Accepted Article
Article Type: Original Article
Comparison of different methods for the recovery of DNA from spores of mycotoxin producing moulds in spiked food samples
Running headline: DNA recovery from spores of moulds
Sabrina Grube1, 2, Jutta Schönling3 and Alexander Prange1,2,4*
1
Microbiology and Food Hygiene, Niederrhein University of Applied Sciences, D-41065
Mönchengladbach, Germany 2
Institute of Microbiology and Virology, University of Witten/Herdecke, D-58448 Witten,
Germany 3
GEN-IAL GmbH, D-53842 Troisdorf, Germany
4
Center for Advanced Microstructures and Devices (CAMD), Louisiana State University, Baton
Rouge, Louisiana 70806
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/lam.12405 This article is protected by copyright. All rights reserved.
Accepted Article
* Corresponding author:
Alexander Prange Microbiology and Food Hygiene Niederrhein University of Applied Sciences Rheydter Str. 277 D-41065 Mönchengladbach Phone: +49-2161-186-5390 FAX: +49-2161-186-5314 Email:
[email protected]
Keywords DNA extraction, moulds, spores, cell disruption, real-time PCR, food safety, food quality assurance
SIGNIFICANCE AND IMPACT OF THE STUDY The choice of “ready-to use” commercial kits and methods has been of great importance regarding the recovery of extracted DNA. However, these commercially available kits are neither effective nor time-efficient when extracting DNA from fungal spores embedded in complex food matrices. Different extraction principles were compared and their effectiveness tested using real-time PCR. The combination of different principles for extraction and concentration of DNA was found as the most efficient method (quantity and purity) to obtain DNA from moulds and their spores from food samples.
This article is protected by copyright. All rights reserved.
Accepted Article
ABSTRACT Several food samples were spiked with fungal conidia in order to test the efficiency of different cell disruption methods and DNA-extraction kits for subsequent molecular detection. For disrupting the firm cell walls of the spores, two different pretreatment methods, namely sonication and bead beating were
tested against no pretreatment. The subsequent DNA-extraction and -purification was performed using three different DNA-extraction methods, which are based on a diverse combination of extraction principles, such as precipitation, thermic-enzymatic lysis, pH-enhancement and bonding on a silica membrane. The aim of the study was, to find out which pretreatment and DNA-extraction method is suitable for the recovery of detectable amounts of fungal DNA from different food matrices.
INTRODUCTION Contamination of raw materials, processed food and food-products with moulds is a serious food safety problem for the entire food industry (e.g. Wan et al. 2006; Dijksterhuis 2007; Ray and Bhunia 2013). Firstly, moulds are ubiquitous owing to their ability to form spores (Ugalde and Corrochano 2007; Geisen 2010). They are easily dispersed by air and may occur in any environment – including farm fields, storage rooms or sanitized food production facilities (Ugalde and Corrochano 2007). Secondly, the spores may remain undetected and germinate at any stage or time in the food processing chain (Chitarra and Dijksterhuis 2007; Geisen 2010). Thirdly, several mould strains are able to form mycotoxins for example aflatoxin B1 (Aspergillus flavus) and ochratoxin A (Aspergillus spp., Penicillium
spp.). These factors have led to great economic losses in the food industry and present potential enormous health risks to the consumers.
This article is protected by copyright. All rights reserved.
Accepted Article
Therefore, a quality control step that uses a rapid method of analysis of raw materials, pre-foods and packed foods is necessary for use in the food industry food producers. Currently, traditional visual and culture methods are used as quality control devices, however they are time-consuming and request detailed knowledge of micro- and macroscopic morphology, and often the results are sometimes imprecise (Shapira et al. 1996; Geisen, 2010). Molecular detection methods present a novel approach to
detect moulds before the formation of mycelium has taken place (Chitarra and Dijksterhuis 2007) as mould spores contain deoxyribonucleic acid (DNA). The greatest challenge impeding and affecting the molecular analysis of contaminated food samples is the DNA-extraction-step. First we must consider that the ingredients of the food sample and its own DNA impede recovery of fungal DNA and disturb subsequent molecular detection (Rossen et al. 1992; Wilson 1997; Geisen 2010). Furthermore, DNA extraction from fungal mycelium and spores is hindered due to the rigid cell walls containing chitin as principal component (Pérez and Ribas, 2013). Several mechanical methods such as bead beating, liquid nitrogen or sonication have already been tested for breaking up the fungal cell wall (e.g. Cenis 1992; AlSamarrai and Schmid, 2000; Griffin et al. 2002; Fredricks et al. 2005; Saß et al. 2005; Rodriguez et al. 2012).
The recovery of mould-DNA from diverse matrices has been tested in two ways of pretreatment, namely bead beating and sonication against no pretreatment. Moreover, there are many different “ready-to-
use” DNA-extraction Kits available on the market for the recovery of DNA from a wide range of matrices and organisms (Motková and Vytřasová 2011). These are based upon diverse extraction principles, such as precipitation, bonding on silica membranes or thermal, chemical and/or enzymatic lysis (Haugland et al. 2002; Fredricks et al. 2005; Motková and Vytřasová 2011; Rodriguez et al. 2012). In this study, DNAextraction was performed with “NucleoSpin® Food” (Machery-Nagel GmbH & Co. KG., Düren, Germany),
This article is protected by copyright. All rights reserved.
Accepted Article
“First-DNA all-tissue Kit” (Genial GmbH, Troisdorf, Germany), and a modified protocol using an alkaline lysis buffer (“Mould extraction”). Sodium hydroxide (abbrev. NaOH) is able to loosen the firm structure of a cell wall and it leads to denaturation of the double stranded DNA. Restoration of the DNA is performed by Tris/HCl (Meeker et al. 2007). After NaOH treatment, a concentration and clarification step by bonding on a silica membrane was added. The mould strains used for the study were obtained by the German Collection of Microorganisms and Cell Cultures (Deutsche Sammlung für Mikroorganismen und Zellkulturen (DSMZ), Braunschweig, Germany. In order to evaluate the results, real-time-PCR was performed and the threshold cycle (Ct) values, compared to each other within one food sample. The Ct-value conveys the amount of cycles needed until the background noise is exceeded for the first time, and the DNA-amplification is exponential (Roche Diagnostics program, Vers. 3.0, 2014). Furthermore, there is a correlation between Ct-value and initial amount of DNA (Konrad and Busch 2010). The aim of the study was to find out which method leads to the highest yield of mould-DNA being extracted from the spiked food samples.
RESULTS AND DISCUSSION
The following results are interpreted by comparison of the Ct-values of each experiment respectively. The results of the different experiments (food samples and mould strains) cannot be compared with each other since the target sequences being amplified by real-time PCR are derived from repetitive rDNA gene sections of the Internal Transcribed Spacer region (ITS). These gene units occur in various amounts within the genome of the respective mould strains (Geisen 2010) leading to different Ct-values of the respective mould strains for the same spore concentration. Furthermore, the spore size and therefore the DNA-amount vary between the mould species (e.g. Samson et al. 2010). Thus, it is not possible to compare Ct values of real-time PCR reactions of different mould species.
This article is protected by copyright. All rights reserved.
Accepted Article
Two aspects are to be considered when evaluating the results: Firstly, the pretreatment method sonication (S). bead beating (GB) or no pretreatment (C) and secondly the DNA-Isolation method (FirstDNA All-tissue (AT), NucleoSpin® Food (NS) or Mould-DNA-Extraction (ME) are compared concerning the
smallest Ct-value. This value is quivalent to the highest initial amount of DNA in the sample. The average Ct-values of the respective experiments are displayed in Table 4. The deviations of the doublepreparations are between 0.2 and 1.3 cycles due to preparation and pipetting errors. In two cases the deviation is 1.8 cycles (AT-S Experiment 2 and ME-GB, Experiment 3).
Pretreatment method: As shown in Table 4, the pretreatment method leading to the earliest Ct- value in most food samples, within the respective DNA-extraction method, is the glass bead beating.
Sonication was successful for pistachios and malt grains as well. The lowest Ct-values were not observed in experiments that did not undergo a pretreatment (C, S, GB), but within one method (AT, NS, ME), vary between 0 and 8 cycles - mainly 3-4 cycles. A Ct-difference of 2 cycles is equal to an initial DNAconcentration difference of 1:4 (Mäde, 2010). In experiment 1, For ME, the deviation of Ct is 4.5 cycles, for AT 4.4 cycles. Considering the possible errors, there is still a difference of 3 cycles, which means an initial DNA-yield of about one ninth for no pretreatment compared to glass bead beating. The highest deviation in Ct for the different pretreatment methods can be observed at experiment 2 (deviation of 8.3 Ct´s), DNA-isolation method AT. However, possible preparation errors cannot be fully excluded. Sonication leaded to an almost ten times higher yield than no pretreatment in experiment 5.
DNA-isolation method: There are significant differences in Ct-value for each of the DNA-isolation methods within the respective experiments. In experiment 1, the AT-method leads to earlier Ct-values than ME of 6.4 cycles (23.02 - 30.89; considered a tolerance of 1.5 cycles as possible error), in
This article is protected by copyright. All rights reserved.
Accepted Article
experiment 2, the Ct´s for NS and ME are approximately 3.5 cycles higher than for AT and in experiment 5, the maximum deviation is 5 cycles. In experiments 3 and 4 however, the differences in Ct for the respective DNA-isolation methods are much smaller (
