Characterization of a cadmium-zinc complex in lettuce ... - Springer Link

9 1995 by Humana Press Inc. All rights of any nature whatsoever reserved. 0163-4984/95/4801~013 $07.40

Characterization of a Cadmium-Zinc Complex in Lettuce Leaves ILDA MELO MCKENNA *'1'2 AND RUFUS L. CHANEY 1

lUSDA-Agricultural Research Service, Environmental Chemistry Laboratory, Beltsville, MD 20705; and 2Present address: Laboratory of Comparative Carcinogenesis, Building 538, Room 205-E, National Cancer Institute, Frederick, MD 21702-1201 Received May 3, 1994; Accepted June 30, 1994

ABSTRACT Vegetable food contributes a higher amount of daily cadmium (Cd) intake in humans than food of animal origin. The bioavailability of plant Cd depends on the content of plant zinc (Zn). The mechanism by which increased plant Zn lowers the intestinal absorption of plant Cd could be mediated by changes in the chemical speciation of Cd or Zn in plant edible tissues, including Zn-induced phytochelatin synthesis. To test this hypothesis we investigated the chemical speciation of Cd and Zn in leaf extracts of lettuce grown under 10 ~tM of Cd accompanied by 0.32 or 31.6 ~tM Zn in nutrient solution. Gel filtration chromatography of the low- or high-Zn leaf extracts yielded a major low molecular weight Cd-Zn complex that eluted at similar elution volume. Compared to low-Zn leaf extracts, high-Zn leaf extracts contained a higher proportion of Zn incorporated into high molecular weight components, and higher content of the amino acids Cys, Glu, Gly, and Asp in the low molecular weight Cd-Zn complex. The peptides isolated by high performance liquid chromatography (HPLC) of the Cd-Zn complex from the low- or high-Zn leaf extracts did not have an amino acid composition identical to phytochelatins. We concluded that 1. Sequestration of Cd or Zn via phytochelatin does not occur in leaves of lettuce containing levels of those metals representatives of Z n - C d or Cd-only contaminated crops; and 2. Higher Cys, Glu, Gly, and Asp content in high-Zn than lowZn leaves could lower Cd absorption in animals fed high-Zn crop diets, by enhancing metallothionein synthesis or changing Cd or Zn speciation in the animal gut. *Author to whom all c o r r e s p o n d e n c e and reprint requests should be a d d r e s s e d . Biological Trace Element Research

13

Vol. 48, 1995

McKenna and Chaney

14

Index Entries: Cadmium; zinc; chemical speciation; cadmium bioavailability; lettuce; phytochelatin.

INTRODUCTION Plant-derived foods are the principal contributors of daily Cd intake to nonsmokers and populations not occupationally exposed to Cd. Gastrointestinal absorption of dietary Cd may vary between less than 1% to higher than 15% of Cd in food, depending on the nutrients in the diet, nutritional status of consumers, and the chemical form of ingested Cd (1-5). Among the nutrients that can interact with Cd bioavailability, Zn is an important factor in terms of assessing human exposure to environmental Cd, because of the presence of high levels of Zn in most Cd-contaminated sites (6). Leafy vegetables, such as lettuce, can readily accumulate both Zn and Cd in leaves when grown in Zn-Cd contaminated sites (7). Previously we found that increased plant intrinsic Zn lowered the retention of plant Cd in the kidney, liver, and jejunum-ileum of Japanese quail fed lettuce or spinach diets (8). We also observed that intrinsic plant Zn had a stronger effect than inorganic Zn in lowering Cd absorption in quail for equal levels of Cd and Zn in the diets. Our results on plant Cd bioavailability to quail suggested two hypotheses: 1. The chemical speciation of Zn or Cd in leaves of plants grown under subphytotoxic levels of those metals differs from that in leaves of plants exposed to low Zn for equal levels of Cd in the growth medium; or 2. High levels of intrinsic Zn in plant leaves changes the chemical speciation of plant Cd in the animal gut, compared to equal extrinsic Zn. In order to test the first hypothesis we conducted the present research on the chemical speciation of Zn and Cd in leaves of lettuce grown under subphytotoxic levels of Cd accompanied by low (nondeficient) or high (subphytotoxic) levels of Zn in nutrient solution. Some of the hypothetical changes of Cd or Zn speciation in leaves of plants grown in Zn-Cd polluted environments, compared to those of Cdonly polluted sites, could be mediated by the synthesis of metal-binding peptides referred to as class III Metallothionein (MT), (poly [gamma-glutamyl-cysteinyl] glycine, or more commonly phytochelatins (PCs). These polypeptides are nontranslationally synthesized, in contrast to the geneencoded class I and class II MTs, and in general they are comprised of the amino acids glutamic acid, cysteine, and glycine [7(Glu-Cys)n Gly] in the proportion of n:n:l. However, in some cases Gly can be replaced by Ala or Ser (9-12).

Biological Trace Element Research

Vol. 48, 1995

Cadmium-Zinc Complexes in Lettuce Leaves

15

Most of the research on PCs has been performed on Cd-resistant plant cells, with a few studies on plant roots exposed to acutely toxic levels of Cd (13,14). Not only do plant cells differ from leaves or roots in the synthesis of metal-binding complexes, but roots also seem to produce thiol-rich polypeptides faster and in greater concentrations than do leaves (12,15). Additionally, little data is available on the ability of various metals other than Cd and Cu to specifically induce and bind to PCs in plants (12,14,16). Moreover, PCs have not been investigated in terms of assessing plant Cd transfer to humans, nor has it been shown that there is induction of Zn- or Cd-PCs in leaves of plants whose roots are exposed chronically to Zn-Cd environmental contamination. Therefore, in the present study we investigated PC induction in leaves of plants grown under Zn and Cd levels representative of Zn-Cd contamination or Cd-only pollution, in an attempt to understand the biochemical mechanism by which plant Zn lowers plant Cd bioavailability to animals.

MATERIAL AND METHODS Plant Growth Seeds of Romaine lettuce (Lactuca sativa, L., var. longifolia, cv. Parris Island) were germinated on moist seed germination papers for 1 wk in a growth chamber. Seedlings were transplanted to polyethylene buckets containing 8 L of complete nutrient solution. There were four seedlings per pot. Seedlings were supported by polyurethane foam in 2-cm holes in the covers, and all openings were covered with black polyethylene to keep light from reaching the nutrient solution and roots. The nutrient solution contained 2.5 mM KNO3, 2.5 mM Ca(NO3)2, 1 mM MgSO4, 10 W~d K2HPO4, 100 ~tM KC1, 15 ~tM H3BO3, 2.0 ~tM Mn, 10 ~tM FeEDDHA [ethylenediamine di (0-hydroxyphenylacetic acid)], 0.2 ~tM Mo, 0.2 ~tM Cu, 0.2 ~ Ni, and 0.2 ~tM Co. EDDHA was used to keep the Fe soluble during the growth period; it is a strong chelator of Fe3+. The chelator EGTA [ethyleneglycolbis(ethylamine)tetraacetic acid] was used to buffer Zn, Cd, and the other micronutrient cations. The use of chelator buffering to control microelement availability to plants is described by Parker et al. (17). In this method, 50 ~tM EGTA in excess of the strongly chelated microelement cations is added to buffer their activity; this "excess" EGTA is filled with Ca, and microelement activity is a function of the ratio of the element to Ca. Calcium is in large supply in the nutrient solution so little change occurs in the activity of the microelements. Because EGTA chelates Cd much strongly than Zn, it can be used to supply controlled low levels of Cd similar to soils, and varied levels of moderate to high Zn. Zinc was supplied at 0.32 and 31.6 ~tM ZnEGTA for the low and high Zn treatments, respectively; activity of


Vol. 48, 1995

16

McKenna and Chaney

Zn 2+ was buffered at 10-7.08 and 10-5.17 mol/L, respectively. Cadmium was added at 10 ~M CdEGTA to both solutions, providing 10-9.20 tool Cd2+/L. Each Zn-Cd treatment combination was replicated. Solution pH was buffered at 6.0 with MES (2 mM) and daily adjusted to that value when needed. The initial nutrient solution was not replaced during plant growth, but a solution that supplied macro- and micronutrients, except Cd, was added daily in all buckets (after Asher and Blarney, 18). In the first 20 d of the experiment, we added the following amount of nutrients daily (~tmol pot-l): 1323N, 523 K, 139 Ca, 120 P, 107 Mg, 1.72 B, 0.044 Cu, 0.678 Mn, and 0.256 Zn. During the remaining growth until harvest (15 d), twice these amount of nutrients were added daily. Nutrient solutions were continuously aerated and deionized water was added as needed to maintain 8 L of nutrient solution in each pot. Plants were grown in a growth chamber under 16 h / 8 h light-dark, a day/night temperature of 27/21~ and 546 ~tE m-2/s at plant level of photosynthetically active radiation provided by cool-white fluorescent and incandescent lamps. Five weeks after addition of Zn and Cd treatments, young leaves (leaves above the first six true leaves) were harvested, pooled per pot, and stored at -20~

Extraction Procedure Plant grinding and leaf extract preparation were conducted in a cold room at 4~ under a stream of N2. Leaves were first ground coarsely with a large pestle and mortar to homogenize plant material and subsamples were placed into small aluminum pouches, weighed, immediately frozen in liquid N2, and stored at -70~ until leaf extract preparation. Subsamples were also taken for total Cd and Zn analysis and mineralized with HNO3:H202. To prepare the tissue extracts frozen leaf tissue was then ground finely in a small mortar and pestle and homogenized with N2 purged 100 mM HEPES (N1:1 v / w e t wt), pH 8.0, containing 1 mM phenylmethansulfonyl fluoride (PMSF) and 1% (v:v) Tween 20 (Rauser, personal communication). The homogenate was centrifuged at 25,000 rpm for 20 min at 4~ Supernatant (extract 1) was applied to a column (83 x 2.5 cm) of Sephacryl 100 HR (103-105 fractionation range) (Pharmacia LKB Biotechnology Uppsala, Sweden) and eluted at constant pressure with 10 mM HEPES containing 0.3M KC1 (-50 mL/h), pH 8.0, at 4~ The pellet resulting from the first centrifugation was suspended in 10 mL of N2 purged 10 mM HEPES, pH 8.0, containing 1% Tween, and centrifuged again at 25,000 rpm for 15 min at 4~ The supernatant (extract 2) was subjected to the same gel-filtration chromatography as extract 1, but it was not further purified or characterized. Three more extracts (3-5) were prepared in 10 mM HEPES, pH 8.0, 1% Tween from the second extract pellet to calculate the total amount of Biological Trace Element Research

Vol. 48, 1995


17

Zn and Cd that were buffer soluble. The buffer insoluble pellet was suspended in 0.1M HC1 twice to recover the acid soluble Zn and Cd. The final acid insoluble (0.1M HC1) pellet was mineralized with HNO3:H202 and analyzed for Cd and Zn. All metal analyses were performed in an atomic absorption spectrophotometer (AAS) equipped with 3-slot burner (to compensate for sample salt concentration) and background correction. To avoid matrix interference effects in AAS analysis, standards were prepared using the same buffer and salt concentration as the samples. Fractions of 100 drops eluting from the Sephacryl column were collected at 4~ and monitored for Cd and Zn. Fraction collector tubes were weighed before and after fraction collection to calculate fraction volume, since the drop volume varied during column elution. Buffers were stirred with Chelex 100 Resin (BioRad, Hercules, CA) and then filtered through 0.22-gm porous filters to remove any Zn or Cd contamination. The gel filtration column void volume was determined using Blue Dextran in 100 mM HEPES. The column was also calibrated with Zn-Cd reduced glutathione complex, Zn-Cd citrate complex, and free Zn and Cd prepared in 100 mM HEPES and chromatographed under the same conditions as leaf extracts. Fractions that eluted later than the column void volume and contained the major Zn-Cd peak from leaf extract were pooled (Zn-Cd binding complex), monitored for UV absorption (Beckman [Fullerton, CA] DU spectrophotometer), and stored at -70~

High Performance Liquid Chromatography (HPLC) The pooled fractions containing the Zn-Cd complex isolated by gel filtration chromatography of leaf extracts were concentrated by lyophilization, resuspended in 0.05% H3PO4, acidified to pH N2 with concentrated H3PO4, filtered through a 0.2-~tm polytetrafluoroethylene (PTFE) syringe filter, and analyzed by HPLC. Peptide separation was resolved by reversed-phase HPLC using an Alltech (Alltech Associates Inc., Deerfield, IL) Econosphere C18, 5 ~t column (25 cm length x 4.6 mm id) and detection at 220 nm. The solvents used were 0.05% phosphoric acid (v/v) (solvent A) and acetonitrile (solvent B). After injection of the aqueous sample (50-200 ~tL), a linear gradient 0-20% solvent B (v/v) in solvent A was applied from 0 to 15 min followed by 20% B in A from 15 to 25 min, at a constant flowrate of 1.0 mL/min. Solvents were filtered through a Corning 0.2-~tm membrane before being used and sparged with N2 during HPLC operation. After each sample run, the column was returned to initial conditions (0-100% A in 10 min) at a flowrate of 1.5 m L / m i n and then allowed to equilibrate in 100% A for 10 min at 1.0 mL/min, before new sample injection. Multiple runs were performed to collect adequate quantities of each peptide. Selected peak fractions were pooled, frozen, and lyophilized in preparation for amino acid composition analysis. To compare the elution profile of samples with known compounds, standards of Biological Trace Element Research

Vol. 48, 1995

18

McKenna and Chaney

reduced and oxidized glutathiones and cysteine were prepared in HEPES, lyophilized, solubilized in 0.05% H3PO4, acidified and passed through HPLC along with the buffer HEPES.

Amino Acid Analysis Amino acid composition analysis was performed by AAA Laboratories (Mercer Island, WA). Peptides were subjected to 20-h 6N HC1/0.05% mercaptoethanol hydrolysis at 115~ derivatized, and separated by ionexchange HPLC. Cysteine was determined as cysteic acid after performic acid oxidation at 50~ for 15 min. This method was found to give identical values for cysteine as the original procedure of performic acid oxidation for 4 h at 4~

RESULTS Lettuce seedlings developed well and leaves of plants grown under low-Zn level contained 36 ~tg Cd and 31 ~tg Z n / g dry wt, whereas leaves of plants grown under high-Zn level had 35 ~tg Cd and 190 ~tg Z n / g dry wt. Leaf extracts used in gel-filtration chromatography did not have the brown coloration characteristic of phenolic compounds as happens in highly metal-stressed plant material. We found it difficult to investigate the chemical speciation of total leaf Zn and Cd under conditions that do not dissociate the native metaMigand complex. In fact, about 48% Cd and 30% Zn present in leaves were not buffer soluble (5 extractions with HEPES, pH 8.0), but soluble only in 0.1M HC1, and about 1% of the total Cd and 4% of the total Zn were insoluble in 0.1M HC1. In addition, buffer soluble Cd was in general lower in high- than low-Zn leaves. From the 51% of Cd and 66% of Zn in leaves that were buffer soluble, about 35% of Cd and 45% of Zn were found in the first extract. Also, we observed that buffers differ in their relative capacity to extract Zn and Cd. For example, with TRIS-HC1, pH 8.6, we could extract more Cd (66 vs 51%) but less Zn (45 vs 66%) than with HEPES, pH 8.0. Similar elution profiles were obtained by gel filtration chromatography of the first and second extracts of the low- or high-Zn leaves. Only results corresponding to the first extract are presented in figures and tables. Gel filtration chromatography of leaf extracts yielded a major lowmolecular-weight Zn-Cd binding complex which eluted at similar elution volumes in the low-Zn and high-Zn leaf extracts and in both treatment replicates (Table 1). Zinc and Cd peaks overlapped, but the maximum of Cd eluted one or two fractions before that of Zn (Table 1). A smaller proportion of Cd and Zn was associated with high molecular components in both low- and high-Zn extracts (Fig. 1). More Zn eluted in the high-molecular-weight fraction in case of high- than low-Zn extracts. In fact, the ratio between the amount of Zn associated with the Biological Trace Element Research

Vol. 48, 1995


19

Table 1 Elution Volume and Percentage of Total Zn and Cd Recovered in Column Eluent Contained in Peaks of Zn-Cd Complexa Elution volume Peak Max, mL

Low-Zn leaves Rep 1 Rep 2 High-Zn leaves Rep 1 Rep 2 Zn-Cd glutathione Free Zn and Cd Zn-Cd citrate

Percent Total metal recovered/peak

Zn

Cd

Zn

Cd

356 356

345 345

84 88

39 42

356 353 384 684 768

345 347 372 785 951

57 71 83 82 73

35 58 97 60 74

aFrom leaf extracts, and standards: metal-glutathione, free metal, and metal-citrate.

low- and high-molecular-weight components was 20-30 in case of lowZn extracts, but only two-three in high-Zn extracts. The high-molecularweight Zn and Cd components eluted broadly in the region of the column exclusion volume (Ve/Vo between 1.0 and 1.5), simultaneously with chlorophyll. The low-molecular-weight Zn-Cd complex from leaf extracts eluted close but earlier than the complex Zn-Cd reduced glutathione (Zn-Cd-GSH), with the maximum Zn or Cd occurring about 30 mL earlier than in Zn-Cd-GSH (Table 1, Fig. 1). Virtually no Zn eluted after the low-molecular-weight complex in both the low- and high-Zn leaf extracts. On the other hand, some amount of unresolved Cd eluted later in the elution profile, in a position corresponding to free Cd or Cd-citrate (Table 1, Fig. 1). Whereas Cd and Zn in the complex Zn-Cd-GSH eluted in sharp peaks, in Zn-Cd-citrate and to a less extent as free metals they eluted broadly and with lower recoveries. Adsorption dependent interactions between the gel (Sephacryl) matrix and Zn or Cd might have occurred not only with free metals but also metal-citrate complex owing to the weak Zn and Cd citrate bonds. Approximately 90% Zn and 73% Cd of leaf metals loaded into S-100 (about 8.85 ~tg Cd and 8.65 ~tg Zn in low-Zn extracts, and 4.57 ~tg Cd and 28.77 ~tg Zn in high-Zn extracts) were recovered in the column eluent. In percentage of the total Zn and Cd recovered in the profile, the low-molecular-weight complex contained about 85% of the Zn and 41% of the Cd in the low-Zn extract, and 64% of the Zn and 45% of the Cd for the high-Zn extract. The Zn-Cd containing fractions of the low-molecularweight complex were pooled as shown in Fig. 1 for further characterization and purification and this component will be referred to as the Cd-Zn complex. Biological Trace Element Research

Vol. 48, 1995

McKenna and Chaney

20

/

~ o

,

~

,~

~.

&

~.~~

/

~

~

~

"~-' - ~

r~ ~.~

.

~

~.~

~~ ~,--~ ~ ~

~

pe~e^ooel:l P 0 lelO.L

,-

.~-

,~

~

~-~.. ~

9 O

o

=

~ ~ ~^~o 0,,~

~

0 - ~

.

~

~. :

~ ~ ~ "~

~

O

T--

6

T--

~

I~ o

~ x: ~ N ~ r~ ~ ~ --.~

O

o

~

o

.~'~ ~

pe4eAOOal:l u z Ira,0.1. %


Vol. 48, I995


21

The ultraviolet absorption spectrum of the Cd-Zn complex isolated from the low or high-Zn extracts showed a maximum absorbance around 220 nm and negligible absorbance at 280 nm, which corresponds to the lack of aromatic amino acids (Fig. 2). The ratio A25o/A28o was 2.42 for the high-Zn and 1.54 for the low-Zn extracts. The absorbance between 250 and 260 nm did not show the characteristic absorption shoulder of Cdmercaptide bonding, as in Cd-MT (19) or Cd-PCs (9). However, the absorbance at 250 nm diminished to 67% of the native complex in highZn Cd-Zn complex and to 86% in low-Zn Cd-Zn complex by acidification to pH N1 and increased to approximately the original value by back titration to pH N8.5 (Fig. 2). The absence of the characteristic UV absorption spectrum of cadmium mercaptide formation in the region 250-260 nm could be related to the low Cd concentration. Lane et al. (20) generated that spectrum in the wheat embryo Zn-containing metallothionein eluted from gel-filtration only when 1 mM Cd was added. A similar chromatogram was obtained by reverse phase HPLC of the Cd-Zn complex from low- or high-Zn leaf extracts (Fig. 3). The retention times of HEPES, cysteine, GSH, and oxidized glutathione (GSSG) are also shown in Fig. 3. The resolution between HEPES and cysteine peaks lowered as their concentration increased, forming a double ramp similar to peak 1 of Cd-Zn complex. The average (8-11 runs) retention times (s) of peaks 1-7 were: 3.28, 6.38, 7.78, 9.90, 19.07, 19.90, and 22.21 in case of the complex from low-Zn extract, and 3.27, 6.39, 7.74, 9.65 19.11, 19.77, and 22.35 for the high-Zn extract complex. The average retention times found for HEPES, cysteine, GSH, and GSSG were: 2.96, 3.53, 7.94, and 18.74, respectively. The elution profile showed some increase in the baseline at 100% of eluent B (20% acetonitrile). Peak 3 was the predominant fraction and eluted close but earlier than the standard GSH. It eluted at approx 50% of eluent B (20% acetonitrile). Peaks 5 and 6 eluted closely and usually not fully resolved. Only peaks 3, 5, 6, and 7 were chosen for amino acid analysis, based on their absorbance and position in the elution profile. The amino acid composition of the HPLC isolated peptides is presented in Table 2. The most abundant amino acid in peak 3 of both lowand high-Zn metal complexes was Val, followed by Glu and Ala. The amino acid composition of peak 7 did not differ significantly between the low- and high-Zn Cd-Zn complex with Ala, Glu, and Gly as the predominant amino acids, but Glu was higher and Ala lower in high- than low-Zn Cd-Zn complex. The major differences in the amino acid composition between the two samples were in peaks 5 and 6. Cysteine in peak 5 was about tenfold higher in the high- than low-Zn Cd-Zn complex, and Glu in peak 6 was 23% in the high-Zn Cd-Zn complex and absent in low-Zn Cd-Zn complex. On the other hand, peak 6 of low-Zn Cd-Zn complex contained predominantly His at a concentration 40-fold higher than in high-Zn Cd-Zn complex. Cysteine, Glu, and Gly were the


Vol. 48, 1995

McKenna and Chaney

22

0 0

~

o

~ow l |

~o:=

~

.

o>o

Od'~

N~

•

~

I

~

C~.~o

:r

"l

I ,o"

I

.~

o

o~

~.%~ .~

~ ~

~:

I i

o. oo

eoueqJosqv


Vol. 48, 1995

Cadmium-Zinc Complexes in Lettuce Leaves 8.1

i

. . . .

i

23

. . . .

i

. . . .

A 7

8.85

8 1

E e-

8.1

.

.

.

.

.

'

.I .

r

.

. . . .

i

i

i

. . . .

i

i

i

i

i

8.85

i

i

. . . .

'I

o

O C

.

B 7

.,Q O

8

< 8.1 CYS GSSG

H

GSX

8.85

8 L--L

8

5

18

15

28

,

A

9

25

Elution Time (min) Fig. 3. HPLC elution profiles of the Cd-Zn complex isolated by gel filtration chromatography of low-Zn (A) and high-Zn (B) leaf extracts, and the standards (C): Hepes, Cys, GSH, and GSSG. Fractions 3, 5, 6, and 7 were collected for amino acid analysis. major amino acids in peak 5 of high-Zn (N88% of the total) in a ratio of 1 Cys:l.7 Glu:l.6 Gly. The approximate concentration (nmoles/mL) of each amino acid in the Cd-Zn complex isolated by gel filtration is presented in Table 3. The


Vol. 48, 1995

McKenna and Chaney

24

Table 2 Amino Acid Composition of Peptides 3, 5, 6, and 7a Proportion of total residues, percentb High-Zn extracts

Low-Zn extracts

Amino acid

(3)

(5)

(6)

(7)

(3)

(5)

(6)

(7)

Alanine Aspartic acid Cysteinec Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Proline Serine Threonine Valine

12.2 6.7 2.9 12.7 7.6 1.1 0.7 0.8 2.3 0.2 0.4 7.6 4.0 40.8

5.5 1.2 20.2 35.1 32.4 0.2 0.5 0.4 0.1 0.0 0.5 1.4 0.6 1.8

25.1 6.0 4.1 23.3 15.6 1.0 1.7 2.2 0.4 0.0 4.0 7.5 3.2 5.9

13.8 9.8 3.6 23.0 14.6 1.6 2.2 2.7 3.5 0.0 6.8 9.5 4.1 4.9

13.4 1.2 4.6 12.5 6.3 1.0 0.7 9.8 1.1 0.0 3.0 5.0 3.0 38.3

13.0 3.2 2.4 36.9 31.8 0.1 1.1 0.2 2.1 0.0 2.4 2.8 1.0 3.1

9.0 0.0 1.7 0.0 12.0 45.6 0.0 0.2 5.0 1.3 8.9 8.1 1.0 7.3

29.5 7.2 3.2 17.8 13.7 1.1 1.8 3.6 0.6 0.8 3.8 9.0 2.0 5.8

aIsolated by HPLC of the Cd-Zn complex from high-Zn and low-Zn leaf extracts. bArginine, phenylalanine, and tyrosine were not detected. cDetermined as cysteic acid after performic acid oxidation.

Table 3 Estimated Amino Acid Concentrations (nmoles/mL) in the Cd-Zn Complexesa

Alanine Aspartic acid Cysteine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Proline Serine Threonine Valine

High-Zn

Low-Zn

5.40 2.35 4.92 11.54 9.25 0.39 0.39 0.45 0.68 0.05 0.56 2.64 1.31 10.31

7.58 1.30 1.71 9.67 7.69 2.15 0.46 2.53 0.81 0.12 1.68 2.55 1.02 9.85

aIsolated by gel filtration chromatography of the high-Zn and low-Zn leaf extracts.


Vol. 48, 1995


25

estimated concentrations were calculated based on the sample volume used in the HPLC for amino acid analysis, and the total amount of each amino acid in the four HPLC peaks analyzed. The Cd-Zn complex from high-Zn extracts contained higher concentration of Cys, Glu, Gly, and Asp, but lower concentration of Ala, His, Leu, and Pro than that from low-Zn extracts. The content of Cys was about threefold higher in the high- than low-Zn Cd-Zn complex, which could explain its higher absorbance at 250 nm.

DISCUSSION Leaf extracts of lettuce exposed chronically to low- or high-Zn levels and identical levels of Cd, showed some differences in the composition of the Zn-Cd binding components isolated by gel filtration chromatography and HPLC. Compared to low-Zn extracts, high-Zn extracts contained a higher proportion of Zn incorporated into highmolecular-weight components, and the low-molecular-weight Cd-Zn complex contained a higher content of the amino acids Cys, Glu, Gly, and Asp. We did not find that Zn or Cd were sequestered via phytochelatin (PC) in leaves of lettuce containing levels of those metals representatives of Zn-Cd contaminated crops (35 ~tg Cd, 190 ~tg Z n / g dry wt) or Cd-only contaminated crops (36 ~tg Cd and 31 ~tg Z n / g dry wt). The absence of PCs in our lettuce leaves is consistent with our lack of success in following the extraction procedure that Rauser (personal communication) found suitable for isolating PCs from Cd-rich corn roots (31 ~tg C d / g wet wt in corn roots vs 2 ~tg C d / g wet wt in lettuce leaves). As per Rauser's procedure, lettuce leaf homogenates were extracted five times with 100 (first extract) and 10 mM Tris-HC1 (next four extracts), pH 8.6, and the supematants were passed over a small anion exchange column (QSepharose Fast Flow). The elution of the column with 10 mM HEPES, 1M KC1 recovered only 1/3 to 1/4 of Zn or Cd retained in the column, which was 30-40% of Zn or Cd loaded. The remaining leaf Zn and Cd retained in the anion exchanger were eluted only with 0.4M HC1, whereas in the case of corn roots containing Cd-PCs all Cd retained in the column was eluted with 10 mM HEPES, 1M KC1 (Rauser, personal communication). The Zn-Cd binding material in lettuce extracts that was eluted only with 0.4M HC1 was bound stronger to the anion exchanger than Cd-GSH, since this was eluted completely with 10 mM HEPES, 1 M KC1. Cysteine, Glu, and Gly in peptide 5 of the HPLC chromatogram of the Zn-Cd complex from high-Zn extracts were the predominant amino acids and their sum (88%) was within the range (75-90%) observed in case of impure or purified plant PCs (13). However, Cys, Glu, and Gly were not in the proportion characteristic of plant PCs (n:n:l, n = 2-11) (11), but in a ratio of 1:1.7:1.6. Also, the content of Cys (20%) in the same peptide was lower than the average value reported in purified or impure


Vol. 48, 1995

26

McKenna and Chaney

plant PCs (30--40 residue percent) (21). The small percentage of other amino acids in peptide 5 could be contaminants that aggregated with the peptide during isolation. The elution of the gel filtration column at high ionic strength (0.3M KC1) should have helped in separating contaminants (11), although the Cd-Zn complex containing fractions still had a yellowish coloration. Peptide 5 of high-Zn Zn-Cd complex contained identical proportion of Glu and Gly which could also indicate the presence of glutathione, if Cys content was identical to Glu and Gly. Reese et al. (22) reported that Zn was ineffective in promoting 7-glutamyl peptide formation in Schizosaccharomyces pombe, and that the major Zn binding component appeared to be glutathione because it contained equimolar ratio of Glx and Gly. However, the Cys content of the Zn-compound was not revealed by those authors. The amino acid composition of the same peptide 5 resembles the Cu-binding complex isolated from roots of Mimulus gluttatus by Robinson and Thurmann (23) with a ratio of Cys:Glu:Gly of 1:1.7: 1.3. This Cu-complex was later identified as a Cu-PC by purification to homogeneity (24). The amino acid composition of the peptides isolated from both the low and high-Zn leaf extracts did not match that of either animal or fungal metallothioneins (MTs), characterized by a cysteine content of nearly 30 mol% (25). The apparent lack of Zn-Cd-PCs or Zn-Cd-Mts in our lettuce leaves corroborates the findings of Palma et al. (26) in leaves of peas exposed to 240 ~ Cu for 14 d. They found that the major Cu-binding protein in Cu-tolerant peas contained only 1% Cys, and the predominant amino acids were Gly (16%), Ser (16%), Glx (14%), and Asp (10%). Similarly, Tukendorf and Baszynski (27) isolated from oat roots exposed to excess of Cu a thiol-poor Cu-binding protein with only 3% Cys and containing predominantly Asp (21%), Glu (19%), and Gly (11%). These authors also reported that Cu-proteins from Cu-treated and control oat roots (185 vs 17.2 ~tg C u / g dry wt) did not differ in the content of Cys, but Glu was higher in the first ones (about a two-fold increase). Also, in an earlier study with Cu-tolerant spinach, Tukendorf et al. (28) did not find evidence of an MT-like protein being produced in response to Cu. The absence of metal-PCs was also reported by Taylor et al. (29), who isolated from citrus phloem a Zn-binding peptide rich in Asp, with only 4% Cys, and containing 1 Cys:l Glu:2 Gly. The role of plant PCs under chronic environmental metal exposure is not well defined (30) and most PC reports have been done under high or very high levels of Cd exposure: 20-1000 ~ Cd (14,16). Furthermore, Zn was reported to be a weaker PC inducer than Cd and substantial metal loss can occur during the purification of Zn-PCs. In Zn-tolerant Silene grown on a zinc-rich mine tailing or nutrient solution, only a few percent (0.08-1.3%) of root Zn was complexed by PCs (15,31), and 1000 ~M Zn induced only 41% of the PCs induced by 100 t~M Cd in R. serBiological Trace Element Research

Vol. 48, 1995


27

pentina (11). Also, no evidence was obtained by Reese and Wagner (32) for Zn induction in vivo of, or Zn binding in vitro, to tobacco Cd-peptide for Zn concentrations as high as 500 ~M. By contrast, animal MTs are induced by Zn or Cd (33,34). It is therefore not surprising that we did not find Cd-Zn-PCs in our leaf extracts due to the relatively lower levels of metal used and the low PC induction capacity of Zn. In general, information is scanty on the metal concentrations in plant tissues that correspond to the presence of PCs and the methodology used in controlling metal chemical activity in plant growth medium. In unbuffered nutrient solutions, root uptake of Cd or Zn could be rapid, which might affect not only metal distribution between plant roots and shoots but also accelerate the synthesis of PCs as a response stress mechanism. Our plant growth conditions in Zn--Cd buffered nutrient solutions mimicked closely the soil buffering capacity and translated to more realistic concentrations of Zn and Cd in plant edible tissues. Although we were not able to identify the Zn-Cd binding peptides isolated from the low- or high-Zn leaf extracts, we can say that sequestration of Zn and Cd in lettuce leaves, at levels caused by environmental pollution, are not explained by PCs. Besides metal chelation through thiol groups, as in cysteine, there are other amino acids (e.g., His, Glu, Ala, Ser) able to chelate Zn or Cd in thiol-poor proteins. In fact, Cd and Zn have been found to be bound in low-cysteine proteins/peptides not only in plants (29), but also in different animal species (35-40). Possibly, the biochemical processes of metal sequestration in plants, as found in animals (40,41), might be different in acute toxicity and chronic environmental exposure conditions. In conclusion, since we did not find Zn- or Cd-PCs in either the low- or high-Zn leaf extracts, this mechanism cannot explain our previous findings (8) of lower bioavailability of plant Cd from Zn-Cd contaminated crops relative to Cd-only polluted plants and the stronger effect of plant Zn than inorganic Zn in lowering intestinal absorption of Cd. Nevertheless, our present finding of higher cysteine content in the Cd-Zn complex from high-Zn than low-Zn leaf extracts could in part explain our prior results (8). Metallothionein synthesis might have been enhanced in the duodenum of animals fed high-Zn plant diets compared to animals fed low-Zn plant diets or even high-Zn control diets with identical Cd levels, thus increasing Cd sequestration in that tissue. In addition, the amino acid composition differences between the Cd-Zn complex from high-Zn and low-Zn leaf extracts here reported, such as higher content of Glu, Gly, and Asp in the first case could have contributed to differences in Cd or Zn speciation in the animal gut and lowered Cd absorption in animals fed high-Zn crop diets as previously reported (8). In view of the results of the present study and the difficulty in accounting for the chemical speciation of total Zn and Cd in edible plant tissues, future research should focus on the chemical speciation of those metals in the intestine of animals fed Zn-Cd contaminated crops or Biological Trace Element Research

Vol. 48, 1995

28

McKenna and Chaney

Cd-only polluted crops, in order to understand the importance of plant Zn in lowering the intestinal absorption of plant Cd.

ACKNOWLEDGMENTS We thank Wilfried E. Rauser, Department of Botany, University of Guelph, Ontario, Canada, for his valuable advice on the preparation and purification of leaf extracts. We are also grateful to John Lydon and Parthasarathy Pillai, Weed Science Laboratory, USDA-Agricultural Research Service, Beltsville, MD, for their suggestions and assistance in the HPLC analyses.

REFERENCES 1. D. A. Cataldo, K. M. McFadden, T. R. Garland, and R. E. Wildung, Plant Physiol. 86, 734-739 (1988). 2. P. R. Flanagan, J. S. McLellan, J. Haist, M. G. Cherian, M. J. Chamberlain, and L. S. Valberg, Gastroenteroly 74, 841-846 (1978). 3. E. C. Foulkes, Critical Reviews in Toxicology 20, 327-339 (1990). 4. M. R. S. Fox, J. Environ. Qual. 17, 175-180 (1988). 5. C. D. Klaassen, L. D. Lehman, and T. Maitani, in Trace Substances in Environmental Health, D. D. Hemphill, ed., University of Missouri, Columbia, MO, pp. 113-120 (1986). 6. I. M. McKenna and R. U Chaney, in Proc. of the Fourth Int. Conference on The Combined Effects of Environmental Factors, Fechter, L. D., ed., Johns Hopkins University, Baltimore, MD, pp. 65-70 (1991). 7. I. M. McKenna, R. L. Chaney, and F. M. Williams, Environ. Pollution 79, 113-120 (1993). 8. I. M. McKenna, R. L. Chaney, S.-H. Tao, R. M. Jr. Leach, and F. M. Williams, Environ. Res. 57, 73-87 (1992). 9. E. Grill, E. L. Winnacker, and M. H. Zenk, Science 230, 674-676 (1985). 10. E. Grill, E. L. Winnacker, M. H. Zenk, FEBS Lett. 205, 115 (1986). 11. E. Grill, E. L. Winnacker, and M. H. Zenk, Proc. Natl. Acad. Sci. USA 84, 439-443 (1987). 12. W. E. Rauser, Annu. Rev. Biochem. 59, 61-86 (1990). 13. A. B. Tomsett, A. K. Sewell, S. J. Jones, J. R. Miranda, and D. A. Thurman, in Inducible Plant Proteins: Their Biochemistry and Molecular Biology, J. L. Wray, ed., Cambridge University Press, pp. 1-24 (1992). 14. G. J. Wagner and R. M. Krotz, in Metal Ion Homeostasis: Molecular Biology and Chemistry, D. H. Hamer and D. R. Winge, eds., A.R. Liss, New York, pp. 325-336 (1989). 15. E. Grill, E. L. Winnacker, and M. H. Zenk, Experientia 44, 539, 540 (1988). 16. A. B. Tomsett and D. A. Thurman, Plant, Cell and Environment 11, 383-394 (1988). 17. D. R. Parker, R. L. Chaney, and W. A. Norvell, in Chemical Equilibrium and Reaction Models, R. Loeppert, A. P Schwab, and S. Goldberg, eds., Soil Science Society of America Special Publ. No. 42. American Society of Agronomy, Madison, WI, (1995). 18. C. J. Asher and M. E. Blamey, J. Plant Nutr. 10, 1371-1380 (1987). Biological Trace Element Research

Vol. 48, 1995

C a d m i u m - Z i n c Complexes in Lettuce Leaves 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35. 36. 37.

38. 39. 40. 41.

29

H. Willner, M. Vasak, and J. H. R. Kagi, Biochemistry 26, 6287-6292 (1987). B. Lane, R. Kajioka, and T. Kennedy, Biochem. Cell Biol. 65, 1001-1005 (1987). R. N. Reese and G. J. Wagner, Biochem. J. 241, 641-647 (1987a). R. N. Reese, R. K. Mehra, E. B. Tarbet, and D. R. Winge, J. Biol. Chem. 263, 4186-4192 (1988). N. J. Robinson and D. A. Thurman, Planta 169, 192-197 (1986). D. E. Salt, D. A. Thurman, A. B. Tomsett, and A. K. Sewell, Proc. R. Soc. Lond. B 236, 79-89 (1989). R. K. Mehra, E. B. Tarbet, W. R. Gray, and D. R. Winge, Proc. Natl. Acad. Sci. 85, 8815-8819 (1988). J. M. Palma, J. Yanes, M. Gomez, and L. A. Rio, Planta 181, 487-495 (1990). A. Tukendorf and T. Baszynski, J. Plant Physiol. 120, 57-63 (1985). A. Tukendorf, S. Lyszcz, and T. Baszynski, J. Plant Physiol. 115, 351-360 (1984). K. C. Taylor, L. G. Albrigo, and C. D. Chase, in Metal ion Homeostasis: Molecular Biology and Chemistry, D. H. Hamer and D. R. Winge, eds., A. R. Liss, New York, pp. 385-394 (1989). M. J. Melancon, R. Alscher, W. Benson, G. Kruzynski, R. R Lee, H. C. Sikka, and R. B. Spies, in Biomarkers: Biochemical, Physiological, and Histological Markers of Anthropogenic Stress, R. J. Hugget, R. A. Kimerle, P. M. Mehrle, Jr., and H. L. Bergman, eds., SETAC Special Publication Series, Lewis Publishers, pp. 87-123 (1992). H. Harmens, P. R. Den Hartog, W. M. Ten Bookum, and A. C. Verkleij, Plant Physiol. 103, 1305-1309 (1993). R. N. Reese and G. J. Wagner, Plant Physiol. 84, 574-577 (1987b). M. G. Cherian and R. A. Goyer, Life Sci. 23 (1), 1-10 (1978). K. T. Suzuki, Y. Tanaka, and R. Kawamura, Comp. Biochem. Physiol. 75C (1), 33-37 (1983). M. P. Waalkes, S. B. Chernboff, and C. D. Klaassen, Biochem. J. 220, 811-818 (1984). A. Nejmeddine, N. Dhainaut-Courtois, J. L. Baert, P. Sautiere, B. Fournet, and R Boulenguer, Comp. Biochem. Phisiol. 89C, 321-326 (1988). R D. Whanger, J. T. Deagen, and M. A. Beilstein, in Metallothionein II, J. H. R. Kagi and Y. Kojima, eds., Proc. of the Second Int. Meeting on Metallothionein and Other Low Molecular Weight Metal-Binding Proteins. Birkhauser Verlag Publ., Basel-Boston, pp. 281-287 (1987). J. W. Ridlington and B. A. Fowler, Chem. Biol. Int. 25, 127-138 (1979). C. L. Post, K. S. Squibb, B. A. Fowler, D. E. Gardner, J. Illing, and G. E. R. Hook, Biochem. Pharmacol. 31, 2969-2975 (1982). D. G. Thomas, J. F. L. G. Solbe, J. Kay, and A. Cryer, Biochem. Biophys. Res. Commun. 110 (2), 584-592 (1983). R. P. Cosson, C. Amiard-Triquet, and J.-C. Amiard, Water, Air, Soil Pollut. 57,58, 555-567 (1991).


Vol. 48, 1995