Development and utilization of analytical methods for rapid GM detection in processed food products: a case study for regulatory requirement

Genetically modified (GM) food crops for desired traits have been approved in some of the countries. The approval status of a GM event varies from country to country. In India, Bt cotton is the only approved GM crop. So far, no GM food crop has got regulatory approval in the country, which may be considered as unauthorized GM (UGM) event in the Indian context. The entry of UGM events of food crops needs to be checked in the imported consignments as well as in the marketplace. In the present study, screening elements were identified based on the genetically modified organism (GMO) matrix developed as decision support system for 22 GM food crops approved globally. Three sets of multiplex PCR assays were developed and validated for GM detection in food crops and products: triplex PCR targeting control elements [Cauliflower Mosaic Virus 35S promoter (P-35S), Figwort Mosaic Virus promoter (P-FMV), nopaline synthase terminator (T-nos)], triplex PCR targeting marker genes [aminoglycoside-3’-adenyltransferase (aadA), neomycin phosphotransferase (nptII), phosphinothricin-N-acetyltransferase (pat)] and duplex PCR targeting Bt genes (cry1Ab/Ac and cry2Ab2). Limit of detection (LOD) ranged from 0.5–0.01% for different targets, which is in compliance with the labelling threshold of many countries. The developed assays were utilized to check the GM status of apple and maize products along with an additional test for ctp2-cp4epsps for herbicide tolerance in maize. Based on the tests conducted, none of these samples were found positive for GM targets screened. These procedures could be efficiently employed as a part of GMO testing to trace GM contamination, if any, in both the imported as well as domestic food products.


Introduction
With the approval of first genetically modified (GM) food crop FLAVR SAVR™ tomato with delayed fruit ripening in 1990s, a number of GM food crops and derived products have got approval presently in some of the countries such as the United States and Canada. Since last 25 years, GM food Aparna Paliwal and Kushaldeep Kaur contributed equally to this work.
crops have been developed for imparting resistance to insect pests, to better withstand salinity, drought and other abiotic stresses, and for improving different characteristics including color, flavor, texture and shelf life of foods along with the nutritional enhancement. The approval of GM food products is regulated in different countries by respective regulatory bodies. Labelling of genetically modified organism (GMO) is voluntary in some of the countries; whereas it is mandatory in several countries as in the European Union (EU) with labelling threshold of 0.9% Querci et al. 2010; https:// ec. europa. eu/ jrc/ en/ resea rch-topic/ gmos).
In India, Bt cotton is the only approved GM crop for planting and GMO labelling threshold has not yet been implemented. In the country, regulation of GMO and GMderived products is being implemented under the "Rules for the manufacture, use, import, export and storage of hazardous microorganisms, genetically engineered organisms or cells, 1989" notified under the Environment (Protection) Act, 1986. In accordance with the Food Safety and Standards Authority of India (FSSAI) Order No. 1-1764/FSSAI/ Imports/2018 (Part-1) dated 21 August 2020 and 1-1764/ FSSAI/Imports/2019 (Part-1) dated 08 February 2021, every imported consignment of 24 selected crops should be accompanied with the non-GMO certification effective from 1 March 2021 and the tolerance limit of adventitious presence of GMOs at 1 per cent is permissible in the consignments of imported food crops (FSSAI 2021).
Many food products are being imported in India from the countries where the GM events of respective food crop are approved for use as food or feed. Therefore, it is necessary to check the unauthorized entry of GM products in the supply chain as well as in the marketplace as GM food products are regulated in the country. Systematic GMO testing is essential for checking the GM status of food products for verification for their non-GM status.
DNA-based methods are preferred over protein-based methods for GMO testing of processed food products or derivatives as there may be the chances of degradation of proteins. However in processed food products, quality of DNA may be hampered during processing steps or due to exposure to high temperature and pressure. Addition of certain ingredients or flavors may also considerably affect the quality of DNA, thereby inhibiting the amplification (Ramos-Gómez et al. 2014). Therefore, DNA extraction has been considered as a crucial step for GMO testing in food derivatives. DNA extraction methods from food products using chemical, thermal, enzymatic, or mechanical lysis, or a combination of these, or through use of column-based purification have been reported (Rantakokko-Jalava and Jalava 2002; Reischl et al. 2000;Van Tongeren et al. 2011;Singh et al. 2021).
Selection of common targets employing GMO matrix approach could facilitate cost-and time-efficient GMO screening by eliminating the need to test for every possible GM event of a particular crop (Waiblinger et al. 2010;Randhawa et al. 2014). The selected targets can be tested using polymerase chain reaction (PCR), multiplex PCR or real-time PCR assays. Multiplex PCR enables simultaneous detection of multiple targets in a test sample. Hexaplex PCR simultaneously detecting six marker genes was developed to check the GM status of seeds (Randhawa et al. 2009). GMO matrix combined with multiplex PCR was reported for GMO screening of globally approved GM events of cotton and maize .
Few reports on the utilization of GM diagnostics for GMO testing of food products from the markets in Brazil, Georgia and Turkey are available (Avsar et al. 2020;Cardarelli et al. 2005;Datukishvili et al. 2015). Cardarelli et al. (2005) checked the GM status of maize and soybean derived food products including raw soybeans, dehydrated and canned soups, powdered soy milk, cookies, french fries, sausages, products of maize meal, canned corn, soy beverages, soy protein, pasta, and pet-food, procured from the Brazilian market using PCR assays for Roundup Ready soybean, Bt176 and MON810 events of GM maize, and some of the soybean derived products were found GM. Datukishvili et al. (2015) utilized multiplex PCR simultaneously targeting Cauliflower Mosaic Virus 35S promoter (P-35S), nopaline synthase terminator (T-nos), 5-enol pyruvylshikimate-3-phosphate synthase (cp4epsps), for GM detection in soy, and targeting P-35S and cry1Ab for GM detection in maize food products. Avsar et al. (2020) checked the presence or absence of GM ingredients in 123 food and feed samples derived from maize, rice, soybean and wheat from retail shops in Turkey. The GM status of samples was checked by the multiplex realtime PCR simultaneously detecting P-35S, T-nos, P-FMV and bar gene, None of the wheat or rice products were found GM positive whereas, several maize and soybean derived samples were found positive for GM elements tested.
In the present study, based on the information available in the GMO databases, common elements were identified for screening of 80% of globally approved GM events (> 130 single or non-stacked GM events) of 22 food crops. Published matrix of maize was considered. Based on the GM coverage and ease of reactions, three sets of multiplex PCR assays were developed and validated for acceptable specificity and limit of detection (LOD) ranging from 0.5-0.01% for different assays/ targets. These assays were utilized to test the GM status of food derivatives of apple and maize collected from the retail shops and markets.

Food products/derivatives
Total of 14 food products of maize and apple were analyzed using developed multiplex PCR assays. As GM maize has been approved for consumption as food or feed in many countries, the samples of six food products were used, which included Corn flour (Brown & Polson), Natural Roasted Puff (Cool Crunchies), Corn Chips (Makino). Masala Munch (Kurkure), Sweet Corn Vegetable Soup (Knorr), and Plain Flour All Purpose (Organ). GM apple has been approved in the United States and Canada, therefore, five samples of apple fruit and three samples of apple products, viz., Apple Pie Cookies (Merba), Puffs Cereal Snack (Strawberry Apple) (Gerber), and Apple Fruit Drink (Amul Tru) were also subjected to the GMO testing. The samples were purchased from the retail shops/ grocery stores/supermarkets based at Delhi or nearby location.

Sample preparation
Genomic DNA from CRM/reference material (RM) was extracted using DNeasy® Plant Mini Kit (Qiagen, Hilden, Germany) protocol (as per the manufacturer's instructions). The contents of whole packet of each food product were homogenized to fine powder for representing whole sample using a Knife Mill Grindomix . In case of apple fruit drink, the contents of the packet was centrifuged at 12,000 rpm for 30 min (centrifugation was done in steps in 50-ml centrifuge tube). After centrifugation, supernatant was discarded and pellet was used for DNA extraction. DNA from food derivatives was extracted using modified CTAB method or DNeasy® Mericon Food Kit (Qiagen) protocol (Singh et al. 2021). Quantity and quality of DNA extracts were estimated using a UV Spectrophotometer (Eppendorf, Hamburg, Germany). DNA extracts were diluted to a concentration of 40 ng/μl for further use. Before method validation, DNA extracts of CRM/RM were validated using respective event-or genespecific protocols.

GMO matrix, selection of targets and multiplex PCR assays
The screening targets for GM detection in 22 GM food crops including alfalfa, apple, Argentina canola, common bean, chicory, cowpea, eggplant, flaxseed, melon (Cucumis melo), papaya, Polish canola, plum, potato, rice, safflower, soybean, squash, sugarbeet, sugarcane, sweet pepper, tomato and wheat, were identified employing GMO matrix approach based on the information compiled from open access GM crop databases, namely, http:// bch. cbd. int/ datab ase/ organ isms, http:// www. isaaa. org/ gmapp roval datab ase/. Selected genetic elements, which can cover most of the GM events of these crops are shown in Supplementary Information 1 and Fig. 1. Previously reported crop-specific GMO matrix was referred for GM maize . Triplex PCR targeting two promoters (P-35S, P-FMV) and a terminator (T-nos); triplex PCR targeting three marker genes (aadA, nptII, pat) and duplex PCR targeting cry1Ab/Ac and cry2Ab2 transgenes were developed and validated. Details of primers used are given in Table 1.
Multiplex PCR was performed in Veriti® Thermal Cycler (Applied Biosystems, Foster City, CA, USA) in 20 µl volume with the composition as follows: The cycling profile comprised of initial denaturation at 94 °C for 5 min followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 60 °C for 50 s, and extension at 72 °C for 1 min; and a final extension at 72 °C for 7 min. The PCR products were analyzed on 4% (w/v) metaphor agarose gels or on 2% (w/v) agarose gels in 1 × TAE buffer, and visualized using UV Gel Imaging System (Alpha Innotech, USA).

GMO testing of food derivatives
Amplifiability of DNA extracts of food products was confirmed by PCR targeting endogenous Adh1 gene of maize and for apple products, PCR assay targeting chloroplast tRNA (cp-tRNA) gene was employed (Mazzara et al. 2013;Taberlet et al. 1991). All the three sets of multiplex PCR were used to test the GM status of food derivatives using same PCR conditions as mentioned in method development and validation. All the samples were tested in replications with appropriate positive and negative controls to reduce the chance of errors due to false negatives or positives. An additional PCR test targeting commonly employed ctp2-cp4epsps region for herbicide tolerance was conducted in case of maize products.

Results and discussion
DNA-based methods, including PCR and real-time PCR may be preferred over protein-based methods for GM detection in food products due to stability of DNA. Commonly present transgenic elements such as P-35S and T-nos are the targets of choice for GMO screening (Lipp et al. 2001;Waiblinger et al. 2010;Holden et al. 2010;Bak and Emerson 2019). However, the inclusion of more screening targets with respect to approved GM events of a particular crop, may enhance the GM coverage in GM detection, for which GMO matrix approach can be applied for selection of suitable targets. In the present study, multiplex PCR assays targeting common transgenic elements identified in GM food crops approved globally using matrix approach were developed and validated. These assays were adopted for preliminary GMO screening in food derivatives of maize and apple as an initiative for GMO testing of food products in India. The screening assays give a fair idea for the GMO status of a sample and further confirmation could be done using eventspecific tests in case of positive screening results. imported for research purposes. Ten targets, namely, P-35S, cry1Ab,cry1Ac. cry1C,cry2Ab, GA20 oxidase1, nptII, bar were identified for screening of these events.

Selection of screening targets
In accordance with the FSSAI (2021), 24 food crops have been identified for assurance of non-GM status in the imported consignments. Among these, information of more than 130 GM events 22 GM food crops including alfalfa, apple, Argentina canola, common bean, chicory, cowpea, eggplant, flaxseed, melon, papaya, plum, Polish canola, potato, rice, safflower, soybean, squash, sugarbeet, sugarcane, sweet pepper, tomato and wheat, was compiled from the GM crop databases. While analyzing the GMO matrix, only non-stacked GM events were considered as the genetic information of stacked GM events is derived from individual events comprising the stack. Nine elements were identified, which can detect 80% of globally approved GM events of 22 GM food crops, as shown in Fig. 1 and Supplementary Information 1. For maize, previously reported crop-specific GMO matrix was referred  where more than 90% of globally approved GM events could be covered using the selected targets. Based on the elements selected, triplex PCR targeting two promoters (P-35S, P-FMV) and a terminator (T-nos); triplex PCR targeting three marker genes (aadA, nptII, pat) and duplex PCR targeting cry1Ab/Ac and cry2Ab2 transgenes were developed and validated. Assay for ctp2-cp4epsps was also selected for GMO teating of specific food crops with Roundup Ready® trait such as maize. One of the crops, namely, pineapple included in FSSAI (2021) was not covered with these elements and single approved event can be individually tested for this.
As the event-specific assays are restricted to identification of particular GM event so testing a sample for all the possible GM events is cumbersome and time-consuming. GMO screening methods could reduce the time and cost of GMO testing by eliminating the need of testing for all the GM events (Randhawa et al. 2014;Waiblinger et al. 2010). Multiplex PCR-based screening assays were reported by our laboratory to detect more than 90% of globally approved GM events of cotton and maize ).
In the present study, based on the common screening targets identified in 22 GM food crops, three sets of multiplex PCR assays were developed and validated. While optimizing the multiplex PCR conditions, choice of DNA polymerase was considered as a key parameter for optimum PCR performance to avoid the formation of non-specific products and primer-dimer for simultaneous detection of targeted products without any primer interference (Randhawa et al. 2009). In this study, the ready-to-use master mix based on Hot Start DNA polymerase was used, Triplex PCR, simultaneously targeting control elements, i.e., 123 bp region of P-35S, 78 bp region of P-FMV and 180 bp region of T-nos, was developed. Another triplex PCR targeting three marker genes, i.e., 406 bp region of aadA, 262 bp of pat, and 215 bp of nptII was optimized and validated. Bt gene specific duplex PCR targeting cry1Ac/Ab and cry2Ab2 was developed, which targeted 74 bp region of cry1Ac/Ab and 260 bp region of cry2Ab2.
Specificity test of duplex PCR targeting cry1Ab/Ac and cry2Ab2 genes showed amplification of expected product of 74 bp for cry1Ab/Ac in the targets, namely, MON15985, Cot102 × MON15985, MON15985 × MON88913, Cot102 × MON15985 × MON88913 and MON531 events of Bt cotton, and Bt11 and Bt176 events of Bt maize, whereas no amplification was detected in respective nontargets and negative control (Fig. 2c). Product of 260 bp for cry2Ab2 gene was detected in the samples of MON15985, Cot102 × MON15985, MON15985 × MON88913, Cot102 × MON15985 × MON88913 and MON89034 and no amplification was detected in rest of samples used as non-targets (Fig. 2c), The Bt gene specific assay was found sensitive to detect as low as 0.01% for each target (Fig. 3c).
The results indicated that the developed multiplex PCR assays showed acceptable specificity over the range of targets and non-targets analyzed ( Fig. 2 and Table 2). The established LOD of developed assays was up to 0.5-0.01% depending on the targets (Fig. 3), which is in compliance with the labelling threshold of many countries including 0.9% in the EU (https:// ec. europa. eu/ jrc/ en/ resea rch-topic/ gmos), or the tolerance limit of adventitious presence of GMOs permissible in the imported consignments as in the case of India (FSSAI 2021). The LOD of different multiplex PCR sets may vary due to level of multiplexing or size of the amplicons. For example, in case of triplex PCR, LOD was 0.5-0.1 whereas in duplex PCR for Bt genes, detection limit was higher, that is, 0.01%. On comparison of triplex PCR assays, difference in the LOD may be attributed to the size of amplicons. More the size may require more template to bind and therby contributing to lower detection limt. These procedures could be efficiently employed as a part of GMO testing to trace GM contamination, if any, in both the imported as well as domestic food products.

Utilization of developed assays for GMO testing of food derivatives
Few reports are available on the utilization of GM diagnostics for checking the GM status of food products from the markets in Brazil, Georgia and Turkey (Avsar et al. 2020;Cardarelli et al. 2005;Datukishvili et al. 2015). Very recently, our laboratory has conducted a study for checking the unapproved GM event CDC Triffid flax using GM  (Singh et al. 2022). As an initiative for GMO testing of food products in India in a systematic manner, in the present study, developed assays were utilized to check the GM status of food products of apple and maize.
A set of 14 food products purchased from local markets or retail shops were allocated codes before subjected to testing (Table 3). DNA extraction procedures as optimized previously by our laboratory were used for most of the matrices (Singh et al. 2021). Quality of DNA from processed products may be hampered by the processing steps when subjected to high temperature and pressure, or presence of inhibitors due to complex composition and amplifiability may get hindered to the inhibitors or complex composition (Ramos-Gómez et al. 2014;Singh et al. 2021). To address this issue and to eliminate the chances of false negatives, quality and amplifiability of DNA extracts was first ensured using PCR assays targeting respective taxon-specific assay or plant-specific marker. The amplicon of expected size of 135 bp for Adh1 gene was detected in the DNA extracts of food samples of maize ( Fig. 4-i) and amplicon of ~ 615 bp for cp-tRNA gene was detected in the DNA samples of fruits and products of apple ( Fig. 5a-I, b), thereby confirming the amplifiability of DNA extracts.
Samples of food products of maize and apple were analyzed in replications and appropriate positive and negative controls were run along with the samples. On multiplex PCR assays, none of the tested samples of maize were found positive for the transgenic elements (P-35S, P-FMV, aadA,nptII,pat,cry1Ab/Ac and cry2Ab2) included in the assays (Fig. 4ii-iv). An additional PCR test for ctp2-cp4epsps was also conducted for more GM coverage in case of maize and none of the samples were found positive for this specific element (Fig. 4v). Amplification was detected in respective positive controls of each target, indicating the accurate working of the assay used. Based on the tests conducted using multiplex PCR assays, none of the samples of apple were found positive for the targeted transgenic elements .
All the samples were tested in two replications along with appropriate positive and negative controls for each assay to avoid type I (false positives) and type II (false negatives) errors. The tests were repeated twice by different analysts. Before GMO testing, the DNA samples were first tested for ensuring quality and amplifiability using endogenous gene specific assays: Adh1 for maize products and cp-tRNA for apple products, in order to minimize the chances of type II error. The acceptance criteria in accordance with the Definition of Minimum Performance Requirements for Analytical Methods of GMO Testing, European Network of GMO Laboratories were referred (JRC Technical Report 2015). In all the replications, none of the food samples were found positive for selected transgenic elements, whereas specific amplicon was detected in respective positive control for each target. The statistical interpretations confirmed that the GMO analysis results complied scientifically and not by any chance factor.
The results indicated that the developed diagnostics can be efficiently utilized to check the GM status of food products as a measure for controlled distribution and use of GM foods based on the approval status of particular GM food crop/product in a country.

Conclusions
More than 20 GM food crops have been approved in different countries (https:// www. isaaa. org/ gmapp roval datab ase/). The approval status of GM food crops in a country has demanded the need for checking the possible UGM event(s) in the derived food products of respective crops for regulatory compliance as well as for addressing consumers' apprehensions to an extent. So far, no GM food crop has been approved for commercial cultivation or for use as food in India, though several GM food crops have been developed or imported for research purposes or were under field trials (https:// geaci ndia. gov. in/ meeti ngs. aspx, . Bt eggplant event EE1 is a classical example, which has been developed by Maharashtra Hybrid Seeds Co. (Mahyco), Jalna and under field trials in the country, in spite of this the event has not got regulatory approval In view of existing regulation or policy for GM crops or derived products in India and other countries, three sets of multiplex PCR assays for GMO detection in food products were developed and validated with the LOD range of 0.5-0.01% depending upon the target, which fulfils the requirement of the labelling threshold of many countries or the tolerance limit of adventitious presence of GMOs permissible in the imported consignments. These assays targeted the common transgenic elements including P-35S, P-FMV, T-nos, aadA, nptII, pat, cry1Ab/Ac and cry2Ab2 that can be utilized for GM detection in food products derived from 22 food crops including alfalfa, apple, Argentina canola, common bean, chicory, cowpea, eggplant, flaxseed, maize, melon, papaya, Polish canola, plum, potato, rice, safflower, soybean, squash, sugarbeet, sugarcane, sweet pepper, tomato and wheat.
These procedures could be efficiently employed as a part of GMO analysis to trace the UGM contamination, if any, in both the imported as well as domestic food products.
Acknowledgements The authors thank the Indian Council of Agricultural Research (ICAR) and the Director, ICAR-National Bureau of Plant Genetic Resources for providing necessary facilities and resources.

Fig. 5
PCR profiles for utilization of developed GM diagnostics for checking the GM status of a apple fruits and b-d apple products. a (i) Checking amplifiability of DNA extracts of apple fruits using cp-tRNA gene specific assay; (ii-iv) PCR profiles for checking the GM status using developed multiplex PCR assays. b Checking amplifiability of DNA extracts of apple products using cp-tRNA gene specific assay. c-e (i-iii) PCR profiles for checking the GM status using developed multiplex PCR assays. Sample Details: AF1: Fruit (SMB), Data availability Sstatement Data sharing is not applicable to this article as no datasets were generated or analyzed. All the data is included and provided in the main text of the manuscript or in the Supplementary Information.

Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Disclaimer Any opinions, findings, or conclusions, expressed in this publication are those of the authors and do not necessarily reflect the view of the concerned organization.
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.