
==== Front
J Agric Food Chem
J Agric Food Chem
jf
jafcau
Journal of Agricultural and Food Chemistry
0021-8561
1520-5118
American Chemical Society

37500067
10.1021/acs.jafc.2c08966
Article
Filling Analytical Gaps in Allergen Detection—Real-Time PCR for the Detection of Commercially Relevant Cephalopods and Gastropods in Food
https://orcid.org/0000-0002-3237-5662
Blaschke Vincent *†
Berten Alea ‡
Sprenger Heike †
Zagon Jutta †
Winkel Matthias †
† Department of Food Safety, German Federal Institute for Risk Assessment (BfR), P.O. Box 330013, D-14191 Berlin, Germany
‡ Technische Universität Berlin, Institute for Food Technology and Food Chemistry, Gustav-Meyer-Allee 25, 13355 Berlin, Germany
* Email: vincent.blaschke@bfr.bund.de.
27 07 2023
09 08 2023
27 07 2024
71 31 1202912042
20 12 2022
06 06 2023
26 05 2023
© 2023 The Authors. Published by American Chemical Society
2023
The Authors
https://creativecommons.org/licenses/by-nc-nd/4.0/ Permits non-commercial access and re-use, provided that author attribution and integrity are maintained; but does not permit creation of adaptations or other derivative works (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Mollusks belong to the group of shellfish, which are considered to be among the elicitors of severe food allergies worldwide. In recent years, numerous PCR detection methods have been developed for other shellfish such as crustaceans. However, cephalopods and gastropods were not considered in the development of these shellfish detection systems. In this study, we have developed highly specific real-time PCR methods for the comprehensive detection of all commercially relevant cephalopod species and the gastropod families Helicidae, Buccinidae, and Muricidae in food matrices. In total, we cross-tested over 100 animal and plant species to show the specificity of our systems. The limit of detection (LOD12) was set at 1 pg of cephalopod and gastropod DNA or 10 ppm (mg/kg) spiked in a vegetarian food product. The robustness of the protocol was confirmed by testing multiple parameters while cooking and autoclaving of samples ensured the practical applicability of the systems.

Cephalopoda
Gastropoda
shellfish
allergen
Mollusca
PCR
Bundesministerium fÃ¼r ErnÃ¤hrung und Landwirtschaft 10.13039/501100005908 281A103016 document-id-old-9jf2c08966
document-id-new-14jf2c08966
ccc-price
==== Body
pmc1 Introduction

Despite modern medicine and the advances humanity has made in recent decades, food allergies still pose a major threat to human health. With a prevalence of 1–3% in adults and 4–6% in children in the general population,1 millions of people worldwide struggle with food allergies. Nevertheless, the current prevalence of food allergies cannot be fully determined due to differences in geographic and cultural food consumption habits.2,3

In particular, fish and shellfish allergies (to crustaceans, mussels, squids, and other mollusks) are considered as one of the most severe food allergies worldwide.4 The consumption of mollusks, such as cephalopods and gastropods, can cause allergic reactions in humans, ranging from rather mild symptoms to life-threatening anaphylactic shocks. For this reason, making food as safe as possible for all allergic patients is of high priority to most governments. Over the past decade, countries and confederations have gradually enacted regulations for the labeling of potential allergens in food to raise consumer awareness in their daily lives.5−7 To ensure and control correct food labeling, new detection methods need to be developed that are suitable for national and international standards. To cover all possible species that cause allergies, methods are required that include families and orders or even classes. Two possible targets are available for this purpose: the first is DNA-based and the second is protein-based. Protein-based methods often show deficiencies when dealing with highly processed foods. Heat and dry treatments denature proteins, which does not always mean a loss of allergenicity but hampers detection by standard protein-based methods, such as immunoblot and lateral flow assays.8,9 Mass spectrometry provides highly sensitive quantification of multiple proteins in food for a single analysis but is greatly affected by food processing.10 Furthermore, protein-based methods, such as the enzyme-linked immunosorbent assay (ELISA), which is a relatively simple and inexpensive method, have been shown to produce less-reliable results due to matrix interference, whereas DNA-based methods seem to have minor matrix effects.10,11 Unlike proteins, DNA is much more stable and can be used more efficiently for species identification in processed food. Therefore, DNA-based real-time PCR systems are much more frequently used to detect any kind of DNA traces resulting from food adulteration or cross-contamination during the production process.12,13

The class of cephalopods (Cephalopoda) consists of 876 species and about 3.7 million (M) metric tons (t) were caught in 2020.14 Currently, cephalopods are not farmed in large-scale aquaculture. Nearly all species caught belong to four orders composed of five families: Oegopsida (Ommastrephidae), Sepiida (Sepiidae and Sepiolidae), Myopsida (Loliginidae), and Octopoda (Octopodidae). Among them, the order Oegopsida accounts for more than 1.8 M t, followed by the order Myopsida (0.5 M t), Sepiida (0.4 M t), and Octopoda (0.4 M t). Due to ongoing food fraud and authenticity issues in the cephalopod food market, much work has been invested in the development of PCR systems to detect specific species or families of cephalopods.13,15−18 Almost all commercially relevant cephalopods can be detected with the available systems. However, these systems use multiple primer sets and, most importantly, not all of them use probes as required by international standards based on molecular food analysis to verify PCR results.19 In contrast to cephalopods, only a few analytical methods have been developed for gastropods in the past,20,21 although their allergenicity has been described in recent decades.22−24 In fact, to the authors’ knowledge, no universal gastropod method has been published yet for the scope of the investigation.

Compared to the Cephalopoda, the Gastropoda (snails) are a much larger class within the phylum Mollusca, comprising about 90,648 species.25 The FAO statistic only provides comprehensive data for global fisheries and aquaculture production of freshwater and sea snails, but not for terrestrial snails. The FAO only reports terrestrial snail production for four countries (Turkey, Tunisia, Morocco, and Ivory Coast) with a quantity of 18,100.14 However, European countries, such as France, Spain, and Italy, which also produce large quantities of snails, are not included in that data set.26,27 Unfortunately, there is no credible scientific source in the literature that could be used to estimate the current production and consumption of terrestrial snails worldwide. In 2020, 740,519 t of freshwater and sea snails were caught or produced in aquaculture.14 Just three families account for more than 75% of the worldwide freshwater and sea snail consumption with about 550,000 t. These three belong to the Muricidae (murex snails), with the important genera Rapana and Bolinus, the Haliotidae (abalone snails), and the Buccinidae (whelk snails).14 To address the existing gaps in the detection of mollusks in food, we focused on the development of a cephalopod and gastropod real-time PCR system. In our study, we investigated multi-copy mitochondrial ribosomal RNA genes as marker genes, since they are frequently used in phylogenetic studies and are useful in the development of large-group spanning PCR methods for the purpose of food investigation.28−31 After evaluating existing sequence data, we focused on the mitochondrial 12S rRNA and 16S rRNA genes. With the 16S rRNA gene real-time PCR system for cephalopods, we have developed a system, which can detect all commercially relevant species of the class Cephalopoda. For the duplex gastropod system, we used the 12S rRNA and 16S rRNA genes to detect the commercially important species of the Helicidae, Buccinidae, and Muricidae.

2 Material and Methods

2.1 Samples

The samples used in this study were provided by the reference material collection of the National Reference Laboratory for Animal Proteins in Feed at the German Federal Institute for Risk Assessment (BfR), Berlin. Additional samples were either purchased from local supermarkets, ordered from internet stores, or provided by FRoSTA AG (Bremerhaven, Germany) as the partner of the German national AQUALLERG-ID research project funded by the Federal Ministry of Food and Agriculture.32 Samples used in this study are listed in Table 1.

Table 1 Results of the Specificity Testinga

 	singleplex system (10 ng DNA)	duplex system (10 ng DNA)	
 	Loligo 16S	Heli/Bucci 12S	
Cephalopoda	 	 	
Octopoda (order)	 	 	
Octopus membranaceus	18.32 ± 0.06	undet.	
Octopus vulgaris	21.01 ± 0.11	undet.	
Octopus maya	21.46 ± 0.18	38.92 (1/3)	
Amphioctopus aegina	18.38 ± 0.27	undet.	
Eledone moschata	17.85 ± 0.16	undet.	
Oegopsida (order)	 	 	
Todarodes pacificus	17.72 ± 0.29	undet.	
Illex argentinus	20.27 ± 0.08	undet.	
Dosidicus gigas	16.11 ± 0.11	n.m.	
Sepiida (order)	 	 	
Sepia officinalis	20.69 ± 0.05	37.33 (1/3)	
Sepia lycidas	17.14 ± 0.09	undet.	
Sepiella japonica	18.61 ± 0.12	undet.	
Myopsida (order)	 	 	
Loligo reynaudii	22.51 ± 0.09	undet.	
Doryteuthis gahi	22.77 ± 0.16	undet.	
Uroteuthis chinensis	18.81 ± 0.08	undet.	
Uroteuthis duvaucelii	18.69 ± 0.03	undet.	
Gastropoda	 	 	
Stylommatophora (order)	 	 	
Helix pomatia	undet	20.38 ± 0.04	
Helix lucorum	undet	18.69 ± 0.06	
Cornu aspersum	n.m.	17.31 ± 0.14	
Lissachatina fulica	undet	36.74 ± 0.18	
Littorinimorpha (order)	 	 	
Littorina littorea	undet.	undet.	
Architaenioglossa (order)	 	 	
Pila virescens	n.m.	37.42 ± 1.11	
Neogastropoda (order)	 	 	
Buccinum undatum	undet.	17.53 ± 0.09	
Buccinum humphreysianum	n.m.	21.87 ± 0.01	
Bolinus brandaris	n.m.	23.86 ± 0.01	
Lepetellida (order)	 	 	
Abalone (Haliotidae spp.)	n.m.	undet.	
Bivalvia	 	 	
Perna canaliculus	undet.	undet.	
Mytilus chilensis	undet.	undet.	
Placopecten magellanicus	37.23 (1/3)	undet.	
Mizuhopecten yessoensis	36.48 ± 0.73	undet.	
Pecten maximus	undet	undet.	
Ruditapes pihilippinarum	undet.	undet.	
Callista chione	undet.	undet.	
Ensis directus	undet.	undet.	
Cerastoderma edule	undet.	undet.	
Crassostrea gigas	undet.	undet.	
Ostrea edulis	undet.	undet.	
Meretrix lyrate	undet.	undet.	
Mimachlamys varia	undet.	undet.	
Crustacea	 	 	
Crangon crangon	undet.	undet.	
Eriocheir sinensis	undet.	undet.	
Nephrops norvegicus	undet.	undet.	
Pandalus borealis	undet.	undet.	
Paralithodes camtschaticus	undet.	undet.	
Pleoticus muelleri	undet.	undet.	
Varuna litterata	undet.	undet.	
Xiphopenaeus kroyeri	undet.	undet.	
Cancer pagurus	undet.	n.m.	
Homarus americanus	undet.	undet.	
Macrobrachium rosenbergii	undet.	undet.	
Clerax quadricarinatus	undet.	undet.	
Penaeus monodon	undet.	undet.	
Crab Rillettes	undet.	undet.	
Lobster confit	undet.	undet.	
Seafood	 	 	
Sea urchin paste (Paracentrotus lividus)	undet.	undet.	
Sauce with squid ink	27.78 ± 0.14	n.m.	
Rice dish with octopus	21.41 ± 0.05	n.m.	
Linguine with squid ink	32.1 ± 0.23	n.m.	
Buttered jumbo squid rings (mix)	22.75 ± 0.05	n.m.	
Buttered jumbo squid rings (crust)	32.82 ± 0.36	n.m.	
Cuttlefish ball (Sepia pharaonis)	17.63 ± 0.09	undet.	
Paella, Frutti di Mare (Costa)	30.53 ± 0.26	n.m.	
Rice noodles with snail extract	n.m.	undet.	
Fish	 	 	
Melanogrammus aeglefinus	undet.	undet.	
Merluccius merluccius	undet.	undet.	
Sebastes marinus	undet.	undet.	
Clupea harengus	undet.	undet.	
Sardinia pilchardius	undet.	undet.	
Oncorhynches mykiss	undet.	undet.	
Anguilla Anguilla	undet.	undet.	
Salmo salar	undet.	undet.	
Cyprinus carpio	undet.	undet.	
Scomber scombrus	undet.	undet.	
Squalus acanthias	undet.	undet.	
Sparus aurata	undet.	undet.	
Oreochromis niloticus	undet.	undet.	
Thunnus albacares	undet.	undet.	
Insects	 	 	
Locusta magratoria	undet.	undet.	
Drosophila hydei	undet.	undet.	
Hermetia illucens	undet.	undet.	
Musca domestica	undet.	undet.	
Callosobruchus maculatus	undet.	undet.	
Acheta domesticus	undet.	undet.	
Alphitobius diaperinus	undet.	undet.	
Tenebrio molitor	undet.	n.m.	
Bombyx mori	n.m.	undet.	
Galleria mellonella	undet.	undet.	
Gryllus assimilis	n.m.	undet.	
Livestock	 	 	
lamb	undet.	undet.	
duck	undet.	undet.	
goat	undet.	undet.	
pork	undet.	undet.	
horse	undet.	undet.	
rabbit	undet.	undet.	
plants	 	 	
wheat	undet.	undet.	
rye	n.m.	undet.	
oat	n.m.	undet.	
rice	undet.	undet.	
corn	undet.	n.m.	
soy	undet.	undet.	
lupine	undet.	n.m.	
peanut	undet.	undet.	
walnut	undet.	undet.	
celery	undet	n.m.	
mustard seeds	undet.	undet.	
apple	undet.	n.m.	
fennel	undet.	undet	
macadamia	undet.	undet.	
nutmeg	undet.	n.m.	
cloves	undet.	n.m.	
bell pepper	undet.	n.m.	
brazil nut	undet.	undet.	
almond	undet.	undet.	
barley	n.m.	undet.	
instant meal	 	 	
vegetable pan—style—Asia curry	undet.	undet.	
procarionts	 	 	
Arthrospira platensis (Spirulina)	n.m.	undet.	
NTC	undet. in all experiments	
a Shows the mean CT-values from triplicates ±mean deviation for the both real-time PCR systems. Loligo for the Cephalopoda system and Heli for the Helicidae and Buccinidae systems. undet. = undetermined, n.m. = not measured, NTC = non template control, (1/3) = one out of three replicates.

2.2 DNA Extraction

DNA extractions were performed either according to the standard cetyltrimethylammonium bromide (CTAB) protocol DIN EN ISO 21571:2013-08, 2013,33 or with the Nucleospin Food Kit (Macherey-Nagel, Düren, Germany). Instructions of the manufacturer were followed for the Nucleospin Food Kit. Extraction controls were run with all DNA extractions. DNA concentrations were measured using the Qubit v4 Fluorometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Nucleic acid purity was estimated by measuring the 260/280 nm absorbance ratio with a NanoPhotometer (NanoPhotometer N60, Implen GmbH, München, Germany). The lack of inhibition was tested for all DNA samples using a universal 18S rRNA PCR system.34

2.3 Primer, Probes, and Real-Time PCR

DNA gene sequences (16S rRNA and 12S rRNA genes) were derived from the National Center for Biotechnology Information (NCBI) genome database (https://www.ncbi.nlm.nih.gov/). The Benchling Alignment tool (www.benchling.com) and the Oligo Calc: Oligonucleotide Properties Calculator (http://biotools.nubic.northwestern.edu/OligoCalc.html) were used to design all primers and TaqMan probes. Here, Loligo reynaudii (Accession No.: KF854035.1) served as the lead species for the singleplex cephalopod system (Loligo). For the duplex gastropod system (Heli/Bucci) Helix lucorum (Accession No.: MG709108.1) and Buccinum undatum (Accession No.: FN677400.1) were used as lead species. For the Heli/Bucci system targeting land and sea snails, 15 sequences were aligned for the mitochondrial 16S rRNA gene (Heli) and 20 for the mitochondrial 12S rRNA gene (Bucci) (Figure 1a,b). For the Loligo system targeting cephalopods, 18 sequences of the mitochondrial 16S rRNA gene were aligned (Figure 1c). Primers and probes were designed based on these alignments. Eurofins Genomics (Ebersberg, Germany) synthesized all primers and probes (Table 2). Prior to validation, the specificity of all oligonucleotides and amplicons was tested in silico using the NCBI Blast tool (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Real-time PCR was performed in a total volume of 25 μL, consisting of 1 μL template DNA (10 ng/μL) and either the Takyon Low ROX Probe MasterMix UNG (Kaneka Eurogentec S.A., Seraing, Belgium) or the primaQUANT PROBE blue qPCR Master Mix (Steinbrenner Laborsysteme GmbH, Wiesenbach, Germany). Primers were concentrated at 0.3 or 0.6 μM depending on the PCR system, while the probe concentration was 0.2 μM for both systems (Table 2). The final volume of the reaction was adjusted using molecular biology grade water (Sigma-Aldrich, Steinfurt, Germany). All samples were analyzed in triplicates in one PCR run and measured on a QuantStudio 6 Flex system, except for the robustness test where two runs were performed for each PCR parameter change. Here, we additionally used the AriaDx qPCR (Agilent Technologies, Santa Clara, CA, USA) or the Stratagene Mx3005P device (Agilent Technologies). All experiments were performed according to the following standard program: Initial step of UNG (Uracil-DNA glycosylase) activation for 2 min at 50 °C and 10 min at 95 °C, followed by 15 s at 95 °C and 1 min at 60 °C for 40 cycles. In all runs, non-template controls (NTC’s) were included with molecular biology grade water instead of DNA. When results of different real-time PCR run had to be compared (e.g., specificity and robustness test), the threshold was manually set to the same value of 0.04 ΔRN, while the baseline was always set automatically.

Figure 1 Alignment of the primer and probe binding sites for the (a) Helicidae (Heli 16S rRNA gene), (b) Buccinidae (Bucci 12S rRNA gene) and (c) Cephalopoda (Loligo 16S rRNA gene) systems with different species belonging to their respective families and classes. Similarities were highlighted with dots. Letters show mismatches in the primer and probe binding sites. Sequences of the alignment were derived from the NCBI genome database.

Table 2 Primers of the Developed Real-Time PCR System with Their According Probes

PCR system–lead species	abbreviation	sequence 5′-3′	conc. [μM]	amplicon length	
Loligo-Loligo reynaudii	Loligo 16S fwd. 1	CAGCTGCGGTATTTTAACTGTAC	0.3	170	
Loligo 16S rev. 1	CTCAATAGGGTCTTCTCGTCC	0.6	
Loligo 16S probe	FAM-AGGCTAGAATGAATGGTTTGACGAAGGT-BHQ1	0.2	
Heli-Helix lucorum	Heli 16S fwd. 1	TAAATAGCCGCAGTACTCTGAC	0.3	87	
Heli 16S rev. 1.2	AGTTTCTAGGGTCTTCTCGTC	0.3	
Heli 16S probe	FAM-TGTGCAAAGGTAGCATAATCAGTTGGCT-BHQ1	0.2	
Bucci-Buccinum undatum	Bucci 12S fwd.1	GAAACTGGGATTAGATACCCC	0.3	152	
Bucci 12S rev.1	CGTGGATTGTCGATTARGAGAC	0.3	
Bucci 12S probe	CY5-AAGGGCTTGGCGGTGTTTTAGACC-BHQ2	0.2	

2.4 Method Validation

Specificity of the target sequences and the corresponding amplicons was examined in silico using the NCBI Blast tool. In addition, the specificity was tested with DNA extracts of around 100 different animals and plants (Table 1) via real-time PCR. In all specificity test runs, 10 ng of pure, native DNA was used. For the method evaluation, the concentration of DNA extracted from the organisms was always expressed in ng/μL. Due to their unknown number of gene copies of 12S rRNA and 16S rRNA genes in their respective genomes, genome equivalents (cp-values) were not used in this study.

LOD12 and efficiency were determined for the singleplex cephalopod system using eight serial tenfold dilutions, starting at 10 to 10–6 ng of native Loligo reynaudii DNA and for the duplex gastropod system with Helix lucorum and Buccinum undatum DNA, respectively. Both tests were conducted with a background of 100 ng of wheat DNA in each dilution. The following equation was used to calculate the efficiency:

The slope (s) of the regression curve was obtained by plotting the specific CT (Cq)-values versus the log10 DNA concentration. The LOD12 was considered valid if all twelve replicates had a CT-value <37. This threshold was empirically set based on the negligible background signals obtained with pure native DNA from the target species for both systems (Table 1). For the duplex system, the detected LOD12 values were additionally validated by testing the asymmetric LOD (LODasym). To determine the LODasym, the previously detected LOD12 for each species was tested in ten replicates in excess of the other target DNA with a factor of  250, following the guidelines published by the European commission, 2021.35 Material of the lead species was cooked in a thermomixer (95 °C for 15 min) and autoclaved (121 °C for 20 min) (Systec 2540 EL, Systec GmbH, Linden, Germany) to determine the system sensitivity for processed foods.

For spiking experiments, a vegetarian dish (Vegetable pan Style Asia Curry, FRoSTA AG, Hamburg, Germany) was used to test the performance of the systems in food matrices. For this purpose, 1 g of raw material of each lead species was mixed with 9 g of the uncooked vegetarian dish to obtain 10% (w/w) mixtures. After homogenization, 1 g of these mixtures was taken and mixed with 9 g of the uncooked vegetarian dish. This was continued until a mixture of 0.0001% was obtained, resulting in a tenfold serial dilution from the highest concentration of 100,000 ppm to the lowest of 1 ppm. Aliquots were taken from all mixtures and cooked at 95 °C for 15 min, and DNA was extracted from all uncooked and cooked mixtures and used in the performance tests. Finally, to evaluate the robustness of the system, we performed a 4-factor test (mastermix/real-time PCR device/annealing temperature/day). Each DNA sample was validated using two different mastermixes and two additional annealing temperatures (58 and 62 °C) on two thermal cyclers (QuantStudio 6 Flex System and AriaDx/Stratagene Mx3005P) on two different days. Validation of the systems was based on the “Guidelines for the single-laboratory validation of qualitative real-time PCR methods” by the BVL, 201636 (Federal Office of Consumer Protection and Food Safety, Germany).

2.5 Statistical Analysis

All statistical analyses were performed using R,37 specifically the rstatix package (version 0.7.2).38 To assess the potential effect of the four factors in the robustness test, pairwise t-tests were performed. p-values were adjusted by the Holm method and considered significant at padj < 0.05. The condition “Quantstudio 6, Takyon mastermix, 60°C, day 1” served as a reference group for pairwise tests. Similarly, potential matrix effects were evaluated by t-tests regarding the processing factor and p-value adjustment by the Holm method.

3 Results

3.1 Primer Design and Species Specificity

For the Heli/Bucci system targeting land and sea snails, 15 sequences were aligned for the mitochondrial 16S rRNA gene (Heli) and 20 for the mitochondrial 12S rRNA gene (Bucci) (Figure 1a,b). For the Loligo system targeting cephalopods, 18 sequences of the mitochondrial 16S rRNA gene were aligned (Figure 1c). Primers and probes were designed based on these alignments.

The specificity of both systems was confirmed by testing DNA samples from different plant and animal phyla by real-time PCR (Table 1). All cephalopod tested species were detected with CT-values ranging from 16 to 23 for individual species. For members of the gastropod families Helicidae, Buccinidae, and Muricidae, we obtained CT-values ranging from 17 to 24. The lead species used for the cephalopod system (Loligo reynaudii) had a CT-value of 22.51 ± 0.09 (Loligo reynaudii), and the lead species used for the gastropod system had CT-values of 18.69 ± 0.06 (Helix lucorum) and 17.53 ± 0.09 (Buccinum undatum). The Loligo system showed a detection at a CT-value of 36 for Mizuhopecten yessoensis, a marine bivalve in the group of scallops. The Heli/Bucci system detected the gastropod Lissachatina fulica, a member of the family Achatinidae (Giant African Snails) at a CT-value of about 36 (Table 1). In foods containing cephalopod material, we detected traces of DNA with CT-values in the range of 18–32 (Table 1 (seafood)). Here, slightly higher CT-values were obtained due to the lower amount of cephalopod DNA in the DNA mixtures extracted from the food matrix samples. The only gastropod-containing food (rice noodles with snail extract) was tested negative (Table 1 (seafood)). To minimize false-positive results, a cut-off at a CT-value of 37 was applied, since we observed single amplifications in both systems above this threshold with the species Placopecten magellanicus, Sepia officinalis, and Octopus maya (Table 1).

3.2 Efficiency and Sensitivity

The efficiency of each system was evaluated by a serial dilution of the corresponding lead species in a background of wheat DNA. By using Loligo reynaudii as the lead species for the Loligo system, we were able to amplify DNA traces down to 1 pg (Figure 2a). An efficiency of 92.96% was calculated from the regression curve with a slope of −3.5031 (Table 3). For the Heli/Bucci system, two separate LOD12 experiments were performed for each single PCR to determine the performance differences in terms of sensitivity and efficiency of both systems and to identify the less sensitive and efficient system which was taken as the overarching efficiency of the duplex system (Figure 2b,c). For the Heli/Bucci system (lead species: Helix lucorum and Buccinum undatum), an efficiency of 95.23% was calculated (slope: −3.4443) with a detection limit down of 1 pg. R2-values of both systems were 0.999 (Table 3). To demonstrate that each of the LOD12 of our duplex system is not affected by the excess of DNA of the other target species, an asymmetric LOD was determined at the level of 1 pg. No changes in the CT-values were observed (data not shown).

Figure 2 (A) LOD12 measurement of the Loligo singleplex real-time PCR system with Loligo reynaudii DNA. (B) LOD12 measurement of the Heli/Bucci duplex real-time PCR system with Helix lucorum DNA. (C) LOD12 measurement of the Heli/Bucci duplex real-time PCR system with Buccinum undatum DNA.

Table 3 Sensitivity (LOD12/DNA [pg and mg/kg]), Slope, R2-Value and the Efficiency of the PCR Systems with Their Respective Lead Speciesa

system	lead species	DNA [pg]	DNA [mg/kg]	LOD [ppm]	linear range [pg]	linear range (food matrix)	slope [s]b	R2-valueb	efficiencyb	
Loligo	Loligo reynaudii	1	10	10	1–104	10–106 ppm	–3.5031	0.999	92.96%	
Loligo (cooked)	Loligo reynaudii	1	10	10	1–104	10–106 ppm	–3.1296	0.988	108.70%	
Loligo (autoclaved)	Loligo reynaudii	10	100	—	10–104	—	–3.074	0.997	111.51%	
Heli/Bucci	Helix lucorum/Buccinum undatum	1	10	10	1–104	10–106 ppm	–3.4443	0.999	95.13%	
Heli/Bucci (cooked)	Helix lucorum/Buccinum undatum	1	10	10	1–104	10–106 ppm	–3.4625	0.999	94.45%	
Heli/Bucci (autoclaved)	Helix lucorum/Buccinum undatum	1	10	—	1–104	—	–3.4246	0.999	95.89%	
a — = undetermined.

b Values correspond to the linear range for the non-food matrix experiments.

All systems met the criteria (efficiency: 90–110%, R2-values: above 0.98, slope: between −3.1 and −3.6) for single-copy GMO (genetically modified organism) detection methods,39 that we applied to our multi-copy systems. These values are also recommended for the validation of qualitative real-time PCR methods.40 To confirm that our system can detect the target species in processed foods, cooked and autoclaved species samples were tested (Table 3). The results showed that cooking and autoclaving the samples had no substantial effect on the sensitivity and efficiency of the Heli/Bucci system. For the Loligo system, cooking led to a minor increase in the efficiency with no effect on the sensitivity of the system. For the autoclaved Loligo reynaudii sample, a decrease in the slope was observed, resulting in a debasement of the efficiency. The sensitivity for the autoclaved Loligo sample was lowered from 1 to 10 pg.

3.3 Performance in Food Matrix

Performance in food matrix was determined for the cephalopod and gastropod system in order to investigate the sensitivity of the systems in a food matrix. Therefore, a vegetarian dish was spiked with 10% down to 0.0001% of the raw lead species sample of each system. The matrices were tested in the unprocessed and cooked state. For both systems, we were still able to detect traces of DNA in unprocessed and cooked food matrices spiked with either 0.001% of our lead species Loligo reynaudii for the Loligo system or Helix lucorum and Buccinum undatum for the Heli/Bucci system, which corresponds to a sensitivity of 10 ppm (Table 4). Results of the statistical analysis can be found in Figure 3, as well as in the Supplementary Tables S1 and S2. For the cephalopod systems, slight differences were observed between the cooked and uncooked matrices. Notably, differences in CT-values were observed between the 10 and 1% mixtures, with the 1% cooked mixture exhibiting a lower CT-value compared to the uncooked mixture. However, no differences were observed for other dilutions (Figure 3 and Table S1). For the gastropod mixtures, distinct differences in the Ct-values were observed for all samples (Figure 3 and Table S2). However, the data suggest that cooking lowered the CT-values and therefore gastropod species were detected easier in food. Only the uncooked 10% mixture of Helix lucorum showed a higher CT-value compared to the cooked matrix.

Figure 3 Results of the statistical analysis (t-test) of the matrix tests using the rstatix package (version 0.7.2)38 in R.37 Figure shows the boxplots for the performance of the different real-time PCR system with their respective lead species (A) Loligo reynaudii, (B) Helix lucorum, and (C) Buccinum undatum in uncooked (blue) and cooked (red) food matrices.

Table 4 Determination of Performance in Food Matrixa

system	food matrix dilutions	
 	10%	1%	0.1%	0.01%	0.001%	0.0001%	
Loligo	24.98 ± 0.09	26.49 ± 0.08	28.50 ± 0.17	31.54 ± 0.13	35.15 ± 0.40	undet.	
Loligo (cooked)	25.41 ± 0.05	25.28 ± 0.11	29.54 ± 0.99	31.15 ± 0.27	36.06 ± 0.94	undet.	
Heli	19.23 ± 0.11	23.38 ± 0.16	26.49 ± 0.18	30.22 ± 0.19	34.30 ± 0.89	undet.	
Heli (cooked)	20.62 ± 0.07	22.81 ± 0.06	27.59 ± .0.50	29.82 ± 0.06	34.65 ± 0.44	undet.	
Bucci	22.01 ± 0.13	24.88 ± 0.10	27.74 ± 0.24	31.13 ± 0.17	34.58 ± 0.32	undet.	
Bucci (cooked)	21.30 ± 0.05	23.15 ± 0.03	26.95 ± 0.12	28.95 ± 0.08	32.62 ± 0.31	undet.	
a DNA was extracted from food mixtures spiked with 10 to 0.0001% of Loligo reynaudii in an instant meal, as well as 10 to 0.0001% of Helix lucorum and Buccinum undatum together in an instant meal. CT-values for Heli and Bucci are deduced from the duplex PCR reaction performed on the Heli/Bucci system. Mean CT-values from triplicates with 10 ng DNA input in PCR.

3.4 Robustness

Robustness tests were performed to demonstrate that the developed real-time PCR systems are robust and can be performed without any limitations in other laboratories with alternative mastermixes and real-time PCR instruments. The results of the robustness tests for the Loligo and Heli/Bucci systems are presented in Tables 5 and 6. The statistical analysis results can be found in Figure 4, as well as in the Supplementary Materials (Figures S2, S3 and Tables S3, S4). Only slight differences between the two mastermixes (Takyon and primaQUANT) were observed in the robustness test for both systems. Also, changes in the annealing temperature had no effect on the detection of 10 ng target DNA. However, the robustness test showed noticeable differences for the Loligo system when comparing the QuantStudio 6 and the AriaDx device. The detection of individual Cephalopoda species in the QuantStudio 6 Flex system is, on average, two CT lower than in the AriaDx real-time PCR device. For some species, such as Octopus aegina (Sandbird octopus) and Sepiella japonica (Japanese spineless cuttlefish), the increase or decrease of temperature from 60 °C to 58 or 62 °C, respectively, showed clear changes with both mastermixes and both devices (Table 5 and Figure 4).

Figure 4 Results of the statistical analysis (pairwise t-test) of the robustness tests using the rstatix package (version 0.7.2)38 in R.37 Figure shows the boxplots for the species (A) Loligo reynaudii, (B) Sepiella japonica of the singleplex cephalopod real-time PCR system, as well as, (C) Helix lucorum and for the duplex gastropod real-time PCR system. The reference group for the pairwise test was the condition “Quantstudio 6 (caption), Takyon mastermix, 60 °C, day 1.” Left panels of each sub-figures show the primaQUANT mastermix, while right panels show the Takyon mastermix. Upper panel of sub-figures show day 1 and lower panels show day 2 as repetition. Quanstudio6 values are shown in blue, while values for AriaDx and Stratagene MX3005P are shown in red. On the y-axis, three different temperatures (58°, 60°, and 62 °C) are depicted. Additional boxplots of the species not shown in this figure can be found in the supplementary Tables S3 and S4.

Table 5 Results of a 4-Factor Robustness Test of the Loligo System (Mastermix/Real-Time PCR Instrument/Annealing Temperature/Day) with DNA from Five Different Cephalopoda Speciesa

 	10 ng DNA, 58 °C	10 ng DNA, 60 °C	10 ng DNA, 62 °C	
mastermix	QuantStudio 6	AriaDx	QuantStudio 6	AriaDx	QuantStudio 6	AriaDx	
 	Loligo 16S	Loligo 16S	Loligo 16S	
Takyon (Loligo reynaudii)	22.39 ± 0.04	24.55 ± 0.09	22.49 ± 0.09	24.49 ± 0.07	22.42 ± 0.17	24.67 ± 0.11	
Takyon (Loligo reynaudii)b	22.50 ± 0.02	24.68 ± 0.05	22.60 ± 0.07	24.53 ± 0.08	22.74 ± 0.16	24.69 ± 0.13	
Takyon (Amphioctopus aegina)	17.52 ± 0.11	19.57 ± 0.12	18.95 ± 0.09	21.07 ± 0.23	21.79 ± 0.24	24.40 ± 0.24	
Takyon (Amphioctopus aegina)b	17.65 ± 0.11	19.75 ± 0.06	18.87 ± 0.15	21.05 ± 0.22	21.51 ± 0.11	23.95 ± 0.33	
Takyon (Eledone moschata)	17.90 ± 0.16	19.94 ± 0.08	18.10 ± 0.11	19.93 ± 0.07	18.11 ± 0.12	20.37 ± 0.09	
Takyon (Eledone moschata)b	18.15 ± 0.08	20.12 ± 0.06	18.47 ± 0.11	20.15 ± 0.10	18.47 ± 0.06	20.37 ± 0.08	
Takyon (Illlex argentinus)	20.28 ± 0.11	22.04 ± 0.29	20.53 ± 0.07	22.26 ± 0.12	20.41 ± 0.09	22.34 ± 0.04	
Takyon (Illlex argentinus)b	20.37 ± 0.06	22.45 ± 0.27	20.62 ± 0.06	22.44 ± 0.20	20.61 ± 0.05	22.38 ± 0.16	
Takyon (Sepiella japonica)	18.31 ± 0.22	20.70 ± 0.33	19.42 ± 0.16	22.19 ± 0.29	20.20 ± 0.25	24.79 ± 0.66	
Takyon (Sepiella japonica)b	18.37 ± 0.26	20.62 ± 0.12	19.55 ± 0.24	22.59 ± 0.42	20.48 ± 0.22	25.20 ± 0.43	
Takyon NTC	 	 	 	 	 	 	
Takyon NTCb	 	 	 	 	 	 	
primaQUANT (Loligo reynaudii)	22.07 ± 0.17	24.35 ± 0.08	22.25 ± 0.19	24.11 ± 0.07	22.26 ± 0.29	24.45 ± 0.12	
primaQUANT (Loligo reynaudii)b	22.07 ± 0.24	24.29 ± 0.14	22.14 ± 0.18	24.23 ± 0.12	22.48 ± 0.15	24.46 ± 0.06	
primaQUANT (Amphioctopus aegina)	17.51 ± 0.41	19.65 ± 0.09	19.01 ± 0.29	21.05 ± 0.09	21.39 ± 0.21	23.98 ± 0.42	
primaQUANT (Amphioctopus aegina)b	17.38 ± 0.41	19.64 ± 0.09	19.45 ± 0.10	21.09 ± 0.12	21.53 ± 0.11	23.55 ± 0.08	
primaQUANT (Eledone moschata)	17.86 ± 0.35	20.07 ± 0.23	18.13 ± 0.15	19.97 ± 0.89	18.30 ± 0.29	20.70 ± 0.36	
primaQUANT (Eledone moschata)b	17.72 ± 0.22	20.07 ± 0.36	18.06 ± 0.31	20.37 ± 0.09	18.49 ± 0.14	20.45 ± 0.89	
primaQUANT (Illlex argentinus)	19.69 ± 0.08	21.99 ± 0.13	20.02 ± 0.28	22.08 ± 0.25	19.86 ± 0.11	22.24 ± 0.07	
primaQUANT (Illlex argentinus)b	19.78 ± 0.07	22.15 ± 0.24	20.16 ± 0.26	22.09 ± 0.11	20.17 ± 0.07	22.19 ± 0.12	
primaQUANT (Sepiella japonica)	18.18 ± 0.21	20.36 ± 0.16	19.18 ± 0.11	21.85 ± 0.18	20.38 ± 0.15	26.04 ± 0.18	
primaQUANT (Sepiella japonica)b	18.14 ± 0.22	20.32 ± 0.06	19.51 ± 0.08	21.83 ± 0.12	20.43 ± 0.33	25.55 ± 0.40	
primaQUANT NTC	 	 	 	 	 	 	
primaQUANT NTCb	 	 	 	 	 	 	
a Mean CT-values from triplicates with 10 ng DNA input in PCR. NTC = non template control.

b Repeat of the experiment on another date.

Table 6 Results of a 4-Factor Robustness Test of the Heli/Bucci System (Mastermix/Real-Time PCR Instrument/Annealing Temperature/Day) with DNA from Three Different Gastropoda Speciesa

 	10 ng DNA, 58 °C	10 ng DNA, 60 °C	10 ng DNA, 62 °C	
mastermix	QuantStudio 6	Stratagene Mx3005P	QuantStudio 6	Stratagene Mx3005P	QuantStudio 6	Stratagene Mx3005P	
 	Heli 16S	Bucci 12S	Heli 16S	Bucci 12S	Heli 16S	Bucci 12S	Heli 16S	Bucci 12S	Heli 16S	Bucci 12S	Heli 16S	Bucci 12S	
Takyon (Helix lucorum)	18.77 ± 0.08	 	19.18 ± 0.25	 	18.67 ± 0.11	 	18.63 ± 0.16	 	18.84 ± 0.10	 	19.13 ± 0.06	 	
Takyon (Helix lucorum)b	18.76 ± 0.09	 	18.75 ± 0.15	 	18.85 ± 0.12	 	18.99 ± 0.16	 	18.80 ± 0.04	 	18.67 ± 0.28	 	
Takyon (Buccinum undatum)	31.91 ± 0.15	17.68 ± 0.23	34.08 ± 2.58	17.52 ± 0.12	38.47 ± 0.13	17.51 ± 0.08	38.38 ± 1.48	17.36 ± 0.33	 	17.61 ± 0.10	 	17.21 ± 0.24	
Takyon (Buccinum undatum)b	32.85 ± 0.37	17.59 ± 0.15	 	17.50 ± 0.25	38.61 ± 0.33	17.60 ± 0.11	39.35 ± 0.53	17.50 ± 0.24	 	17.64 ± 0.02	 	17.55 ± 0.13	
Takyon (Bolinus brandaris)	 	23.84 ± 0.08	 	23.53 ± 0.09	 	23.66 ± 0.09	 	23.37 ± 0.20	 	23.80 ± 0.10	 	23.32 ± 0.19	
Takyon (Bolinus brandaris)b	 	23.75 ± 0.08	 	23.57 ± 0.15	 	23.76 ± 0.07	 	23.19 ± 0.17	 	23.72 ± 0.10	 	23.45 ± 0.17	
Takyon NTC	 	 	 	 	 	 	 	 	 	 	 	 	
Takyon NTCb	 	 	 	 	 	 	 	 	 	 	 	 	
primaQUANT (Helix lucorum)	18.08 ± 0.26	 	18.33 ± 0.31	 	18.26 ± 0.10	 	17.83 ± 0.29	 	18.42 ± 0.23	 	18.40 ± 0.28	 	
primaQUANT (Helix lucorum)b	18.26 ± 0.20	 	17.83 ± 0.11	 	17.99 ± 0.25	 	18.37 ± 0.20	 	18.43 ± 0.14	 	18.24 ± 0.22	 	
primaQUANT (Buccinum undatum)	35.45 ± 0.24	17.28 ± 0.21	31.51 ± 2.27	17.17 ± 0.19	38.47 ± 0.13	17.51 ± 0.08	34.59 ± 1.54	17.24 ± 0.30	 	17.59 ± 0.14	 	17.39 ± 0.16	
primaQUANT (Buccinum undatum)b	34.41 ± 2.72	17.25 ± 0.29	29.97 ± 2.19	17.19 ± 0.17	39.16 ± 0.50	17.34 ± 0.24	36.07 ± 1.28	17.22 ± 0.16	 	17.50 ± 0.17	39.56 ± 0.54	17.50 ± 0.17	
primaQUANT (Bolinus brandaris)	 	23.32 ± 0.14	38.47 ± 0.56	23.27 ± 0.27	 	23.24 ± 0.15	 	22.90 ± 0.13	 	23.40 ± 0.14	 	22.93 ± 0.09	
primaQUANT (Bolinus brandaris)b	 	23.30 ± 0.33	37.28 ± 0.92	22.83 ± 0.12	 	23.36 ± 0.26	 	22.89 ± 0.16	 	23.49 ± 0.07	 	23.22 ± 0.12	
primaQUANT NTC	 	 	 	 	 	 	 	 	 	 	 	 	
primaQUANT NTCb	 	 	 	 	 	 	 	 	 	 	 	 	
a Mean CT-values from triplicates with 10 ng DNA input in PCR. NTC = non template control.

b Repeat of the experiment on another date.

Almost no differences were observed for the Heli/Bucci system regarding the real-time PCR device or the annealing temperature. Even though the statistical test revealed clear differences for different devices for the Loligo system, all guides for qualitative real-time PCRs just point out to show positive results in the robustness that was given.36,40

4 Discussion

Both systems developed in this study were tested for about 100 different species. Among them, Mizuhopecten yessoensis (Yesso or Ezo scallop, a marine bivalve of the cold coasts in Northern Japan) revealed a cross-reaction with the cephalopod-specific Loligo system of 36.48 CT. However, using the NCBI blasting tool, we could not find any significant similarities in the primer binding sites. These findings suggest that this cross-reaction might not be due to similarities in primer binding but rather represents contamination of the sample used in this experiment. The contamination of the clam with cephalopod DNA is not unlikely, as both products are often produced and packaged in the same facilities. For the Heli/Bucci system, one cross-reaction with Lissachatina fulica (terrestrial snail distributed mostly in East Africa) was detected at a CT-value of 36.74. In this case, NCBI blasting for similarities revealed several matches in the primer and probe binding site, suggesting that this cross-reaction is due to genetic similarities rather than contamination with other snails. Similarities in the sequence of the 16S rRNA gene of Lissachatina fulica (family Achatinidae) and species of the Helicidae family can be expected as both families belong to the same order (Stylommatophora). It is important to note that the species we used for specificity testing were mostly commercial products, for which a cross-contamination cannot be ruled out.

Nevertheless, both CT-values 36.48 for Mizuhopecten yessoensis and 36.74 for Lissachatina fulica are very close to the introduced cut-off at 37. Therefore, we would consider these cross-reactions negligible as specificity tests were performed with low amounts of pure native DNA and the contamination of small traces will not affect the intended detection of cephalopods and gastropods. In our experience, a cut-off value is advisable when analyzing highly abundant targets like mitochondrial 12S rRNA and 16S rRNA genes, since background signals due to the pronounced sensitivity of the methods plus possible contamination issues in market samples are quite common. It is important to note that sometimes cut-off values can limit the sensitivity of detection methods. In our case, the CT-values obtained for LOD12 and matrix tests were always below this threshold (Table 1) and therefore our introduced cut-off value of 37 does not limit the sensitivity of our systems.

The LOD12 experiments showed only minor differences between the unprocessed and the cooked species for the Loligo system, with no loss in sensitivity. Only the detection limit of the autoclaved Loligo reynaudii sample was lower at 10 pg compared to 1 pg for cooked and unprocessed samples. The efficiency of the autoclaved Loligo reynaudii sample just missed the adopted values applicable to single-copy GMO detection methods,39 but was in the range of suggested qualitative real-time PCR efficiencies.40 Nevertheless, autoclaving is a really strong and destructive treatment for food that is not often applied, but still a detection of 10 pg is a sufficient amount to detect DNA traces of allergenic species in food to ensure the safety of the consumers. No significant changes were observed for the Heli/Bucci system.

The “Guidance document on multiplex real-time PCR methods” published by the European Commission in 202135 recommends the use of an asymmetric LOD (LODasym) instead of a LOD12. The LODasym uses copy numbers to estimate the influence of excess DNA from the DNA target of the second PCR system on the sensitivity. Since the copy number of the 12S rRNA gene in members of the family Buccinidae and of the 16S rRNA gene in members of the family Helicidae is unknown, we considered that the application of the LOD12 would also be accurate to describe the sensitivity of our system. Especially, considering, that the aim of our system is to detect any kind of allergenic trace in food in order to provide safe food for all allergic people, we are convinced that our LOD12 experiments are sufficient enough to describe the sensitivity of our system.

Finally, we have to mention that the results discussed are based on the results of the LOD12 experiments performed on the lead species of our systems (Loligo system: Loligo reynaudii, Heli/Bucci system: Helix lucorum and Buccinum undatum). In the case of the Loligo system, the lead species had one of the highest CT-values and is therefore suitable to show the highest limits of the PCR-system. For the Heli/Bucci system, the lead species were in the lower range of the CT-values, therefore laboratories that have to evaluate the method should test other lead species that showed later amplification. Of course, the use of species with higher CT-values during real-time PCR compared to our lead species may lead to a different result for the LOD12 experiments. However, it should be kept in mind, that the obtained CT-values are not always due to differences in the match of the primer and probe binding sites. The sequence alignment in Figure 1 shows that, Octopus membranaceus and Amphioctopus aegina have a total of 4 mismatches in the primer and probe binding sites. The CT-values obtained in this study for these two species are approximately 4 CT lower than the CT-value obtained for Loligo reynaudii, which has only one mismatch in the reverse primer. Instead, the quality of the extracted DNA from the sample, and thus the CT-value obtained in the real time PCR, can be significantly affected by the quality of the sample. Therefore, older samples and samples that have been treated with substances, such as citric acid or other food additives and preservatives may alter the result of the real-time PCR.41,42 This is the case for the Bolinus brandaris sample. For this species, we could only obtain a commercial product that was pickled in lemon juice, salt, and onions. Therefore, a new sample without any ingredients has to be tested to finally estimate or indicate a more accurate LOD for this species.

In Europe, snails such as members of the family Helicidae and Buccinidae are often sold as whole animals. They are usually served with some sort of herb butter or in a jar of vegetable broth, rather than in paellas or various pasta dishes like other molluscan species. Therefore, it is difficult to obtain food products containing snails. Only one package of rice noodles with a snail extract of an unknown species could be purchased in an Asian supermarket. The snail extract used in that product was undetectable by our system. In this case, the extract was either heavily processed, so that the DNA was too much damaged to be detected, or the unknown species does not belong to the families that we detect with our system. For example, the Heli/Bucci system could not detect the gastropod species Littorina littorea (family: Littorinidae) and Haliotis tuberculata (family: Haliotidae). These species belong to two different orders Littorinimorpha (Littorina littorea) and Lepetellida (Haliotis tuberculata). In contrast, the families Buccinidae and Muricidae belong to the same order of Neogastropoda and the Helicidae to the order Pulmonata. As the Haliotidae (abalone) and the Littorinidae species are two important families of gastropod sold on the European market, the development of methods to detect these missing families will be of a high priority.14 The detection of the family Muricidae could be confirmed experimentally with only one species (Bolinus brandaris). Nevertheless, in silico analysis showed that the primers used in this study should detect all members of the family Muricidae (Figure S1). In the NCBI database, we found that the members of the Muricidae family only have a maximum of two mismatches in the probe and one mismatch in the reverse primer binding site. These mismatches were always located in the 5′ end of the primer. It has been shown that mismatches in the 5′ end region have less impact on the priming efficiency than in the 3′ end region43 and therefore we should be able to amplify these bivalves’ DNA using the developed real-time PCR system. No mismatches were observed in the forward primer binding site.

As described previously, four orders (Oegopsida, Sepiida, Myopsida, and Octopoda) of the class Cephalopoda are commercially relevant. In this study, we were able to detect all species tested, including at least three species from each order, giving a good overview of the detection capability of this system for the entire class. To our knowledge, this real-time PCR system is the first system published, that can detect all commercially relevant cephalopods. This system will fill the existing gaps in the detection of cephalopods in food, as previously published systems have generally focused only on the detection of single species or families.13,15,17,18

For example, Velasco et al.17 focused only on the specific authentication of the common cuttlefish (Sepia officinalis). On the other hand, Santaclara et al.44 and Espiñeira et al.18 have developed Forensically Informative Nucleotide Sequencing (FINS)-methodologies for sequencing and authentication of the aforementioned Cephalopoda orders. This technique is used to identify samples based on informative nucleotide sequences by combining DNA sequencing with phylogenetic analysis.45 However, they did not include probes into their studies and therefore these systems are not suitable for the detection of allergens in food according to the international standards for molecular food analysis.19

The length of an amplicon of a system has a significant impact on the sensitivity and efficiency of a real-time PCR system.46 The cephalopod system from Santaclara et al.44 and Espiñeira et al.18 uses a larger amplicon of 208 bp compared to the 170 bp amplicon from our study. Still, the 170 bp amplicon of our cephalopod system must be critically considered in terms of its length. In order to obtain highly sensitive systems, it is important to use amplicons with smaller lengths of ≤150 bp while sometimes even larger amplicons (198 bp) still performed well in real-time PCR assays.46 The detection of a short fragment is less affected by processing, as they are more likely to be still intact afterward and therefore can still be detected in a real-time PCR reaction. This is reflected by the decrease of sensitivity from 1 to 10 pg of the detection limit for our cephalopod system after autoclaving. In case of our study, it was not possible to develop a shorter amplicon because the differences in the genomes of the cephalopod species did not allow for a different constellation without losing some of the specificity. Even the usage of an MGB probe as used by Velasco et al.17 would have not been beneficial. Despite this amplicon length, we were able to achieve a high and sufficient sensitivity of 10 ppm in a cooked food matrix containing the lead species Loligo reynaudii.

Recently, doses that elicit a response to objective symptoms in 1% of the allergic population (ED01) have been determined for 14 species using discrete and cumulative dosing schemes.47,48 Allergic reactions have been consistently reported by allergy sufferers despite a strict avoidance diet.49−51 Due to sharing of processing equipment, the contamination is common in the complex food supply and processing chain. Therefore, the ED01 is often used in food management, as zero risk cannot always be achieved.52 Yet, there are no data sets for mollusk species (Gastropoda, Bivalvia, and Cephalopoda). Since molluscan and crustacean allergies often occur as cross-allergies,53,54 we can assume that the same values published for crustaceans can also be applied for mollusks. Here, Houben et al.55 published an amount of 26.2 mg of crustacean protein for the use of discrete and 30.8 mg for the use of cumulative dosing schemes for the ED01. According to these values, the sensitivity of our systems of 10 ppm guarantees sufficient food safety for allergic consumers. Nevertheless, in order to protect the consumers health, it is of great importance to obtain the correct eliciting doses for all mollusks. Consulting the publication of Holzhauser et al.10 and considering the shrimp as a reference value for mollusks, we can further conclude that the sensitivity of our system would still be able to detect enough DNA in a portion size of more than 1 kg.

Overall, we can conclude that the development of the Gastropoda system, which covers the families Helicidae, Buccinidae, and Muricidae, is a first step toward the detection of gastropods in food. The system continues to exhibit high specificity, sufficient efficiency, sensitivity of 10 ppm, and a LOD12 of 1 pg. Furthermore, we were still able to detect DNA traces down to 1 pg in autoclaved samples without compromising the efficiency of the real-time PCR reaction. Despite some significant changes, the robustness test confirmed that the system can be easily performed in other laboratories with alternative mastermixes and qPCR equipment. The largest deviation observed was the increase of the CT-value up to 26 for Sepiella japonica within the cephalopod detection system using the AriaDx qPCR device compared to the CT-value of 20–21 with QuantStudio 6. For the robustness test, we initially used the QuantStudio 6 device and then moved to the AriaDx.

Since the robustness was first tested on the QuantStudio 6 device, which showed no large differences, we wanted to rule out that damaged DNA was causing these results. We additionally tested a second frozen DNA aliquot and a newly extracted DNA sample, however, no changes in the CT-values were observed. Unfortunately, we were unable to obtain a new sample of the species and therefore could not test whether our sample could have caused this CT-value shift. Since this shift was only observed in one species (Sepiella japonica) and similar results were not observed in the robustness test of the gastropod system, we assume that this event occurred due to a poorer sample quality and may not occur for other cephalopod species.

With the establishment and validation of the Loligo system, a real-time PCR system for the detection of all commercially relevant cephalopods was developed using only one primer set including a probe for PCR verification. The system exhibits high specificity, sufficient LOD12 of 1 pg and a sensitivity of 10 ppm. As for the Gastropoda system, the robustness of this system could be confirmed. Therefore, both systems are ideally suited for the analysis of foods potentially contaminated with Cephalopoda or Gastropoda.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jafc.2c08966.Alignment of the primer and probe binding sites of the Buccinidae (Bucci 12S rRNA gene) systems with different species belonging to Muricidae family and results of the statistical analysis (pairwise t-test) of the robustness tests using the rstatix package (version 0.7.2) in R (PDF)

Results of the statistical analysis of the performance in food matrix using pairwise comparison by t-test with Holm-Bonferroni correction for the uncooked and cooked matrices containing 100,000 ppm down to 10 ppm of cephalopod (Loligo reynaudii) sample material and for the uncooked and cooked matrices containing 100,000 ppm down to 10 ppm of gastropod (Helix lucorum and Buccinum undatum) sample material; results of the statistical analysis of the 4-factor robustness test of the Loligo system and the Heli/Bucci system (XLXS)

Supplementary Material

jf2c08966_si_001.pdf

jf2c08966_si_002.xlsx

The project is supported by the funds of the Federal Ministry of Food and Agriculture (BMEL) based on the decision of the Parliament of the Federal Republic of Germany via the Federal Office for Agriculture and Food (BLE) under the innovation support program (Funding reference: 281A103016).

The authors declare no competing financial interest.

Acknowledgments

We gratefully thank Ralf Winter and Finja König for the performance of several RT-PCR experiments and their excellent technical assistance.
==== Refs
References

Jablonski J. E. ; Fu T. J. ; Jackson L. S. ; Gendel S. M. Determination of protein levels in soy and peanut oils by colorimetric assay and ELISA. J. AOAC Int. 2010, 93 , 213–220.20334183
Sharp M. F. ; Lopata A. L. Fish allergy: in review. Clin. Rev. Allergy Immunol. 2014, 46 , 258–271. 10.1007/s12016-013-8363-1.23440653
Tham E. H. ; Leung D. Y. M. How Different Parts of the World Provide New Insights Into Food Allergy. Allergy, Asthma Immunol. Res. 2018, 10 , 290–299. 10.4168/aair.2018.10.4.290.29949829
Matricardi P. M. ; Kleine-Tebbe J. ; Hoffmann H. J. ; Valenta R. ; Hilger C. ; Hofmaier S. ; Aalberse R. C. ; Agache I. ; Asero R. ; Ballmer-Weber B. ; Barber D. ; Beyer K. ; Biedermann T. ; Bilo M. B. ; Blank S. ; Bohle B. ; Bosshard P. P. ; Breiteneder H. ; Brough H. A. ; Caraballo L. ; Caubet J. C. ; Crameri R. ; Davies J. M. ; Douladiris N. ; Ebisawa M. ; EIgenmann P. A. ; Fernandez-Rivas M. ; Ferreira F. ; Gadermaier G. ; Glatz M. ; Hamilton R. G. ; Hawranek T. ; Hellings P. ; Hoffmann-Sommergruber K. ; Jakob T. ; Jappe U. ; Jutel M. ; Kamath S. D. ; Knol E. F. ; Korosec P. ; Kuehn A. ; Lack G. ; Lopata A. L. ; Makela M. ; Morisset M. ; Niederberger V. ; Nowak-Wegrzyn A. H. ; Papadopoulos N. G. ; Pastorello E. A. ; Pauli G. ; Platts-Mills T. ; Posa D. ; Poulsen L. K. ; Raulf M. ; Sastre J. ; Scala E. ; Schmid J. M. ; Schmid-Grendelmeier P. ; van Hage M. ; van Ree R. ; Vieths S. ; Weber R. ; Wickman M. ; Muraro A. ; Ollert M. EAACI Molecular Allergology User’s Guide. Pediatr. Allergy Immunol. 2016, 27 , 1–250. 10.1111/pai.12563.27288833
Sheth S. S. ; Waserman S. ; Kagan R. ; Alizadehfar R. ; Primeau M. N. ; Elliot S. ; St Pierre Y. ; Wickett R. ; Joseph L. ; Harada L. ; Dufresne C. ; Allen M. ; Allen M. ; Godefroy S. B. ; Clarke A. E. Role of food labels in accidental exposures in food-allergic individuals in Canada. Ann. Allergy, Asthma, Immunol. 2010, 104 , 60–65. 10.1016/j.anai.2009.11.008.20143647
Costa J. ; Mafra I. ; Carrapatoso I. ; Oliveira M. B. Almond allergens: molecular characterization, detection, and clinical relevance. J. Agric. Food Chem. 2012, 60 , 1337–1349. 10.1021/jf2044923.22260748
Fiocchi A. ; Risso D. ; DunnGalvin A. ; Gonzalez Diaz S. N. ; Monaci L. ; Fierro V. ; Ansotegui I. J. Food labeling issues for severe food allergic patients. World Allergy Organ. J. 2021, 14 , 100598 10.1016/j.waojou.2021.100598.34703523
Monaci L. ; Brohee M. ; Tregoat V. ; van Hengel A. Influence of baking time and matrix effects on the detection of milk allergens in cookie model food system by ELISA. Food Chem. 2011, 127 , 669–675. 10.1016/j.foodchem.2010.12.113.23140717
Gomaa A. ; Boye J. I. Impact of thermal processing time and cookie size on the detection of casein, egg, gluten and soy allergens in food. Food Res. Int. 2013, 52 , 483–489. 10.1016/j.foodres.2013.01.019.
Holzhauser T. ; Johnson P. ; Hindley J. P. ; O’Connor G. ; Chan C.-H. ; Costa J. ; Fæste C. K. ; Hirst B. J. ; Lambertini F. ; Miani M. ; Robert M.-C. ; Röder M. ; Ronsmans S. ; Bugyi Z. ; Tömösközi S. ; Flanagan S. D. Are current analytical methods suitable to verify VITAL® 2.0/3.0 allergen reference doses for EU allergens in foods?. Food Chem. Toxicol. 2020, 145 , 111709 10.1016/j.fct.2020.111709.32866515
Eischeid A. C. Detection of Krill in Foods Using Real-Time PCR. Food Anal. Methods 2022, 15 , 3189–3195. 10.1007/s12161-022-02357-5.
Wilwet L. ; Shakila R. J. ; Sivaraman B. ; Nayak B. B. ; Kumar H. S. ; Jaiswar A. K. ; Jeyasekaran G. Rapid detection of fraudulence in seven commercial shrimp products by species-specific PCR assays. Food Control 2021, 124 , 107871 10.1016/j.foodcont.2021.107871.
Lee Y. M. ; Lee G. Y. ; Kim H. Y. Development of a multiplex PCR assay for the simultaneous detection of big blue octopus (Octopus cyanea), giant Pacific octopus (Enteroctopus dofleini), and common octopus (Octopus vulgaris). Food Sci. Biotechnol. 2022, 31 , 497–504. 10.1007/s10068-022-01051-w.35464245
FAO; Fishery and Aquaculture Statistics. Global production by production source 1950–2020 (FishstatJ); FAO Fisheries and Aquaculture Division [online]: Rome, 2022, www.fao.org/fishery/statistics/software/fishstatj/en (accessed December 12, 2022).
Chapela M. J. ; Sotelo C. G. ; Perez-Martin R. I. Molecular identification of cephalopod species by FINS and PCR-RFLP of a cytochrome b gene fragment. Eur. Food Res. Technol. 2003, 217 , 524–529. 10.1007/s00217-003-0788-y.
Suh S. M. ; Kim M. J. ; Kim H. I. ; Kim H. J. ; Kim H. Y. A multiplex PCR assay combined with capillary electrophoresis for the simultaneous detection of tropomyosin allergens from oyster, mussel, abalone, and clam mollusk species. Food Chem. 2020, 317 , 126451 10.1016/j.foodchem.2020.126451.32109655
Velasco A. ; Ramilo-Fernandez G. ; Sotelo C. G. A Real-Time PCR Method for the Authentication of Common Cuttlefish (Sepia officinalis) in Food Products. Foods 2020, 9 , 286 10.3390/foods9030286.32143375
Espiñeira M. ; Vieites J. M. ; Santaclara F. J. Species authentication of octopus, cuttlefish, bobtail and bottle squids (families Octopodidae, Sepiidae and Sepiolidae) by FINS methodology in seafoods. Food Chem. 2010, 121 , 527–532. 10.1016/j.foodchem.2009.12.042.
International Organization for Standardization; Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection of food-borne pathogens—general requirements and definitions. International Standard ISO 22174:2005; ISO: Geneva, Switzerland, 2005
Borgo R. ; SoutyGrosset C. ; Bouchon D. ; Gomot L. PCR-RFLP analysis of mitochondrial DNA for identification of snail meat species. J. Food Sci. 1996, 61 , 1–4. 10.1111/j.1365-2621.1996.tb14712.x.
Hammoud C. ; Mulero S. ; Van Bocxlaer B. ; Boissier J. ; Verschuren D. ; Albrecht C. ; Huyse T. Simultaneous genotyping of snails and infecting trematode parasites using high-throughput amplicon sequencing. Mol. Ecol. Resour. 2022, 22 , 567–586. 10.1111/1755-0998.13492.34435445
Ishikawa M. ; Ishida M. ; Shimakura K. ; Nagashima Y. ; Shiomi K. Purification and IgE-binding epitopes of a major allergen in the gastropod Turbo cornutus. Biosci., Biotechnol., Biochem. 1998, 62 , 1337–1343. 10.1271/bbb.62.1337.9720216
Martins L. M. ; Peltre G. ; da Costa Faro C. J. ; Pires E. M. ; da Cruz Inacio F. F. The Helix aspersa (brown garden snail) allergen repertoire. Int. Arch. Allergy Immunol. 2005, 136 , 7–15. 10.1159/000082579.15591808
Emoto A. ; Ishizaki S. ; Shiomi K. Tropomyosins in gastropods and bivalves: Identification as major allergens and amino acid sequence features. Food Chem. 2009, 114 , 634–641. 10.1016/j.foodchem.2008.09.100.
World register of marine species (WoRMS). https://www.marinespecies.org/ (accessed August 8, 2022).
Segade P. ; García-Estévez J. ; Arias C. ; Iglesias R. Parasitic infections in mixed system-based heliciculture farms: dynamics and key epidemiological factors. Parasitology 2013, 140 , 482–497. 10.1017/S0031182012001795.23279777
Manea D. ; Ienciu A. A. ; Stef R. ; Pet I. ; Smuleac L. ; Grozea I. ; Carabet A. ; Draghici G. A. ; Nica D. V. The ″Sandwich″ System: A Potential Solution for Protecting Overwintering Cornu aspersum Snails Reared in Semi-Intensive Heliciculture Farms in Colder Climates. Animals 2021, 11 , 1420 10.3390/ani11051420.34063451
Fernandes T. J. R. ; Costa J. ; Oliveira M. B. P. P. ; Mafra I. A new real-time PCR quantitative approach for the detection of shrimp crustaceans as potential allergens. J. Food Compos. Anal. 2018, 72 , 7–14. 10.1016/j.jfca.2018.05.012.
Cottenet G. ; Blancpain C. A new real-time PCR assay to specifically detect crustaceans in vegan raw materials and vegan shrimps. Food Addit. Contam.: Part A 2022, 39 , 645–652. 10.1080/19440049.2022.2036822.
Eischeid A. C. ; Stadig S. R. ; Rallabhandi P. Comparison of real-time PCR and ELISA for the detection of crustacean shellfish allergens. Food Addit. Contam.: Part A 2021, 38 , 563–572. 10.1080/19440049.2021.1874061.
Fernandes T. J. R. ; Silva C. R. ; Costa J. ; Oliveira M. B. P. P. ; Mafra I. High resolution melting analysis of a COI mini-barcode as a new approach for Penaeidae shrimp species discrimination. Food Control 2017, 82 , 8–17. 10.1016/j.foodcont.2017.06.016.
German Federal Institute for Risk Assessment Home Page. https://www.bfr.bund.de/en/development_and_validation_of_new_methods_for_the_qualitative_detection_and_quantification_of_fish__crustacea_and_molluscs_besides_of_insects_as_potential_food_allergens__aquallerg_id_-241848.html (accessed March 10, 2023).
DIN EN ISO 21571:2013-08. Foodstuffs-Methods of analysis for the detection of genetically modified organisms and derived products-Nucleic acid extraction (ISO 21571:2005 + Amd 1:2013), 2013. dx.doi.org/10.31030/1932004
Filzmaier M. Detection of animal meal in feed by real-time PCR. Diploma thesis, Institute of Analytical Chemistry and Food Chemistry, UF 809 Technical Chemistry, Biotechnology, Bio-, Food Chemistry, 2003.
Commission, E.; Centre J. R. ; Köppel R. ; Scholtens I. ; Sovová T. ; Marmin L. ; Narendja F. ; Weidner C. ; Owen H. ; Eriksson R. ; Mazzara M. ; Zdeňková K. ; Hulin J. ; Grohmann L. ; Perri E. ; Sowa S. ; Georgieva T. ; Marchesi U. ; Huber I. ; Barbante A. ; Gatto F. Guidance document on multiplex real-time PCR methods; Publications Office: 2021, 10.2760/243914.
BVL, Federal Office for Consumer Protection and Food Safety; Guidelines for the single-laboratory validation of qualitative real-time PCR methods; 2016. online publication, © BVL. March 2nd 2016 https://www.bvl.bund.de/SharedDocs/Downloads/07_Untersuchungen/Guidelines%20for%20the%20single%20laboratory.html (Accessed December 15, 2022).
R Core Team; R: A language and environment for statistical computing, R Foundation for Statistical Computing: Vienna, Austria, 2020.
Kassambara A. rstatix: Pipe-Friendly Framework for Basic Statistical Tests, 0.7.2; 2023.
Hougs L. ; Gatto F. ; Goerlich O. ; Grohmann L. ; Lieske K. ; Mazzara M. ; Narendja F. ; Ovesna J. ; Papazova N. ; Scholtens I. M. J. ; Zel J. Verification of analytical methods for GMO testing when implementing interlaboratory validated methods. Version 2; European Network of GMO Laboratories. 2018
Broeders S. ; Huber I. ; Grohmann L. ; Berben G. ; Taverniers I. ; Mazzara M. ; Roosens N. ; Morisset D. Guidelines for validation of qualitative real-time PCR methods. Trends Food Sci. Technol. 2014, 37 , 115–126. 10.1016/j.tifs.2014.03.008.
Pandir D. DNA damage in human germ cell exposed to the some food additives in vitro. Cytotechnology 2016, 68 , 725–733. 10.1007/s10616-014-9824-y.25501537
Abd-Elhakim Y. M. ; Anwar A. ; Hashem M. M. ; Moustafa G. G. ; Abo-El-Sooud K. Sodium Acetate, Sodium Acid Pyrophosphate, and Citric Acid Impacts on Isolated Peripheral Lymphocyte Viability, Proliferation, and DNA Damage. J. Biochem. Mol. Toxicol. 2018, 32 , e22171 10.1002/jbt.22171.30019796
Stadhouders R. ; Pas S. D. ; Anber J. ; Voermans J. ; Mes T. H. ; Schutten M. The effect of primer-template mismatches on the detection and quantification of nucleic acids using the 5′ nuclease assay. J. Mol. Diagn. 2010, 12 , 109–117. 10.2353/jmoldx.2010.090035.19948821
Santaclara F. J. ; Espineira M. ; Vieites J. M. Genetic identification of squids (families Ommastrephidae and Loliginidae) by PCR-RFLP and FINS methodologies. J. Agric. Food Chem. 2007, 55 , 9913–9920. 10.1021/jf0707177.17979229
Li M. ; Zhang K. Y. ; But P. P. ; Shaw P. C. Forensically informative nucleotide sequencing (FINS) for the authentication of Chinese medicinal materials. Chin. Med. 2011, 6 , 42 10.1186/1749-8546-6-42.22153058
Debode F. ; Marien A. ; Janssen E. ; Bragard C. ; Berben G. The influence of amplicon length on real-time PCR results. Biotechnol., Agron., Soc. Environ. 2017, 21 , 3 10.25518/1780-4507.13461.
Remington B. C. ; Westerhout J. ; Meima M. Y. ; Blom W. M. ; Kruizinga A. G. ; Wheeler M. W. ; Taylor S. L. ; Houben G. F. ; Baumert J. L. Updated population minimal eliciting dose distributions for use in risk assessment of 14 priority food allergens. Food Chem. Toxicol. 2020, 139 , 111259 10.1016/j.fct.2020.111259.32179163
Taylor S. L. ; Baumert J. L. ; Kruizinga A. G. ; Remington B. C. ; Crevel R. W. R. ; Brooke-Taylor S. ; Allen K. J. ; Establishment of Reference Doses for residues of allergenic foods: report of the VITAL Expert Panel. Food Chem. Toxicol. 2014, 63 , 9–17. 10.1016/j.fct.2013.10.032.24184597
Gupta R. S. ; Warren C. M. ; Smith B. M. ; Blumenstock J. A. ; Jiang J. ; Davis M. M. ; Nadeau K. C. The Public Health Impact of Parent-Reported Childhood Food Allergies in the United States. Pediatrics 2018, 142 , e20181235 10.1542/peds.2018-1235.30455345
Michelsen-Huisman A. D. ; van Os-Medendorp H. ; Blom W. M. ; Versluis A. ; Castenmiller J. J. M. ; Noteborn H. ; Kruizinga A. G. ; Houben G. F. ; Knulst A. C. Accidental allergic reactions in food allergy: Causes related to products and patient’s management. Allergy 2018, 73 , 2377–2381. 10.1111/all.13560.30004595
Zurzolo G. A. ; Allen K. J. ; Peters R. L. ; Dharmage S. C. ; Tang M. L. K. ; Said M. ; Field M. J. ; de Courten M. ; Mathai M. L. ; Campbell D. E. Self-reported anaphylaxis to packaged foods in Australia. J. Allergy Clin. Immunol.: Pract. 2019, 7 , 687–689. 10.1016/j.jaip.2018.09.006.30240886
Turner P. J. ; Jerschow E. ; Umasunthar T. ; Lin R. ; Campbell D. E. ; Boyle R. J. Fatal Anaphylaxis: Mortality Rate and Risk Factors. J. Allergy Clin. Immunol.: Pract. 2017, 5 , 1169–1178. 10.1016/j.jaip.2017.06.031.28888247
Zhang Y. ; Matsuo H. ; Morita E. Cross-reactivity among shrimp, crab and scallops in a patient with a seafood allergy. J. Dermatol. 2006, 33 , 174–177. 10.1111/j.1346-8138.2006.00040.x.16620221
Lopata A. L. ; Kleine-Tebbe J. ; Kamath S. D. Allergens and molecular diagnostics of shellfish allergy. Allergo J. Int. 2016, 25 , 210–218. 10.1007/s40629-016-0124-2.28239537
Houben G. F. ; Baumert J. L. ; Blom W. M. ; Kruizinga A. G. ; Meima M. Y. ; Remington B. C. ; Wheeler M. W. ; Westerhout J. ; Taylor S. L. Full range of population Eliciting Dose values for 14 priority allergenic foods and recommendations for use in risk characterization. Food Chem. Toxicol. 2020, 146 , 111831 10.1016/j.fct.2020.111831.33166672
