Zearalenone Uptake and Biotransformation in Micropropagated

Jan 25, 2018 - A model was set up to elucidate the uptake, translocation, and metabolic fate of zearalenone (ZEN) in durum wheat. After treatment with...
2 downloads 10 Views 1MB Size
Article pubs.acs.org/JAFC

Cite This: J. Agric. Food Chem. 2018, 66, 1523−1532

Zearalenone Uptake and Biotransformation in Micropropagated Triticum durum Desf. Plants: A Xenobolomic Approach Enrico Rolli,† Laura Righetti,‡ Gianni Galaverna,‡ Michele Suman,§ Chiara Dall’Asta,*,‡ and Renato Bruni‡ †

Deparment of Department of Chemistry, Life Sciences and Environmental Sustainability, University of Parma, Via G.P. Usberti 11/a, Parma, Italy ‡ Department of Food and Drug, University of Parma, Viale delle Scienze 17/A, I-43124 Parma, Italy § Advanced Laboratory Research, Barilla G.R. F.lli SpA, via Mantova 166, Parma, Italy ABSTRACT: A model was set up to elucidate the uptake, translocation, and metabolic fate of zearalenone (ZEN) in durum wheat. After treatment with ZEN, roots and shoots were profiled with LC-HRMS. A comprehensive description of in planta ZEN biotransformation and a biotechnological evaluation of the model were obtained. Up to 200 μg ZEN were removed by each plantlet after 14 days. Most ZEN and its masked forms were retained in roots, while minimal amounts were detected in leaves. Sixty-two chromatographic peaks were obtained, resulting in 7 putative phase I and 18 putative phase II metabolites. ZEN16Glc and ZEN14Glc were most abundant in roots, sulfo-conjugates and zearalenol derivatives were unable to gain systemic distribution, while distinct isomers of malonyl conjugates were found in leaves and roots. This study underlines the potential ZEN occurrence in plants without an ongoing Fusarium infection. Micropropagation may represent a tool to investigate the interplay between mycotoxins and wheat. KEYWORDS: risk assessment zearalenone, biotransformation, masked mycotoxins, biofactories



INTRODUCTION Zearalenone (ZEN, Figure 1) is a lipophilic mycotoxin produced by some cosmopolitan members of Fusarium genus like F. cerealis, F. crookwellense, F. culmorum, F. equiseti, F. graminearum, and F. semitectum. These molds are nonobligate pathogens that may affect cereals or behave as spoilage agents in stored grains, but may as well thrive as saprophytes in soil, where they act as a primary inoculum for crop colonization.1,2 Cereal crops represent by large the most relevant food and feed source, as upon infection ZEN and its metabolites accumulate in the caryopsis of barley, corn, oats, rye, triticale, and wheat. Although not displaying the severe acute toxicity of other Fusarium metabolites, ZEN is considered an endocrine disruptor with estrogen-like properties and thus represents a public health concern in case of chronic exposure in mammalians, with clear relapses for risk assessment and food safety.3,4 Overall toxicity is the combined result of ZEN and its metabolites as, in both vegetal and animal life forms, this substance undergoes reductive phase I metabolism with the formation of α- and β-zearalenol (ZELs), of the saturated form zearalanone (ZAN) and its reduced metabolites α- and βzearalanol (ZALs), which may provide a higher affinity for mammalian estrogen receptors.5 When Fusarium infection occurs, these biotransformations lead to the production of socalled masked mycotoxins, that is the result of regioselective and stereospecific reactions mediated by P450 cytochromes and other enzymes, acting as the plant detoxification system against xenobiotics.6 To facilitate (or prevent) translocation, compartmentation, storage, and eventual disposal, these biochemical machineries catalyze the conjugation of malonyl, hexose or pentose moieties to organic xenobiotics carrying hydroxyl © 2018 American Chemical Society

groups. Further hydroxyls may be directly added to promote cell wall bounding or vacuolar segregation, producing a dynamic and variable portfolio of structures here defined as the “xenobolome”.7 Senso strictu, masked mycotoxins are those phase II conjugates resulting from metabolic pathways activated by the biochemical interplay between pathogenic fungi and infected plants and have recently become a prominent issue due to the increasing awareness of their toxicity.5 At least in infected cereals, ZEN undergoes phase II conjugation through glycosylation, being zearalenone-14-glucoside (ZEN14Glc) the most common masked form, while the detection of the isomer ZEN16Glc is the result of more recent investigations.8−10 Sulfation products like ZEN14Sulf have been reported in naturally infected cereals as well, but it is not yet clear to what extent these compounds should be considered as wheat or fungal metabolites, given the fact that Fusarium species are autonomously capable of their biosynthesis.11,12 Glutathione-conjugated forms of ZEN have not been reported so far, while malonyl moieties are often added in unspecified positions of the carbon resorcylic acid lactone scaffold.13 Quantifying ZEN conjugates is currently deemed critical to determine safe consumption levels. Upon ingestion, the parent form may be released in the digestive tract, thereby increasing the total exposure to the original mycotoxin and in some occasions up to 60% of food and feed matrices were found to contain some ZEN phase I or II metabolites.14,15 On the other Received: Revised: Accepted: Published: 1523

October 28, 2017 January 24, 2018 January 25, 2018 January 25, 2018 DOI: 10.1021/acs.jafc.7b04717 J. Agric. Food Chem. 2018, 66, 1523−1532

Article

Journal of Agricultural and Food Chemistry

Figure 1. Zearalenone and its most common masked forms.

propagated Triticum durum Desf. cultivars by monitoring the production of masked mycotoxins with a metabolomics approach. Goals included both a comprehensive description of phase I and II metabolites of ZEN, a first estimation of their distribution in plant organs and an evaluation of the possible biotechnological exploitation of the model.

hand, the unavailability of pure compounds is seriously affecting both the scientific and the regulatory efforts in the field. In 2014, EFSA CONTAM Panel considered appropriate to assess human exposure to masked mycotoxins noticing that “only a few validated direct methods have been reported in the literature so far, owing to the common lack of analytical standards besides DON3Glc”.16 As a consequence, masked mycotoxins are not monitored in routine food control and their actual toxicity may be uncertain.13 Moreover, some mycotoxins have been detected also in uninfected crops, likely as a consequence of the radical uptake of those with optimal lipophilicity, which can be released by molds in the soil. A growing evidence shows that aflatoxins, ochratoxins, patulin, citrinin and fumonisins may be absorbed by roots and translocated to above ground organs as noticed in rice, lettuce, sugar cane, asparagus, corn, coffee, and peanut crops not blatantly infected by fungi.17−22 Contrarily to most trichothecenes, ZEN is biosynthesized also during nonpathogenic fungal growth and may be therefore found as an environmental micropollutant, thus increasing the chance of its absorption or uptake.23 Despite its biodegradation by soil microflora, ZEN has been detected in drainage water and streams, in soil and in manure, therefore becoming a potential candidate for uptake in plants with a shallow radical apparatus like wheat.24−26 In vitro techniques represent a consolidated approach to investigate both the metabolic fate of xenobiotics in plants and the obtainment of natural compounds of difficult semisynthesis, but their application to mycotoxins is limited.27,28 The literature available, including isolated cell cultures and model plants like Arabidopsis thaliana, do not describe the fate of ZEN or of its derivatives in an agriculturally relevant crop or alternatively do not consider conjugated compounds, rarely taking full advantage of modern metabolomic tools.29−31 In this regard, the biosynthesis of masked mycotoxins in isolated wheat organs has been recently investigated by our group, highlighting a potential radical uptake and an intensive biotransformation both in roots and leaves.32 Therefore, the present research was performed to investigate radical uptake and metabolic fate of ZEN in two micro-



MATERIALS AND METHODS

Chemicals and Reagents. Analytical standards of ZEN (100 μg/ mL in acetonitrile), α-ZEL (solution in acetonitrile 10 μg/mL), and βZEL (solution in acetonitrile 10 μg/mL) were obtained from SigmaAldrich (Taufkirchen, Germany). Zearalenone-14-glucoside (ZEN14Glc), cis-zearalenone, and zearalenone-14-sulfate (ZEN14Sulf) were synthesized and purified in our laboratory. Zearalenone-16glucoside (ZEN16Glc) was kindly provided by Prof. Franz Berthiller (IFA-Tulln, University of Natural Resources and Life Science, Vienna). HPLC-grade solvents were purchased from Sigma-Aldrich (Taufkirchen, Germany); bidistilled water was obtained using a Milli-Q System (Millipore, Bedford, MA). MS-grade formic acid from Fisher Chemical (Thermo Fisher Scientific Inc., San Jose, CA) and ammonium acetate (Fluka, Chemika-Biochemika, Basil, Switzerland) were also used. Culture Medium and ZEN Solutions. Cultures were carried out as previously described by Righetti.32 Briefly, the medium added with 3% (w/v) sucrose was solidified with 0.8% (w/v) phyto agar and autoclaved at 121 °C for 20 min. All the experiments were carried out in triplicate and repeated three times. Micropropagation, Sample Preparation, and ZEN Treatment. Kofa and Svevo commercial durum wheat (Triticum durum Desf.) varieties were selected for their different Fusarium Head Blight resistance.33 Caryopses were soaked 70% (v/v) ethanol for 5 min and then rinsed three times in sterile distilled water. After rinsing, seeds were kept in the dark in distilled water for 5 h at 28 °C. Surface disinfection was performed with 2.5% (v/v) sodium hypochlorite for 25 min, followed by six washes with sterilized distilled water. The sterilized caryopses were cultured individually in glass culture tubes containing about 15 mL of 1/4 strength MS medium. Cultures were maintained in a growth chamber at 25° ± 1 °C with a 16 h photoperiod under fluorescent tubes at a light intensity of 27 μmol m−2 s−1. One week after germination, seedlings grown above 5 cm in length were selected and segments of 10 mm containing apical meristems and leafs were obtained by transverse cuts with a scalpel blade. Explants were cultured on shoot multiplication medium (SM), 1524

DOI: 10.1021/acs.jafc.7b04717 J. Agric. Food Chem. 2018, 66, 1523−1532

Article

Journal of Agricultural and Food Chemistry

Figure 2. Residual ZEN in culture medium during growth of micropropagated plantlets of T. durum cultivars Kova and Svevo. Data are given in terms of residual ZEN% in the medium (n = 4). Initial amounts were 25 μg (A) and 200 μg (B) diluted at 0.78 μM (A) and 6.3 μM (B). containing MS basal salt medium, added with 8.88 μM N6benzyladenine (BAP) and 2.2 μM 2,4-dichlorophenoxyacetic acid (2,4-D). Multiple shoot clumps arising from the shoot apexes were divided and subcultured in SM every 3 weeks. Samples were prepared and treated according to the protocol already reported elsewhere.32 Plant samples were freeze-dried for 24 h using a laboratory lyophylizator (LIO-5PDGT, 5 Pascal s.r.l., Trezzano sul Naviglio, Italy) and then milled. Homogenized plant materials were extracted with acetonitrile/water/formic acid (79:20:1, v/v) as previously reported.32 All medium samples were diluted with water/methanol (80:20, v/v) to achieve a final ratio of 1:1 (v/v), vortexed for 1 min and then subjected to LC-MS analysis. Targeted UHPLC-MS/MS Analysis of Mycotoxins. UHPLC Dionex Ultimate 3000 separation system coupled to a triple quadrupole mass spectrometer (TSQ Vantage; Thermo Fisher Scientific Inc., San Jose, CA, USA) equipped with an electrospray source (ESI) was employed. For the chromatographic separation, a reversed-phase C18 Kinetex column (Phenomenex, Torrance, CA, USA) with 2.10 × 100 mm and a particle size of 2.6 μm heated to 40 °C was used. A full description of the target LC-MS method used has been already reported by our group.32 Quantification of target analytes (ZEN, ZEN14Glc, ZEN16Glc, αand β-ZEL) involved two different calibration sets. Mycotoxins content of roots and leaves was quantified by matrix matched calibration standards of ZEN, ZEN14Glc, α and β-ZEL prepared by dissolving aliquot portion of standard solution in the blank wheat extract. From the acetonitrile standards solution (1000 ng/mL) 6 diluted matrix solutions were prepared (calibration range 1−500 ng/ mL). Whereas medium ZEN content was quantified by using an additional matrix matched calibration standards prepared by diluting blank medium with water/methanol (80:20, v/v) (calibration range 10−1000 ng/mL). A good linearity was obtained for all the considered mycotoxins (R2 > 0.99). Quantification of ZEN, ZEN14Glc, ZEN16Glc, α and β-ZEL was performed employing Thermo Xcalibur 2.2.SP1 QuanBrowser software. HPLC-HRMS Analysis. Roots and leaves extracts were subjected to HRMS in order to investigate the formation of ZEN biotransformation products, for which analytical standards were not available. Chromatographic conditions were the same used for the targeted analysis. LCHRMS full scan spectra were recorded using Q-ExactiveTM high resolution mass spectrometer (Thermo Scientific, Bremen, Germany) equipped with electrospray ionization (ESI). A full description of the operating method was reported elsewhere by our group.32

Putative Identification of ZEN Metabolites. The identification of ZEN, α-ZEL, and β-ZEL was confirmed by comparison with commercial standards. Similarly, cisZEN, ZEN14Sulf, ZEN14Glc, and ZEN16Glc were accurately identified by comparison with authentic standards, obtained by chemical or enzymatic synthesis. For other metabolites, the annotation process previously described32 involved; (i) the measured accurate mass, (ii) similarity of experimental and theoretical isotopic pattern, (iii) HRMS-MS spectra of conjugated metabolites should give rise to product ion of intact ZEN (m/z 317.1389) and ZEL (m/z 319.1550) Only for few metabolites, fragmentation spectra could not be collected, due to parent ion abundance below the threshold. In this case, a tentative annotation based on accurate mass and elemental formula was performed, as already proposed by other authors34 Statistical Analysis. All statistical analyses were performed using IBM SPSS v.23.0 (SPSS Italia, Bologna, Italy). Data were analyzed by Kruskal−Wallis test followed by Duncan posthoc test (α = 0.05).



RESULTS AND DISCUSSION The presented approach is based on the assumption that uninfected wheat plants can absorb and quantitatively metabolize ZEN, with micropropagation culture serving as a model for biotechnological and investigation purposes. Three separate steps were undertaken: an estimation of the actual absorption from culture media, a determination of ZEN and of those masked forms available as standard references and an untargeted monitoring of ZEN phase I and II plant metabolites. Separate results were collected for below ground (i.e., roots) and above ground organs (i.e., leaves and culm) in Triticum plantlets belonging to cultivars with a higher (Kofa) and lower (Svevo) capability to biotransform Fusarium mycotoxins.33 Evidence of Plant Absorption. It has been suggested that ZEN may be a good candidate for radical uptake and that T. durum cultured organs could absorb this substance from sterile growth medium.18,32 This lead to the hypothesis that wheat plants may absorb, biotransform, accumulate and eventually transport ZEN or its modified forms via xylematic tissue without being necessarily infected by Fusarium. At the same time, other investigations have in part described some of these steps in vitro for ZEN in A. thaliana, corn, and barley or for other Fusarium mycotoxins, like fumonisins in maize.9,27,29,35 1525

DOI: 10.1021/acs.jafc.7b04717 J. Agric. Food Chem. 2018, 66, 1523−1532

Article

Journal of Agricultural and Food Chemistry A screening was performed by exposing micropropagated T. durum plantlets with calibrated amounts of pure ZEN and monitoring its disappearing in a 14 days period (Figure 2). Two separate concentrations (0.78 and 6.3 μM, corresponding to a total amount of 25 and 200 μg per treated seedling, respectively) were administered after previously checking the tolerance of cultured plants. At the higher exposure level, wheat plantlets showed no evident symptoms of stunting, necrosis, wilting or chlorosis and provided growth patterns similar to those of untreated control wheat. A previous, not yet replicated report also hypothesized that ZEN may be an endogenous product of plant metabolism during vernalization, acting as a growth and flowering regulator.36 To avoid any interference on this regard, controls with untreated plants were set up and resulted negative both at the beginning and at the end of the experiment. At the tested conditions (LOD 0.35 ppb) we may exclude the presence of endogenous ZEN in wheat plantlets. The removal of ZEN from the medium by a single plantlet was quick and followed a nonlinear, quasi-logarithmic curve (Figure 2), exceeding 60% yet after 12 h and becoming almost complete at 14 days with no statistically significant difference neither between tested cultivars nor between treatments. Nonlinearity at low concentration as those used in our experiments are reputed consistent with a dual-phase sorption model for soil, which posit the coexistence of a high affinity sorbent in low amounts along with a much more abundant phase providing lower affinity, which in the case of ZEN may be respectively represented by the lipophilic and the hydrophilic components of root cells and of tegumental or parenchimatous tissues.37 The sorption trend is in agreement with similar experiments made with A. thaliana, where, even if at much higher concentrations (50 μM, for a total amount of 318 μg) and by a much larger set of plants (n = 40), an almost complete absorption was noticed after 24h.27 Similarly, the absorption of ZAN (2 μM, for a total amount of 6.3 nmol) by a single maize plantlet grown in hydroponic conditions was almost exhaustive after 22 days while three barley plantlets exhibited the capability to absorb more than 95% of ZEN at 5 mg/L (78.5 nmol) in 24 h9,30 A previous investigation provided a slower absorption in isolated roots if compared to excised leaves, likely as a consequence of the larger surface available in the latter and to the effect of lipophilic leaf cuticle.32 In plantlets, ZEN was instead absorbed almost at the same pace of excised leaves, likely as a consequence of the active hydraulic translocation promoted by a higher transpiration rate, which increase the uptake of water and solutes from culture medium. It may be useful to remember that, during in vitro growth, transpiration is regulated by the equilibrium with atmospheric moisture in a closed environment. While this may represent a key difference with actual in field performance for wheat, it also assures more reproducible results from a biotechnological perspective. Moreover, due to their more active growing nature, plantlets had on average an overall higher biomass than isolated roots previously used (0.14 mg in plantlets, 0.04 mg in isolated roots) and therefore their uptake potential was higher. It is well-known that xenobiotic substances may be modified both by intracellular plant enzymes and by enzymatic pools secreted into the soil, or may also be diffused back in the rhizosphere once biotransformed. In some cases, this led us to hypothesize the presence in plants of a sort of an excretion system for unwanted compounds, acting through one-way excretion. Therefore, growing media was carefully monitored, but regardless of exposure level and of sampling time, neither

ZEN nor any of its known modified forms were exuded or produced in the growing media at values exceeding the LOD of our method (0.35 ppb). No degradation occurred during the whole experiment in positive controls, with the unique exception being an expected 40:60 isomerization ratio from trans- to cis-ZEN.32 Previous reports have noticed for A. thaliana the presence in the growth medium of both ZEN14Sulf and α, β-ZAL in traces, while α-ZAL was actively accumulated outside of the plant roots in maize roots exposed to low ZAN quantities (2 μM) and both ZEN16Glu and ZEN14Glu were detected in barley growing media 24 h after exposure.9,30 It remains to be verified if this behavior is related to biological differences (e.g., a different detoxification pattern determined at species-specific level) or induced by a different experimental design, as for instance the 10-fold larger concentration used in the Arabidopsis experiment may activate a stronger metabolic response. Given the aseptic nature of in vitro systems, the gradual disappearing of ZEN from growth medium and the subsequent evidence of biotransformation are to be considered as the exclusive result of a plant-mediated activity. In this regard, no significant differences were noticed between Kofa and Svevo cultivars, a trend that was already noticed in cultured organs.32 This suggests the hypothesis of a limited intraspecific variability in wheat, albeit the minimal size of the sample do not allow conclusive remarks. This would also fit with the hypothesis that ZEN is not a virulence factor for Fusarium infection but rather a substance acting as a plant-growth regulator and therefore more well-tolerated in wheat.38 At the end of the experiment, wheat plantlets had an average biomass of 0.58 ± 0.17 mg/dw and were capable to absorb up to 200 μg of ZEN each during a time span shorter than the half-life of this substance in soil (11 days).39 If actually available in soil, an adult wheat plant with a developed root system could theoretically withdraw consistent amounts of ZEN as a consequence of simple uptake rather than by (or in addition to) active Fusarium infection. As shown later, however, the differential accumulation between roots and above ground organs drastically reduces the risk for wheat, but nevertheless some caution may be paid and further investigation should be performed for asymptomatic crops with edible roots or tubers, in particular if grown in fields with a previous history of Fusarium presence. Overall, the evidence collected confirms that wheat plants may quantitatively absorb up to 200 μg ZEN, which besides its toxicological implications represents a necessary starting point for the evaluation of in vitro plant cultures as potential biotechnological tools for masked mycotoxin studies. Semiquantitative Screening of ZEN Conjugates and Their Distribution in Roots and Aerial Parts. ZEN being a xenobiotic prone to phase I and II biotranformation in planta, its chemical fate and organ distribution should be carefully investigated to describe which biotransformations occur and if some form of selective accumulation is present. Plants may actively prevent the systemic distribution of foreign substances and therefore it may influence the presence of ZEN in edible parts or determine which organ may be more suitable for biotechnological purposes. Since few modified forms of ZEN were available as reference compounds (i.e., ZEN14Glc, ZEN16Glc, α- and β-ZEL), we followed a targeted-untargeted approach, producing semiquantitative data only for the available compounds and providing separately a qualitative description of the whole “xenobolome”. 1526

DOI: 10.1021/acs.jafc.7b04717 J. Agric. Food Chem. 2018, 66, 1523−1532

Article

Journal of Agricultural and Food Chemistry

Table 1. Semiquantitative Data of ZEN and Derived Plant Metabolites Detected in Planta After 14 Days of Exposure, Expressed as Percent of Total Derivatives ZEN [%] cultivar

toxin [μg]

Svevo Kofa Svevo Kofa

200 25

roots

leaves

± ± ± ±