Anal Bioanal Chem (2004) 379 : 495–503 DOI 10.1007/s00216-004-2592-3
O R I G I N A L PA P E R
Rodolfo G. Wuilloud · Sasi S. Kannamkumarath · Joseph A. Caruso
Speciation of nickel, copper, zinc, and manganese in different edible nuts: a comparative study of molecular size distribution by SEC–UV–ICP–MS
Received: 17 November 2003 / Revised: 3 March 2004 / Accepted: 10 March 2004 / Published online: 27 March 2004 © Springer-Verlag 2004
Abstract Molecular size distribution patterns of Cu, Mn, Ni, and Zn were determined in several nut species by sizeexclusion liquid chromatography (SEC) coupled on-line to UV and inductively coupled plasma mass spectrometry (ICP–MS) for detection. The molecular weight (MW) fractionation of the different metals was performed with a Superdex Peptide column, injecting 100 µL of the extracted solutions. The association of the elements with different MW fractions was observed with sequential detection by UV and ICP–MS. Various separation conditions were evaluated to obtain proper resolution and reproducible results with the size-exclusion column. Complete MW information of the elemental fractions in the nut samples was obtained within a retention time of 30 min. Fractionation of the above mentioned elements was done in nine different nut species commonly found in commercial markets. Variability of the fractionation patterns for two different extraction media, 0.05 mol L–1 NaOH and 0.05 mol L–1 HCl, was evaluated for every nut sample. Differences in the elemental fractionation patterns were found depending on the extraction procedure, nut species, and the type of element studied. It was also observed that the elements studied showed predominant association with high MW fractions when extracted with basic solution whereas with acidic extraction media only low MW fractions were obtained. Keywords Multielemental speciation · Nuts · Fractionation · Metalloproteins · SEC–ICP–MS
Introduction Nuts are complex plant foods that are not only rich sources of unsaturated fat, but also contain several nonfat con-
R. G. Wuilloud · S. S. Kannamkumarath · J. A. Caruso (✉) Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221-0172, USA e-mail:
[email protected]
stituents such as plant protein, fiber, micronutrients, plant sterols, and phytochemicals [1]. Among foods with favorable fatty acid profiles, nuts have received particular attention because of the epidemiological association of their frequent intake with protection from coronary heart disease [1, 2, 3, 4] Despite their high calorie content, consumption of nuts and nut products is highly popular [5]. They are used extensively by manufacturers in fruit and nut cereal combinations such as granola bars, chewy fruit bars, breakfast cereals, and desserts. Furthermore, nuts are the third most heavily used ingredient in confectionery manufacture [6]. The elemental composition of plant foods are important for varied reasons such as, nutritional value, toxicity, pollution, geographic origin of plants, etc. [7, 8, 9]. Essential trace elements are vital for various metabolic processes, and toxic elements if present in relatively high amounts, adversely affect these processes [10, 11]. On the other hand, the presence of metals in living organisms, and more specifically in plants, is of particular concern for adequate nutrition of other organisms higher in the food chain. Therefore, information about trace elemental distribution is necessary to estimate the intake of essential elements and to evaluate the potential health risks caused from exposure to toxic elements in foodstuffs [12]. Despite the significance of the presence of metals in food, the number of studies reported to determine total elemental concentrations in nuts are relatively few [7, 11, 13, 14, 15, 16, 17, 18, 19, 20]. Studies about elemental speciation in nuts have not been developed with the exception of selenium. Speciation studies of selenium in nuts were developed in our research group [21, 22]. Selenium was analyzed in several nuts to study its distribution among different molecular weight fractions using size-exclusion chromatography (SEC) coupled to inductively coupled plasma mass spectrometry (ICP–MS). The results showed that selenium was mainly associated to high molecular weight fractions, which in the particular case of nuts, likely corresponded to proteins. Because the chemical composition of nuts is complex (lipids 50–70%, proteins 10–20%, and carbohydrates
496
10–20%) trace elemental analysis is difficult and a suitable sample preparation step has to be developed prior to the analysis. Several wet-digestion procedures and detection systems were used for analysis of total elements in nuts. These experiments showed that incomplete digestion of the sample was the main problem giving poor reproducibility and accuracy in the results [23]. Moreover, care has to be taken in the sample preparation when speciation studies are to be performed in order to preserve the natural composition of the elemental species. SEC is a very useful technique to study the distribution of elements in different molecular weight fractions and gives information about the association of elements with the different compounds in the sample [24, 25]. As mentioned earlier in the case of nuts, SEC–ICP–MS has only been applied for distribution studies of Se. However, numerous other elements of nutritional and toxicological interest exist in nuts, which have not been studied. Although the application of SEC–ICP–MS coupling for metal speciation has some limitations [26], it has been accepted as a very useful hyphenated technique in speciation studies to estimate the association of elements to the compounds present in the sample. These studies are considered as the starting point for a deeper evaluation of the nature of elemental species and subsequent studies [27, 28]. Regarding the importance of the elements studied in this work, Cu is needed for the transport of iron and also involved in the synthesis of connective tissue, lipid metabolism and antioxidant protection while also Mn is a component of the antioxidant enzyme superoxide dismutase [29]. Zinc is a cofactor for many enzymes and also affects protein synthesis through gene expression while Ni is thought to play an important role in folate metabolism [29]. Due to the already known importance that these elements have for living organisms, distribution patterns of Cu, Mn, Ni, and Zn among the different molecular weight fractions corresponding to the compounds present in most consumed nuts were studied. Size-exclusion chromatography was coupled sequentially to UV and ICP–MS detection systems for the separation and on-line detection of these elemental fractions in different edible nuts. Different extraction procedures in basic and acidic conditions were evaluated and optimized. The performance of the separation was studied by modifying the different SEC variables involved in the system.
Experimental Instrumentation A high performance liquid chromatographic system (HPLC) was used. This was a Shimadzu (Shimadzu Scientific Instrument, Columbia, MD, USA) LC-6A pump with a 100 µL loop (Rheodyne 7725 injection valve, Rheodyne, Cotati, CA, USA), and an SPD6A UV–visible spectrophotometric detector coupled through an R9 232 Dionex Interface with 33 MHz PC equipped with AI-450 release 3.21 software. Elemental detection was performed using an ICP–MS Perkin– Elmer Sciex Elan 6000 (Ontario, Canada) with a Gem Tip crossflow nebulizer (Perkin–Elmer) and Ryton spray chamber (Perkin–
Table 1 ICP–MS and SEC chromatography instrumental conditions ICP–MS conditions Forward power Plasma gas flow rate Auxiliary gas flow rate Carrier gas flow rate Sampling depth Sampling and skimmer cones Dwell time Isotopes monitored SEC chromatography conditions Column Resolution range Mobile phase Flow rate Injection volume UV–visible wavelengths studied
1350 W 15.0 L min–1 0.87 L min–1 0.975 L min–1 6 mm Nickel 0.1 s per isotope 63Cu, 55Mn, 58Ni, 68Zn Superdex peptide HR 10/30 100–20,000 Da 50 mmol L–1 Tris-HCl buffer, pH 8.0 0.6 mL min–1 100 µL 200–500 nm (230 nm working λ)
Elmer). The nebulizer gas flow and ion lens setting were optimized using on-board computer algorithms with the Elan 6000 software. Optimization of ICP–MS instrumental conditions is critical when multielemental determinations are performed. The sensitivity of the ICP–MS instrument was optimized using a multielemental standard solution containing the following elements: 10 µg L–1 Ba, Cd, Ce, Cu, Ge, Mg, Pb, Rh, Sc, Tb, and Tl of each element. The analytical conditions were as shown in Table 1. Possible polyatomic interferences in the analytical solutions were considered. The m/z values selected are shown in Table 1. For SEC–UV–ICP–MS coupling, the outlet of the SEC column was connected to the liquid sample inlet of the nebulizer using 0.25 mm i.d. PEEK tubing of 30 cm in length. The instrumental operating conditions were as given in Table 1. The closed vessel microwave digestion system used was a MES 1000 (CEM Corp., Matthews, NC, USA). A model RC5C centrifuge (Sorvall Instruments, DuPont) was used to accelerate the phase separation process in the extraction of the compounds. Reagents A 0.05 mol L–1 Tri(hydroxymethyl)aminomethane (TRIS) (Fisher Scientific, Fairlawn, NJ, USA) mobile phase solution was prepared by dissolving solid TRIS in 1000 mL of water adjusting it to pH 8.0 with HCl (Merck, Darmstadt, Germany) solution. A 0.1 mol L–1 3-cyclohexylamino-1-propanesulfonic acid (CAPS) (Aldrich, Milwaukee, WI, USA) mobile phase solution was prepared by weighing the right amount of the reagent dissolved in ultrapure water. The pH of the solution was adjusted to 10 with NaOH (Merck) solution. Lower concentrations were prepared by serial dilution with ultrapure water. A 1.0 mol L–1 sodium phosphate (Aldrich, Milwaukee, WI, USA) mobile phase solution was prepared by dissolving the reagent in ultrapure water and adjusting the pH to 9.0 with NaOH (Merck) solution. Lower concentrations were prepared by serial dilution with ultrapure water. All water used was deionized (18 MΩ cm) and prepared by passing through a NanoPure treatment system (Barnstead, Boston, MA, USA). Commercial chemicals were of analytical reagent grade and were used without further purification. All reagents were of analytical reagent grade and the presence of trace elements was not detected in the working range. Collection and preparation of nut samples Nut samples commonly consumed in USA were purchased from nearby markets, identified and fresh and/or processed nut samples
497 were selected: cashew (Anacardium occidentale), pecan (Carya pecan), peanut (Arachis hypogea), Brazil nut (Bertholletia excelsa), pine nut (Pinus edulis), almond (Prunus dulcis), black walnut (Juglans nigra), white walnut (Juglans cinerea), and sunflower (Helianthus annuus). Nut quality could be strongly affected by different parameters, such as stage of growth and pre- and post-harvest conditions. To somewhat compensate for the sample variability, samples were purchased on different days from local markets and prepared. The nut samples were washed with demineralized water, dried at 60oC overnight without addition of other ingredients, and finely ground using a household coffee grinder. All the instruments used were previously washed with a 10% (v/v) HNO3 water solution and then with ultrapure water. Elemental analysis of nut samples Lipids in the nuts were eliminated initially with organic solvents. Ground nut samples (20 g) were extracted with 100 mL of a chloroform–methanol (2:1) mixture after mixing them together for 15 min. The organic solvents were filtered and the residue was dried at room temperature. Three sub-samples were precisely weighed (1.0 g), transferred to PTFE vessels, and nitric acid (50%, 10 mL) was added. Digestion was performed in a microwave oven using the following procedure: microwave power was increased over four steps at 5-min intervals, starting at 25%, 45%, 55%, and ending at 65% where 100% power was 1000 W. Temperature limits of 120, 140, 150, and 160 °C were set for each of the four steps. Pressure limits for the four steps were 20, 80, 120 and 170 psig. A portion of this solution was filtered through a 0.2 µm surfactant-free cellulose acetate (SFCA) membrane filter. Total Cu, Mn, Ni, and Zn were determined by ICP–MS (conditions as stated in Table 1), using internal calibration with germanium (10 µg L–1) for analytical signal correction and two-point standard addition technique. Additionally, a recovery study was performed by adding 10 µg L–1 of each element. Recovery values were highly satisfactory and in the range of 95–105%. Total elemental content in different nut samples are shown in Table 2. It can be observed that copper concentrations were around the same value (~100–300 µg g–1) for different types of nut. On the other hand, manganese presented a more varied concentration distribution among the different nut samples with concentrations ranging from 9 up to 4780 µg g–1. These differences in the manganese concentration could be due to not only the type of nut but also some other factors such as the environmental conditions under which the nuts grew or variations in the mineral concentration of soils. Concentrations for Ni and Zn showed little difference among the nut samples analyzed and ranged between 44–141 µg g–1 and 189–671 µg g–1, respectively. It can be said that total elemental concentrations found in this work are in agreement with the results obtained by other authors for similar types of nut sample [11, 15, 19, 20]. In this work, multielemental determinations were also performed using ICP–MS on different nut samples extracted with dif-
Table 2 Total element concentrations in the nut samples analyzed in this work (95% confidence interval, n=6) Nut sample
Cu (µg g–1)
Mn (µg g–1)
Ni (µg g–1)
Zn (µg g–1)
Almond Black walnut Brazil nut Cashew Peanut Pecan Pine nut Sunflower White walnut
200±16 289±21 186±12 290±17 194±10 198±9.1 309±15 223±13 222±18
4780±110 546.2±22 12±0.4 9±0.2 2110±98 143±5.2 478±20 179±8.2 68±3.4
75±3.8 93±6.0 59±2.2 141±6.3 93±3.9 61±4.2 44±1.8 64±3.0 84±5.4
236±12 262±16 547±24 346±19 337±17 543±26 671±35 298±14 189±9.8
ferent media (NaOH and HCl) to calculate the extraction efficiency. It should be pointed out that an important variable that was not addressed in this study was the possibility of differing metal concentrations arising from different collection sites likely to have different soil metal concentrations. Since the number of nut samples in this study was limited, the elemental concentration values obtained cannot be generalized as “normal” concentrations commonly found in these nut species. Extraction of elemental species from nuts Dried and powdered nut sample (without lipids) were precisely weighed (0.1 g) in plastic tubes. The extraction of the elemental species from the nut samples was carried out by adding 2 mL of 0.1 mol L–1 sodium hydroxide. After agitating (Vortex) for 30 min at room temperature, the mixture was centrifuged at 3000 rpm for 10 min and filtered through a 0.45 µm filter. A volume of 100 µL of the supernatant phase was introduced into the SEC–UV–ICP–MS system for the fractionation analysis. SEC–UV–ICP–MS coupling for speciation analysis Elemental fractionation in the nut sample extractions was performed using a Superdex Peptide size-exclusion column (dimensions 300 mm height×10 mm diameter) (Amersham Biosciences, Piscataway, NJ, USA). Molecular masses of eluted compounds were estimated by column calibration using a set of calibration standards with MW within the interval of 189.2 to 13,680 Da. The samples were injected (100 µL) onto the Superdex column utilizing a Rheodyne 9025 injector with a 100 µL polyether ether ketone (PEEK) sample loop. The on-line use of the UV–visible detector prior to ICP–MS did not produce any significant dispersion of the analyte signal. Percentage distribution of the individual elements among different molecular size fractions was evaluated by relating the area of a particular peak to the total area under the chromatogram. Blank solutions consisting of each extractant solution were separately injected in the SEC column and no peaks corresponding to the elements analyzed in this work were found.
Results and discussion Extraction behavior of different media for elemental speciation The extractions of the elemental species under acidic and basic conditions were studied in this work. The elemental compounds were extracted from nut samples using 0.05 mol L–1 NaOH and 0.05 mol L–1 HCl solutions. An extraction time of 30 min with constant agitation was used to perform the extractions. Typical extraction efficiencies for NaOH and HCl are shown in Table 3. The extractability of elements for other types of nuts were found to be similar to those obtained for almond and for this reason is not shown in this work. Table 3 Typical extraction efficiency of different solutions for element speciation in nut samples (almonds)
Element
Cu Mn Ni Zn
Extractability (%) NaOH
HCl
86.6 95.2 92.0 82.6
23.3 34.9 13.8 33.4
498
It was expected that the 0.05 mol L–1 NaOH solution would extract both low molecular weight (LMW) and high molecular weight (HMW) elemental compounds. On the other hand, HCl solution mainly extracted LMW compounds due to the lower solubility of protonated compounds such as proteins [22]. In Table 3, it can be observed that the highest extractabilities were obtained when NaOH solution was used. In this case, the extraction values were in the range of 82.6–95.2%. The HCl extractions did not show such efficiency and in general were 13.8–34.9%.
Elemental speciation of nuts by SEC with sequential UV and ICP–MS detection The SEC separation of the elemental species was monitored sequentially with UV and ICP–MS detection. Initially, two SEC columns, Superdex 75 for resolution of HMW compounds (separation range 3000 to 70,000 Da) and Superdex Peptide for separation of LMW compounds, were tested for the separation of the elemental fractions in the nut extracts. Separation obtained by both columns was similar in the HMW region, whereas in the LMW region, Superdex Peptide column allowed a better fractionation of the elements. Therefore, the Superdex Peptide column was chosen for the remaining experiments. In order to obtain an accurate measurement of the relative MW obtained in the nut sample extracts, the Superdex Peptide column was calibrated with appropriate calibrants. Calibration standards included, lysozyme (14,400 Da), aprotinin (6500 Da), Gly6 (360 Da), and Gly3 (189.2 Da). The relation between retention time and the logarithm of MW was linear in the MW range of 0.19 to 14.4 kDa. All those compounds with relative MW equal or higher than 14.4 kDa eluted in the dead volume of the column whereas for MW lower than 0.19 kDa, the linear response was lost. Chromatographic separation conditions were evaluated for three different mobile phases; Tris buffer solution at pH 8.0, phosphate buffer solution at pH 7.5, and CAPS buffer solution at pH 10.0. The best resolution and minimal retention times were obtained when 50 mmol L–1 Tris buffer solution at pH 8.0 was used. This buffer solution avoided a possible protein precipitation and reduced hydrophobic interactions between these compounds with the column material that can affect the relative MW results. The effect of the mobile phase flow rate was also considered in this work in the 0.3 to 1.0 mL min–1 range. Flow rate values higher than 0.6 mL min–1 produced a significant deterioration of the resolution, whereas lower flow rates did not improve the resolution and the chromatographic times were increased considerably. Therefore, a mobile phase flow rate of 0.6 mL min–1 was used for the remaining experiments. Complete chromatograms were obtained within 25–30 min under these conditions. The chromatographic profiles for acidic (HCl) and basic (NaOH) extraction media with UV detection (230 nm) are shown in Fig. 1. It can be seen that in all cases HMW fractions were obtained for the NaOH extract. LMW fractions were separated in the retention time range of
Fig. 1 Typical fractionation profiles obtained after extraction with 0.05 mol L–1 NaOH solution: (a) Brazil nut; (b) cashew; and (c) white walnut. On-line UV–visible monitoring of the absorbance was performed at 230 nm. Other conditions were as shown in Table 1
20–25 min. The prominent fraction with relative MW higher than 13.7 kDa is likely to be proteins because of the high content of proteins in nuts; on the order of 10–20% in the whole sample including lipids (50–70%). But in the sam-
499 Table 4 Molecular size distribution of Cu, Mn, Ni, and Zn in different media for the extraction of fractions from common edible nut samples Nut sample
Almond
Cu
Mn
Ni
Zn
NaOH
HCl
NaOH
HCl
NaOH
HCl
NaOH
HCl
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
MW Area (kDa) (%)
13.8 10.0 2.5 1.2 0.1
31.8 53.7 9.0 4.6 0.9
7.5 1.2
31.4 68.6
13.4 4.1 1.8
43.3 3.0 34.0 1.7 22.7
57.6 13.6 42.4 1.7 1.3 0.9 0.5
58.6 8.1 11.4 8.4 13.4
7.3 1.6 1.2 0.9 0.6 0.4
15.9 12.7 15.8 4.1 35.6 1.7 12.0 1.3 1.8 18.8
81.8 7.4 8.3 2.5
11.3 6.9 3.1 1.6 1.2
2.4 8.5 35.2 12.3 41.5
Black walnut 12.8 9.9 4.6 1.4
10.3 49.0 28.8 11.9
3.8 1.3
27.1 72.9
10.4 3.6
92.2 0.3 7.8
100.0 10.4 1.3
50.7 49.3
4.0 1.2 0.7 0.4
9.3 12.6 22.4 9.9 19.4 4.5 48.9 1.4
13.3 50.1 27.1 9.5
3.3 1.4 0.3
29.3 18.5 52.1
Brazil nut
10.3 2.3
42.6 57.4
8.0 1.4 0.2
10.9 65.7 23.4
13.0 1.8
58.7 3.5 41.3 0.9
55.6 44.4
9.1 1.8 1.3
39.2 13.3 47.5
0.6
100.0 13.2 1.3
20.0 80.0
8.7 3.8 1.3
2.8 70.0 27.2
Cashew
12.8 8.3 1.4
45.7 37.4 16.9
6.9 1.3
33.4 66.6
13.0 4.0 1.9
18.6 3.3 63.9 1.7 17.5
75.8 12.8 24.2 1.3
17.2 82.8
1.1
100.0 13.0 4.0 1.4
42.0 32.2 25.7
8.9 3.3 1.3
7.6 57.8 34.6
Peanut
7.2 1.1 0.1
11.0 58.6 30.4
4.6 1.3
34.4 65.6
12.1 3.9 1.8
47.1 3.9 36.7 1.9 16.2
29.9 12.5 70.1 1.8 1.3 0.4
58.6 23.8 13.7 3.9
1.7 1.3 0.9 0.4
3.6 12.3 24.3 4.0 10.8 1.2 61.3
41.1 8.0 50.9
3.7 1.3
6.6 93.4
Pecan
11.9 5.4 1.4
80.5 12.4 7.1
4.8 1.4
28.6 71.4
11.9 4.1
84.2 3.6 15.8
100.0 12.6 1.3
77.9 22.1
–
12.8 4.2 1.4
65.9 6.3 27.9
1.4
100.0
Pine nut
13.6 9.3 1.4
50.5 46.6 2.9
10.3 1.3 0.1
8.0 69.1 22.9
13.0 3.8 1.8
7.6 3.3 88.8 1.7 3.6 1.2
30.5 13.2 35.6 1.8 33.8 1.3
49.3 25.5 25.2
1.7 1.2 0.6 0.4
17.9 13.2 21.0 3.9 12.4 1.3 48.7
55.3 21.4 23.3
3.1 1.3
28.1 71.9
Sunflower
12.7 1.6
82.5 17.5
7.0 1.3
33.0 67.0
12.0
13.4 1.3
75.9 24.1
1.3 0.4
92.5 12.3 7.5 1.3
51.0 49.0
1.3
100.0
White walnut 11.5 4.5 1.5
49.7 41.8 8.5
4.0 1.5
23.3 76.7
12.6 3.8
100.0 12.6 1.4
69.2 30.8
1.6
100.0 12.6 4.1 1.4
60.3 22.5 17.2
7.6 3.6 1.4
13.6 32.4 54.0
100.0 – 52.0 3.6 48.0
ples analyzed the content is higher as they were defatted to remove the bulk of the matrix. On the other hand, the HCl solution extracted only LMW compounds. In Fig. 1, it can be observed that primarily one fraction that corresponded to a relative MW about 7 kDa appeared. This fraction could be due to the presence of some metal-containing peptides or other similar LMW compounds which are extracted by using NaOH or HCl solutions. In some cases a LMW fraction at a retention time of 36 min was present in both NaOH and HCl. However, the MW for this fraction could not be determined due to the reason that it was eluted outside the calibration range. The chromatographic run time was extended to 60 min to ensure the elution of all fractions. However, no peaks were observed at retention times higher than 36 min. It has to be pointed out that especially in the LMW region no major differences
–
–
were observed in the fractionation profiles of compounds in different nut samples. Chromatographic fraction profiles of Cu, Mn, Ni, and Zn in NaOH and HCl extractions obtained for different nut samples analyzed are shown in Table 4. The fractions found are expressed in terms of relative MW and area percentage of each peak with respect to the total area of the chromatogram. The SEC separation of the extracted solutions was performed according to the analytical conditions given in Table 1. Appropriate chromatographic resolution was obtained in most of the cases. However, in those where the resolution was incomplete deconvolution software was used for the calculation of the areas.
500
Typical chromatographic profiles obtained for 63Cu are shown in Fig. 2. Copper was found to be mainly associated to a fraction with a relative MW equal or higher than 10–14 kDa (~50%) in the NaOH extracts. Due to the restriction of the Superdex Peptide column, these fractions could correspond to compounds of MW higher than
10–14 kDa. In fact, some publications have demonstrated the presence of HMW proteins, but the association of copper to these compounds has not yet been confirmed [2, 30, 31]. The rest of the 63Cu present in the injected solution was distributed among several LMW fractions and their MW depended on the different nut samples studied (Table 4). Interestingly, only in white and black walnut samples, about 30–40% of 63Cu was found to be associated with a
Fig. 2 Fractionation profiles of 63Cu in different extraction media: (a) black walnut, (b) peanut, and (c) white walnut. Other conditions were as shown in Table 1
Fig. 3 Fractionation profiles of 55Mn in different extraction media: (a) white walnut, (b) cashew nut, and (c) peanut. Other conditions were as shown in Table 1
Copper fractionation profiles
501
fraction of ~4.5 kDa, which could be useful for rough identification of this type of nut from the rest of the nuts. A LMW fraction of 1.1–2.3 kDa was found to be present in every nut sample analyzed. Fractionation of 63Cu in the HCl extracts showed an almost exclusive association of this element to a LMW fraction of 1.2–1.5 kDa (65–77% for the various nuts) as can be observed in Fig. 2 and Table 4. The remaining 23–35% of 63Cu was distributed among different HMW fractions, which are smaller fractions when compared to the LMW fraction around 1.3 kDa. These findings indicate that acidic solutions are more effective media for extraction of LMW and inorganic elemental forms than the basic NaOH solution. Manganese fractionation profiles Fractionation profiles of 55Mn in nut samples can be summarized into mainly two fractions of different MW (Table 4). Typical fractionation of 55Mn in the HCl and NaOH extracts is shown in Fig. 3. In the case of basic extraction, 55Mn was associated to a HMW fraction of 10.4–13.4 kDa (7.6–100% for the various nuts) and a LMW fraction of 3.6–4.1 kDa (7.8–88.8%) among the different types of nut (Table 4). Another LMW fraction was also observed, however for most of the nut samples the association of 55Mn to this fraction was not significant compared to the other higher MW peaks. No 55Mn-containing HMW fraction was obtained when the HCl extract was injected onto the size-exclusion. In this case all 55Mn was found to be associated in two fractions with a relative MW around 0.9– 1.9 kDa (25–70%) and 3.0–3.9 kDa (30–100%) among the different nuts samples studied (Table 4). The chromatographic profiles of 55Mn-containing fractions in ICP–MS and UV detection at 230 nm matched well. 55Mn observed in the ICP–MS chromatogram is as a metal associated organic compound. A comparison of the results obtained for HCl and NaOH extractions confirmed that 55Mn is likely associated to proteins, which are effectively solubilized in NaOH but not in acidic media. Nickel fractionation profiles The chromatographic profiles of 58Ni were significantly different from all other elements studied in this work depending on the extraction media used (Fig. 4). Thus, 58Ni was found only in a HMW fraction around 9.1–13.6 kDa and a LMW fraction around 1.3–1.8 in the NaOH extraction solution (Table 4). The percentage areas for HMW and LMW fractions varied among the several nuts samples analyzed between 17.2 to 77.9% and ~50%, respectively. However, when the HCl extract was injected onto the size-exclusion column almost no 58Ni-containing fractions were found. These results are in good agreement with the extractability values shown in Table 3, where only ~13% of the 58Ni present in nut samples was extracted with HCl media. These 58Ni-containing fractions were observed in the LMW region of 0.4–1.8 kDa. 58Ni in nut
Fig. 4 Fractionation profiles of 58Ni in different extraction media: (a) pecan nut, (b) peanut, and (c) white walnut. Other conditions were as shown in Table 1
samples is mainly attached to HMW proteins, which are not extracted in acidic conditions. Zinc fractionation profiles Chromatographic profiles of 68Zn in NaOH and HCl extracts for different nut samples are shown in Fig. 5. De-
502
tion with a MW of 4.0–4.5 kDa. Extraction of nut samples in acidic conditions significantly reduced the number of 68Zn-containing fractions in the chromatograms (Fig. 5). Only the LMW fraction around 1.3 kDa was extracted with the 0.05 mol L–1 HCl solution. Fractions with MW higher than 1.3 kDa could also be observed, however in those cases the peak area of 68Zn was insignificant compared to the 1.3 kDa fraction. Comparison of the fractionation patterns of 68Zn in basic and acidic extraction media confirmed that this element as well as for Cu, Mn, and Ni is mostly associated to HMW compounds, which could be proteins based on their solubilities in basic conditions.
Conclusions Speciation studies of Cu, Mn, Ni, and Zn in different types of nut sample including, cashew, pecan, peanut, Brazil nut, pine nut, almond, black walnut, white walnut, and sunflower were performed using SEC–UV–ICP–MS. A broad distribution of these elements among several MW fractions was observed. A basic association of all the elements analyzed to HMW fractions was observed indicating the possible attachment of these elements to proteins, already known to be in high amounts in nuts. The elemental fractionation by size using SEC is only an initial step in identifying and characterizing the organometallic species associated with each of these fractions. The application of multidimensional separation techniques and the use of identification tools such as other mass spectrometries are required to identify these unknown species. Future studies will be undertaken to characterize and determine the individual elemental species, including proteins, present in the analyzed nuts. Acknowledgments The authors would like to thank Agilent Technologies for continuing support of our work. We would also like to acknowledge NIEHS grant #ES04908 for partial funding of this research.
References
Fig. 5 Fractionation profiles of 68Zn in different extraction media. (a) white walnut, (b) brazil nut, and (c) sunflower nut. Other conditions were as shown in Table 1
tailed information on the fractionation patterns of 68Zn is also given in Table 4. This element was found to be distributed differently among two or three MW fractions depending on the type of nut studied when NaOH extract was injected. For only sunflower and Brazil nut samples, 68Zn was distributed in a HMW fraction of 12–13 kDa and another with a relative MW of 1.3 kDa. For the rest of the nut samples it was found that there is an additional frac-
1. Kris-Etherton PM, Zhao GX, Binkoski AE, Coval SM, Etherton TD (2001) Nutr Rev 59:103 2. Jenkins DJA, Kendall CWC, Marchie A, Parker TL, Connelly PW, Qian W, Haight JS, Faulkner D, Vidgen E, Lapsley KG, Spiller GA (2002) Circulation 106:1327 3. Fraser GE (2000) Asia Pac J Clin Nutr 9:S28 4. Willett WC (2000) Proc Soc Exp Biol Med 225:187 5. Cardozo MS, Li BW (1994) J Food Compos Anal 7:37 6. Dziezak JD (1989) Food Technol 43:94 7. Schwartz RS, Hecking LT (1991) J Anal At Spectrom 6:637 8. Sadecka J, Polonsky J (1999) J Chromatogr A 834:401 9. Koletzko B, Aggett PJ, Bindels JG, Bung P, Ferre P, Gil A, Lentze MJ, Roberfroid M, Strobel S (1998) Br J Nutr 80:S5 10. Roberts HR (1981) Food safety. Wiley, New York 11. Zaidi JH, Arif M, Fatima I, Qureshi IH (2002) J Radioanal Nucl Chem 253:459 12. Seiler HG, Sigel A, Sigel H (1994) Handbook on metals in clinical and analytical chemistry. Marcel Dekker, New York 13. Barclay MNI, MacPherson A, Dixon J (1995) J Food Compos Anal 8:307
503 14. Chang JC, Gutenmann WH, Reid CM, Lisk DJ (1995) Chemosphere 30:801 15. Pennington JA, Schoen SA, Salmon GD, Young B, Johnson RD, Marts RW (1995) J Food Compos Anal 8:171 16. Wakeling LT, Mason RL, D’Arcy BR, Caffin NA (2001) J Agric Food Chem 49:1277 17. Ridge C, Akanle O, Spyrou NM (2001) J Radioanal Nucl Chem 249:67 18. Wei YY, Chung C (1997) J Radioanal Nucl Chem 217:45 19. Vallilo MI, Tavares M, Aued-Pimentel S, Campos NC, Neto JMM (1999) Food Chem 66:197 20. Plessi M, Bertelli D, Monzani A, Simonetti MS, Neri A, Damiani P (1999) J Food Compos Anal 12:91 21. Vonderheide AP, Wrobel K, Kannamkumarath SS, B’Hymer C, Montes-Bayon M, De Leon CP, Caruso JA (2002) J Agric Food Chem 50:5722
22. Kannamkumarath SS, Wrobel K, Vonderheide A, Caruso JA (2002) Anal Bioanal Chem 373:454 23. Fecher P, Ruhnke G (1998) At Spectrosc 19:204 24. de Leon CAP, Montes-Bayon M, Caruso JA (2002) J Chromatogr A 974:1 25. Szpunar J, Pellerin P, Makarov A, Doco T, Williams P, Lobinski R (1999) J Anal At Spectrom 14:639 26. Szpunar J, Lobinski R (1999) Pure Appl Chem 71:899 27. Sanchez LF, Szpunar J (1999) J Anal At Spectrom 14:1697 28. Szpunar J (2000) Analyst 125:963 29. Smolin LA, Grosvenor MB (2000) Nutition: science and application, 3rd edn. Saunders, Orlando 30. Morgan WA, Clayshulte BJ (2000) J Am Diet Assoc 100:312 31. Almario RU, Vonghavaravat V, Wong R, Kasim-Karakas SE (2001) Am J Clin Nutr 74:72
