey, which was not Halal, clearly shows the need for furthersurveillance of retail gelatine-containing foods and possible regulato-ry action by the authorities.

4. Conclusions

Using a commercial DNA extraction kit and a commercial real-time PCR kit it has been demonstrated that porcine DNA can be reli-ably detected in gelatine at a level below 1.0% w/w. It has also beenshown that porcine DNA can be detected in a variety of gelatine-containing confectionary products at the levels of 1.0% w/w at afalse negative rate of 2.0% gelatine (all data for the gelatines used inthis study). In a small survey, one retail product from Turkey wasfound to contain porcine gelatine whereas the consumer expectationwas that this was a Halal product.

Acknowledgement

The authors gratefully acknowledge the support of CONGEN Bio-technologie GmbH (Berlin, Germany) and Sincer Dis Ticaret (Izmir,

Turkey) for the supply of SureFood® kits used for the developmentand validation work reported in this paper.

References

Aida, A. A., Che Man, Y. B., Raha, A. R., & Son, R. (2007). Detection of pig derivatives infood products for halal authentication by polymerase chain reaction–restrictionfragment length polymorphism. Journal of the Science of Food and Agriculture, 87,569–572.

Binke, R., Spiegel, K., & Schwägele, F. (2007). Identifizierung von tierischen Bestandtei-len in Fleischerzeugnissen mittels PCR. Fleischwirtschaft, 2(2007), 93–98.

Boran, G., & Regenstein, J. M. (2010). Chapter 5. Fish gelatin. Advances in Food and Nu-trition Research, 60, 119–143.

Calvo, J. H., Zaragoza, P., & Osta, R. (2001). Technical note: A quick and more sensitivemethod to identify pork in processed and unprocessed food by PCR amplificationof a new specific DNA fragment. Journal of Animal Science, 79, 2108–2112.

Che Man, Y. B., Aida, A. A., Raha, A. R., & Son, R. (2007). Identification of pork derivativesin food products by species-specific polymerase chain reaction (PCR) for halal ver-ification. Food Control, 18, 885–889.

Jonker, K. M., Tilburg, J. J. H. C., Gele, G. H. H. A., & De Boer, E. (2008). Species identifi-cation in meat products using real-time PCR. Food Additives and Contaminants, 25,527–533.

Karim, A. A., & Bhat, R. (2008). Gelatin alternatives for the food industry: Recent devel-opments, challenges and prospects. Trends in Food Science and Technology, 19,644–656.

Kesmen, Z., Gulluce, A., Sahin, F., & Yetim, H. (2009). Identification of meat species byTaqMan-based real-time PCR assay. Meat Science, 82, 444–449.

Mafra, I., Ferreira, I. M. P. L. V. O., & Oliveira, M. B. P. P. (2008). Food authentication byPCR-based methods. European Food Research and Technology, 227, 649–665.

Murugaiah, C., Noor, Z. M., Mastakim, M., Bilung, L. M., Selamat, J., & Radu, S. (2009).Meat species identification and Halal authentication analysis using mitochondrialDNA. Meat Science, 83, 57–61.

Riaz, M. N., & Chaudry, M. M. (2004). Halal food production. Baca Raton USA: CRC Press.Soares, S., Amaral, J. S., Isabel Mafra, I., & Oliveira, M. Beatriz P. P. (2010). Quantitative

detection of poultry meat adulteration with pork by a duplex PCR assay. Meat Sci-ence, 85, 531–536.

Tasara, T., Schumacher, S., & Stephan, R. (2005). Conventional and real-time PCR-basedapproaches for molecular detection and quantitation of bovine species material inedible gelatin. Journal of Food Protection, 68, 2420–2426.

Zhang, C. -L., Fowler, M. R., Scott, N. W., Lawson, G., & Slater, A. (2007). A TaqMan real-time PCR system for the identification and quantification of bovine DNA in meats,milks and cheeses. Food Control, 18, 1149–1158.