A Stable Isotope Approach to Assessing Water Loss in Fruits and

Jan 31, 2015 - (4) What is the potential of the methodology for the practical assessment of freshness of fruits and vegetables? .... CA) equipped with...
0 downloads 7 Views 1MB Size
Article pubs.acs.org/JAFC

A Stable Isotope Approach to Assessing Water Loss in Fruits and Vegetables during Storage Markus Greule,*,†,‡ Andreas Rossmann,§ Hanns-Ludwig Schmidt,§ Armin Mosandl,‡ and Frank Keppler†,‡ †

Institute of Earth Sciences, Ruprecht Karls University Heidelberg, Im Neuenheimer Feld 234-236, D-69120 Heidelberg, Germany Max Planck Institute for Chemistry, Hahn-Meitner-Weg 1, D-55128 Mainz, Germany § Isolab GmbH, Laboratory for stable isotope analysis, Woelkestrasse 9/I, D-85301 Schweitenkirchen, Germany ‡

ABSTRACT: Plant tissue water is the source of oxygen and hydrogen in organic biomatter. Recently, we demonstrated that the stable hydrogen isotope value (δ2H) of plant methoxyl groups is a very reliable and easily available archive for the δ2H value of this tissue water. Here we show in a model experiment that the δ2H values of methoxyl groups remain unchanged after water loss during storage of fruits and vegetables under controlled conditions, while δ2H and δ18O values of tissue water increase. This enhancement is plant-dependent, and the correlation differs from the meteoric water line. The δ18O value is better correlated to the weight decrease of the samples. Therefore, we postulate that the δ2H value of methoxyl groups and the δ18O value of tissue water are suitable parameters for checking postharvest alterations of tissue water, either addition or loss. KEYWORDS: hydrogen stable isotopes, oxygen stable isotopes, methoxyl groups, δ2H values, cell water, post-harvest alterations



terpenoids is restricted to the investigation of fossil matter.8 On the other hand, methyl groups from ether and ester functionalities (which we both define as methoxyl groups within this paper) of lignin and pectin occur in most fossil and recent plant biomass. Because of the nonpolar nature of the C− H bond, hydrogen atoms of methoxyl groups, once they are established, do not exchange with those of water during ongoing metabolic and diagenetic reactions. The stable hydrogen isotope value of methoxyl groups (δ2HOCH3) reflects the isotopic signature of source water.9−11 Recently, a simple and rapid method for its determination, which can be directly applied to native samples, has been developed and validated.12,13 For example, it has been shown by this procedure that methoxyl groups from plant matter are generally depleted of 2H relative to their source water but that their isotope fractionation (ε) relative to the meteoric water in question is dependent on the plant species. Nevertheless, a method based on the determination of the δ2H value of the methoxyl groups of a sample should be a reliable tool for the reconstruction of the isotopic characteristics of the (plant) tissue water during the biosynthesis of the organic matter and hence provide an intrinsic standard for the identification of secondary changes of it like illegal addition of tap water to juices or wine or its diminution by evaporation, e.g., in context with aging or drying of the tissue in question. To establish a reliable procedure for the control of water loss from plant tissues by evaporation in the course of aging on the basis of the δ2H value of methoxyl groups as an intrinsic standard and to check its potential and suitability for practical

INTRODUCTION The primary source of plant tissue water is the local meteoric water. In leaves, it is enriched with the heavy isotopes of hydrogen and oxygen by isotope effects on evapotranspiration, and after partial equilibration with fresh xylem water (Péclet effect), it is distributed to the individual plant tissues.1 Here, it serves as the source of any bound hydrogen and most oxygen in organic matter. The process is accompanied by defined kinetic (hydrogen) and thermodynamic (oxygen) isotope effects. Therefore, the isotopic characteristics of organically bound hydrogen and oxygen are archives for the water present during their biosynthesis, correlated to the local meteoric water and climate conditions. Therefore, the δ18O value of tree cellulose is a generally used biomarker for the reconstruction of recent and prehistoric climate conditions.2−4 However, quite recently, it has been recognized that the generally used correlation between the cellulose δ18O value and the local precipitate and climate is not universal but also influenced by the temperature dependence of the equilibrium constant.5 A disadvantage of the method is also that the isolation of cellulose is quite laborious and timeconsuming and that not all plants and plant tissues, especially those from annual plants, contain this biomarker. Another biomarker for the original isotope characteristics of tissue water is the δ18O value of soluble sugars. By fermentation, it is transferred to ethanol, which serves as an intrinsic standard for the proof of illegal addition of water to wine or fruit juices.6,7 However, the results can be influenced by the conditions of the fermentation, and again, the procedure requires a large amount of effort and work. Also, the determination of the δ2H value can be used for the reconstruction of the hydrogen isotope characteristics of the water present during biosynthesis. However, so far, the isotope analysis of the biomarkers n-alkyl lipids, sterols, hopanoids, and © XXXX American Chemical Society

Received: October 30, 2014 Revised: January 30, 2015 Accepted: January 31, 2015

A

DOI: 10.1021/jf505192p J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Table 1. Collection Sites of Potatoes and Carrots and Related Isotopic Compositions of Methoxyl Groups, Meteoric Water, and Isotope Fractionation site/code

geographical location

Pamplona, Spain/C4 Narbonne, France/C6 Avignon, France/C8 Mainz, Germany/C9 Dettelbach, Germany/C11 Hamburg, Germany/C12 Tomaszów Lubelski, Poland/C13 Umeå, Sweden/C14 Mainz, Germany/C16 Umeå, Sweden/C17 Belfast, Northern Ireland/C18

La Linea de la Concepciòn, Spain/P1 Murcia, Spain/P2 Valencia, Spain/P3 Pamplona, Spain/P4 Narbonne, France/P6 Villeneuve, France/P7 Noirmoutier, France/P8 Sicily, Italy/P10 Mainz, Germany/P11 Albertshofen, Germany/P12 Gleschendorf, Germany/P13 Tomaszów Lubelski, Poland/P14 Umeå, Sweden/P15

altitude (m)

Carrots 42°49′N, 01°38′W 427 43°11′N, 03°00′E 13 43°57′N, 04°49′E 23 49°58′N, 08°21′E 90 49°47′N, 10°10′E 193 53°30′N, 10°09′E 5 50°27′N, 23°25′E 273 63°49′N, 20°15′E 9 49°58′N, 08°21′E 90 63°49′N, 20°15′E 9 54°35′N, 05°56′W 22 mean isotope fractionation ε Potatoes 36°09′N, 05°20′W 9 37°58′N, 01°07′W 49 39°29′N, 00°20′W 5 42°49′N, 01°38′W 427 42°41′N, 02°53′E 37 43°53′N, 05°51′E 480 46°59′N, 02°15′W 3 37°00′N, 14°00′E nn 49°57′N, 08°22′E 85 49°46′N, 10°09′E 193 54°01′N, 10°39′E 23 50°27′N, 23°25′E 273 63°49′N, 20°15′E 9 mean isotope fractionation ε

δ2HOCH3 (‰)

δ2Htissue water (‰)

εm/wb

−213.0 −208.2 −213.5 −217.8 −233.4 −254.0 −241.1 −262.1 −203.7 −233.5 −214.2

± ± ± ± ± ± ± ± ± ± ±

0.9 0.6 0.5 2.1 0.6 0.6 1.4 1.0 1.4 2.0 2.2

−40.0 −37.0 −37.0 −54.0 −53.7 −54.4 −68.4 −75.0 −41.2 −80.2 −38.4

± ± ± ± ± ± ± ± ± ± ±

3.0a 5.0a 6.0a 3.0a 1.0 1.0 1.0 1.0 1.0 1.0 1.0

−180 −178 −183 −173 −190 −211 −185 −202 −169 −167 −183 −184 ± 13

−208.2 −210.6 −218.9 −177.0 −172.9 −192.2 −201.8 −200.0 −190.4 −202.5 −203.9 −195.6 −246.6

± ± ± ± ± ± ± ± ± ± ± ± ±

2.8 1.5 4.0 4.2 1.3 2.8 4.9 1.7 2.0 2.2 1.9 7.2 0.8

−49.1 −40.6 −54.0 −33.4 −33.0 −64.8 −52.0 −37.0 −54 −54.3 −52.3 −55.7 −92.0

± ± ± ± ± ± ± ± ± ± ± ± ±

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 3.0a 1.0 1.0 1.0 1.0

−167 −177 −174 −149 −145 −136 −158 −169 −144 −157 −160 −148 −170 −158 ± 13

Determination of δ2Htissue water was not possible. Values calculated using OIPC (cf. Hydrogen Isotope Fractionation and Correlation between Tissue Water and Plant Matter Methoxyl Groups). bIsotope fractionation was calculated according to eq 3. a

5°56′W). After being harvested, both sample sets were stored under controlled conditions at 4 °C and processed within 2 days. Three apples and carrots were selected to represent a sample (to be analyzed after different storage time periods). The samples were stored at 22 °C and a relative humidity of 55%. Water loss relative to the fresh weight was calculated from the difference between the initial weight and that after storage before squeezing. Every 4 and 7 days, samples of carrots and apples, respectively, were processed by being squeezed in a commercial juice extractor; tissue water was collected and the residual organic matter lyophilized. Both samples were stored at −20 °C until stable isotope ratio analysis was conducted. Moreover, during a sample campaign from Gibraltar, Spain, to Mainz, Germany, in May 2009 at different locations, several carrot and potato samples were collected along a south−north gradient covering a latitude range of ∼14° (36°−50°N) (see Table 1). These samples were stored at 4 °C and analyzed in the same manner described above. Isotope Ratio Analysis of Samples. Sample Preparation and Instrumentation for the 2H Analysis of Methoxyl Groups. Hydrogen isotope signatures of the methoxyl groups from fruit and vegetable matter were measured as CH3I released upon treatment of the lyophilized samples with HI by the procedure of Greule et al.12 HI (0.5 mL, 55−60%) was added to the sample (150−300 mg) in a crimp glass vial (1.5 mL). The vials were sealed with crimp caps containing PTFE-lined butyl rubber septa (thickness of 0.9 mm) and incubated for 30 min at 130 °C. After being heated, the samples were allowed to equilibrate at room temperature (22 ± 0.5 °C, air-conditioned room) for at least 30 min before an aliquot of the headspace (10−90 μL) was collected and directly injected into the analytical system. The δ2H values (see below) of CH3I released from methoxyl groups in the samples (δ2HOCH3) were measured using an HP 6890N gas chromatograph (Agilent, Santa Clara, CA) equipped with an

application, the following questions should be investigated by suitable experiments. (1) How does the δ2H value of methoxyl groups correlate with the original plant tissue water, and does the value remain unchanged and constant during the evaporative processes? (2) How do the δ2H and the δ18O values of plant tissue water behave and correlate during the evaporative processes? (3) How do the δ2H and the δ18O values of tissue water correlate to the loss of water from biomass as measured by weight loss? (4) What is the potential of the methodology for the practical assessment of freshness of fruits and vegetables? These questions will be investigated in a model experiment by measuring the δ2H values of methoxyl groups, the δ2H and the δ18O values of tissue water, and the weight loss under controlled conditions from apples, carrots, and potatoes.



EXPERIMENTAL PROCEDURES

Chemicals. Hydriodic acid (puriss. p.a., 55−60%, not stabilized) and methyl iodide (99.5%) were purchased from Sigma-Aldrich (Seelze, Germany, and Gillingham, U.K., respectively). Fruit and Vegetable Samples. Apples (Malus domestica var. Jonagold, 31 apples from a single tree) were harvested on September 29, 2009, at a local farm in Germany (geographic position, 49°57′34″N, 08°12′50″E; altitude, 190 m). Carrots (40 carrots per set) were provided from three different locations. One sample set, purchased at the local market in Mainz on August 17, 2010, had been grown nearby (geographic position, 49°58′17″N, 8°21′17″E; altitude, 86 m). The two other carrot sample sets were obtained from Umeå, Sweden (63°49′N, 20°15′E), and Belfast, Northern Ireland (54°35′N, B

DOI: 10.1021/jf505192p J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

geochemical questions, definitions for ε and Δ are as usual in this discipline.8,16 Therefore, the hydrogen isotope fractionation between plant tissue water (index w) and methoxyl groups (index m) during the synthesis of biomass, the “isotope enrichment”, is (R is here the isotope ratio)

autosampler (A200S, CTC Analytics, Zwingen, Switzerland), coupled to a DeltaPLUSXL isotope ratio mass spectrometer (ThermoQuest Finnigan, Bremen, Germany) via a thermo conversion reactor [ceramic tube (Al2O3), length of 320 mm, 0.5 mm inside diameter, reactor temperature of 1450 °C] and a GC Combustion III Interface (ThermoQuest Finnigan). The gas chromatograph (GC) was fit with a HP-5MS capillary column (Agilent) (30 m × 0.25 mm inside diameter; df = 0.5 μm), and the following conditions were employed: split injection (4:1), initial oven temperature of 30 °C for 3.8 min, ramp at 30 °C/min to 100 °C. Helium was used as the carrier gas at a constant flow of 0.6 mL/min. A tank of high-purity hydrogen (hydrogen 5.0, Linde, Höllriegelskreuth, Germany) with a known δ2H value of −295‰ (V-SMOW) was used as the working reference gas. The H3+ factor, determined daily during the measurement period, was in the range of 4.1−4.4. Samples were analyzed five times (n = 5), and the average standard deviations of the Gas chromatography-high temperature conversionisotope ratio mass spectrometry (GC-HTC-IRMS) measurements were in the range of 0.3−4.2‰. All δ2HOCH3 values were mathematically corrected by adjustment for an offset derived from the difference between the nominal and measured δ2H values of a single CH3I working standard relative to V-SMOW. The δ2H value of CH3I was calibrated against international reference substances [IA-R002 (δ2HV‑SMOW = −111.2‰), IAEA-CH-7 (δ2HV‑SMOW = −100.3‰), and NBS-22 (δ2HV‑SMOW = −118.5‰)] using an offline EA-IRMS (elemental analyzer-isotopic ratio mass spectrometer, Iso-Analytical Ltd., Sandbach, U.K.). The calibrated δ2H value versus V-SMOW for CH3I was −179.0 ± 2.9‰ (n = 15; 1σ). The standard was measured after every fifth sample injection. As this procedure represents solely a one-point calibration, it has to be pointed out that the δ2H data might be affected by an additional error (“scale compression”). Unfortunately, CH3I working standards with distinct isotopic signatures spanning the full range of measured δ2H values were not available for this study to eliminate or minimize such an error. The authors are very aware that traceability and therefore international comparability of stable isotope abundance measurements ideally require a two-scale anchor calibration with accepted isotope abundance values as recommended by the IUPAC guidelines.14,15 δ2H and δ18O Analysis of Tissue Water. Stable isotope analysis of tissue water was conducted at Isolab (Laboratorium für Stabile Isotope, isolab GmbH, Schweitenkirchen, Germany). In brief, samples of tissue water were homogenized, and the liquid was separated from solids by centrifugation. This liquid was subsequently used to measure δ2H (after distillation) and δ18O (after equilibration and dual inlet measurement) values of tissue water. A homemade equilibration unit coupled to a dual-inlet IRMS (Thermo MAT 250) was employed for oxygen isotope analysis and an equilibration unit based on platinumcatalyzed hydrogen exchange between water and hydrogen gas for hydrogen isotope analysis. The reliability of the isotopic analyses was checked by measuring in-house reference water of a known isotopic content (δ18O, −10.3‰; δ2H, −72‰) regularly in each sample batch. The reference water had been calibrated versus V-SMOW and V-SLAP international standard waters. To prevent microbial degradation of the dissolved organic compounds (CO2 formation) during oxygen isotope analysis by equilibration, sodium azide was added to the samples before they were placed into the equilibration vessels. Definition of δ Values and Calculation of Isotope Fractionation ε. Throughout this paper, the “delta” (δ) notation, the relative difference of the isotope ratio of a material to that of a standard VSMOW (Vienna Standard Mean Ocean Water), is used. Values of δ2H and δ18O relative to that for V-SMOW are defined by the following equations:

δ 2 H = (2 H/1H)sample /(2 H/1H)standard − 1

(1)

δ18O = (18O/16 O)sample /(18O/16 O)standard − 1

(2)

εm/w = R m/R w − 1 = (2 H/1H)m /(2 H/1H)w − 1 = (δ 2 H m + 1)/(δ 2 H w + 1) − 1

(3)

The authors are aware that this is an approximation and would like to refer to the studies of Coplen et al. and Schmidt et al. for further reading.14,17 Throughout the paper, we use the short form εm/w for εmethoxyl/water.



RESULTS AND DISCUSSION Hydrogen Isotope Fractionation and Correlation between Tissue Water and Plant Matter Methoxyl Groups. In our initial experiments, we determined the hydrogen isotope fractionation between tissue water and methoxyl groups of fresh potatoes and carrots, collected in different geographic locations in Europe (cf. Fruit and Vegetable Samples). For some samples (carrot samples C4−C9), it was not possible to analyze the δ2H values of tissue water as the amount of juice obtained was not sufficient to allow isotope analysis. For these samples, δ2H values of tissue water were calculated using the Online Isotope Precipitation Calculator [OIPC (http://www.waterisotopes.org/)], employing the IAEA database and interpolation algorithms.18−20 For this approach, we made the simplification and/or assumption that the δ2H values of the water taken up by the potatoes and carrots were similar to that of the annual precipitation at the sampling site. The mean isotope fractionations between tissue water and methoxyl groups (εm/w) were −184 ± 13‰ and −158 ± 13‰ for the carrots and potatoes, respectively (Table 1). This shows the necessity of determining the isotope fractionation for each species of fruit and vegetable separately. The slopes of the linear regressions for data of both potatoes and carrots are close to 1 with coefficients of determination (R2) close to 0.6 (Figure 1a; for carrots, R2 = 0.59, n = 11, and p < 0.006; for potatoes, R2 = 0.56, n = 13, and p < 0.003). Furthermore, the mean isotope fractionation εm/w of the potato samples is in excellent agreement with data previously been reported by Keppler et al. (εm/w = −161 ± 11‰).21 Via combination of the latter data with this data set (Table 1 and Figure 1b), R2 increases to ∼0.85 (R2 = 0.85, n = 30, and p < 0.001), highlighting the importance of larger data sets. Relationship between δ2H and δ18O in Tissue Water of Fruits and Vegetables. The δ2H and δ18O values of meteoric water show a linear correlation, well-known as the “meteoric water line” (MWL) roughly expressed by the equation δ2H = 8 × δ18O + 10.22 Because of transpiration, tissue water in plants becomes enriched in δ2H and δ18O relative to source water.23 This isotope enrichment is transferred to fruits and vegetables24 but shows, as compared to MWL, a relative larger enrichment of 18O and then of 2H, which might be due to the relative mass difference of the water isotopologues that is twice as large for oxygen [H218O (mass of 20) − H216O (mass of 18) = 2] as it is for hydrogen [2HHO (mass of 19) − H2O (mass of 18) = 1]. More probably, however, different diffusion constants of the stomata lead to the observed difference in the enrichments of 18 O and 2H. Thus, the slope of the resulting relationship

However, for practical reasons, we use δ values in ‰, and as our method has originally been developed and is still preferably applied for C

DOI: 10.1021/jf505192p J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 1. (a) Relationship between δ2H values of tissue water and organic matter methoxyl groups of potatoes (○) and carrots (■) harvested at different geographic locations. Error bars indicate (i) for methoxyl groups the standard deviation of mean values (n = 4) and (ii) for δ2H values of tissue water the confidence intervals calculated by the Online Isotope Precipitation Calculator (OIPC). (b) Relationship between δ2H values of tissue water and organic matter methoxyl groups of potatoes harvested at different geographic locations. Combination of the current data set and that from a previous study.10 Error bars indicate (i) for methoxyl groups the standard deviation of mean values (n = 4) and (ii) for δ2H values of tissue water the confidence intervals calculated by the OIPC.

Figure 2. Relationship between δ2H and δ18O values in tissue water of (a) carrots and (b) apples. In analogy to the meteoric water line, we refer to the relationship as the “fruit water line”. Error bars show the precision of sample measurement.

Relationship among δ18O and δ2H of Tissue Water, δ2H of Methoxyl Groups, and Storage Time. A main factor to be controlled during the storage of fruit and vegetable is the loss of water by evaporation. The degree of water loss depends on storage time, temperature, and humidity. In this study, the controlled conditions were 22 °C and a relative humidity of 55%. Details of sample storage and measurement of the δ18O and δ2H values of tissue water of carrots and apples and the respective δ2H values of the methoxyl groups during storage are described in the Experimental Section. Our results (Figure 3) show that with storage time and water loss both δ18Otw and δ2Htw of the residual tissue water become more positive, because of the preferred loss of the lighter isotopes during transpiration. As described in the previous section, δ18Otw correlates significantly better with water loss than δ2Htw, for both carrots (R2 = 0.90 ↔ 0.49) and apples (R2 = 0.92 ↔ 0.18). No significant change in the δ2H values of the methoxyl groups (δ2HOCH3) was observed for the same samples during storage for up to 20 and 69 days for carrots and apples, respectively (Figure 4). Thus, methoxyl groups of fruit and vegetable can be considered a plant archive that records the δ2H values of the plant tissue water before harvest. The results further prove that the hydrogen atoms remain fixed to the nonpolar C−H bond of the methoxyl groups and do not exchange with water once they are bound, and hence, their isotope value remains stable. At this point, it is important to note that this is not the case for the CO−H bond of carbohydrates. These results would indicate that most reliable

between δ2H and δ18O in tissue water that we define as the “fruit water line” is lower than the meteoric water line: in the literature, slopes of 2−4 can be found.24−27 Although the “fruit water line” must be determined individually for each of the different fruits and vegetables, once it has been established the isotopic signatures of 2H and 18O of tissue water can be deduced from one another. Nevertheless, we cannot completely rule out the possibility that storage conditions might have an influence on the relationship between δ2H and δ18O in tissue water (δ18Otw and δ2Htw) and thus on the slope of the “fruit water line”. This possibility should be further considered and investigated in the future. Here, we determined δ2H and δ18O values of the “fruit water line” of carrots and apples. The slopes of the linear regressions for data of both carrots and apples are significantly lower than the MWL {4.72 and 3.76, respectively [panels a (carrots) and b (apples) of Figure 2]} and in good agreement with literature data (Lupinus angustifolius, 3.1;26 grapes, 3.924). The carrot samples showed an R2 (R2 = 0.84, n = 24, and p < 0.001) considerably higher than that of the apple samples (R2 = 0.24, n = 27, and p < 0.01). D

DOI: 10.1021/jf505192p J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 3. Relationship between water loss (%) and stable isotope values of tissue water from carrots [(a) δ18O and (b) δ2H] and apples [(c) δ18O and (d) δ2H].

Figure 4. δ2H signatures of methoxyl groups {stable parameter [(a) carrots and (c) apples]} and δ18O signatures of tissue water (changing parameter) during storage time [(b) carrots and (d) apples]. E

DOI: 10.1021/jf505192p J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry and promising use for the reconstruction of the tissue water loss would be a combination of δ2HOCH3 and δ18Otw. In the subsequent sections, we will demonstrate in detail how this can practicably be undertaken. Application of the Method for the Determination of the Initial δ2H/δ18O Ratios of Tissue Water from the δ2H Values of Methoxyl Groups. The complete isotopic signature of tissue water of plant tissue prior to harvest can be deduced from the δ2H values of its methoxyl groups. The δ2H value can be calculated from the correlation between the isotope signatures of methoxyl groups and tissue water (Figure 1a,b), and the δ18O value from the “fruit water line” that is derived from δ2H and δ18O values of tissue water (eq 5 and Figure 2a,b). Because water loss is best reconstructed from δ2HOCH3 and δ18Otw in the first step, it is necessary to calculate the initial δ2H value of tissue water (δ2Htwi) using the respective food specific constant isotope fractionation between the δ2H values of tissue water and methoxyl groups (eq 4). δ 2 H twi = δ 2 HOCH3 − isotope fractionation εm/w

or, as

δ18Otwi <