Chemical profiling of gramineous exudates using high -resolution … › em52 › eprints ›...

1

1

2

Chemical profiling of gramineous plant root exudates using high-resolution NMR and MS Teresa W-M. Fan a *, Andrew N. Lane b , Moshe Shenker c , John P. Bartley d , David Crowley e , and Richard M. Higashi f

a * Department of Land, Air and Water Resources, f Crocker Nuclear Lab., University of California, Davis, CA 95616, USA

b National Institute for Medical Research, Mill Hill, London NW71AA, UK c Department of Soil and Water Sciences, The Hebrew University of Jerusalem, Rehovot 76-100, Israel d Queensland University of Technology, GPO Box 2434, Brisbane 4001, Queensland, Australia. e Dept. of Soil and Env. Sciences, University of California, Riverside, CA 92521-0424, USA Metal ion ligands in root exudates of different plant types were examined using multinuclear 2-D NMR and high-resolution MS. Uptake of nutrient transition metals was related to the mugineic acid phytosiderophores while uptake of pollutant Cd appeared to be mediated by a different mechanism.

Root Exudate

Fe, Cu, Mn, Zn uptake

Cd uptake?

N N H OH

O O O OH OH OH e.g.

2

TITLE: COMPREHENSIVE CHEMICAL PROFILING OF GRAMINEOUS PLANT ROOT 2

EXUDATES USING HIGH-RESOLUTION NMR AND MS 3

4

Teresa W-M. Fana*, Andrew N. Laneb, Moshe Shenkerc, John P. Bartleyd, David Crowleye, and 5

Richard M. Higashif. 6

7 a* Department of Land, Air and Water Resources, University of California, One Shields Ave., Davis, 8

CA 95616, USA. Tel: 530-752-1450; Fax: 530-752-1552; E-mail: [email protected] 9 b Division of Molecular Structure, National Institute for Medical Research, Mill Hill, London 10

NW71AA, UK 11 c Department of Soil and Water Sciences, Faculty of Agricultural, Food and Environmental Quality 12

Sciences, The Hebrew University of Jerusalem, Rehovot 76-100 Israel. 13 d School of Physical Sciences, Queensland University of Technology, GPO Box 2434, Brisbane 4001, 14

Queensland, Australia. 15 e Dept. of Soil and Env. Sciences, University of California, Riverside, CA 92521-0424, USA. 16 f Crocker Nuclear Laboratory, University of California, One Shields Ave., Davis, CA 95616, 17

USA. 18

19

ABSTRACT 20

Root exudates released into soil have important functions in mobilizing metal micronutrients and for 21

causing selective enrichment of plant beneficial soil microorganisms that colonize the rhizosphere. 22

Analysis of plant root exudates typically has involved chromatographic methods that rely on a priori 23

knowledge of which compounds might be present. In the research reported here, the combination of 24

multinuclear and 2-D NMR with GC-MS and high-resolution MS provided de novo identification of a 25

number of components directly in crude root exudates of different plant types. This approach was 26

applied to examine the role of exudate metal ion ligands (MIL) in the acquisition of Cd and transition 27

metals by barley and wheat. The exudation of mugineic acids and malate was enhanced by Fe 28

deficiency, which in turn led to an increase in the tissue content of Cu, Mn, and Zn. The presence of 29

elevated Cd maintained at a free activity pCd of 8.8 (10-8.8 M), resulted in reduced phytosiderophore 30

production by Fe deficient plants. The buffer morpholinoethane sulfonate (MES), which is commonly 31

used in chelator-buffering nutrient solutions, was detected in the root exudate mixture, suggesting uptake 32

and re-secretion of this compound by the roots. The ability to detect this compound in complex mixtures 33

3

containing organic acids, amino acids, and other substances suggests that the analytical methods used 34

here provide an unbiased method for simultaneous detection of all major components contained in root 35

exudates. 36

37

KEYWORDS 38

Hordeum vulgare, Triticum aestivum, multinuclear and 2-D NMR, high-resolution MS, Fe 39

deficiency, metal sequestration, metal ion ligands, cadmium, phytosiderophores, mugineic acids. 40

41

INTRODUCTION 42

Plant root exudates can constitute up to 30-40 % of the plant’s photosynthetic productivity 43

Whipps, 1990; Sauerbeck et al., 1981; Oades, 1978; Samtsevich, 1965). They are known to play 44

important roles in a number of rhizospheric processes, including nutrient (e.g. Fe and Zn) acquisition 45

(Hopkins et al., 1998, Romheld, 1991; Zhang et al., 1991), plant-microbe associations (e.g. Rhizobium 46

symbiosis with leguminous roots) (Janczarek et al., 1997; Scheidemann and Wetzel, 1997), regulation 47

of plant growth (Einhellig et al., 1993), and determination of microbial community structure in the plant 48

rhizosphere (Grayston et al., 1995; Yang and Crowley, 2000). It is also expected that root exudate 49

components interact with both organic and inorganic contaminants to regulate their bioavailability and 50

transport in the soil environment. Moreover, root exudation may be an important mediator in altering 51

the species composition of the rhizosphere microflora that function in nutrient transformations, 52

decomposition and mineralization of organic substances, and formation of soil organic matter, all of 53

which are related to soil quality (Hodge and Millard, 1998). This latter aspect is particularly in need of 54

clarification, in view of the ever-increasing threat of global environmental change and soil pollution 55

caused by anthropogenic activity. However, the relationship between changes in exudate composition 56

and microbial community structure and function in soils is poorly understood. The lack of 57

comprehensive knowledge of exudate chemistry has been a major barrier to such mechanistic 58

understanding. 59

The slow progress in exudate characterization can be largely attributed to the lack of analytical 60

approaches that are comprehensive, require neither sample fractionation nor a priori knowledge, and 61

that are capable of rigorous structure identification. Using a combination of multi-nuclear and 2-D 62

NMR with GC-MS, we have previously demonstrated a comprehensive determination of barley root 63

exudates (Fan et al., 1997b) based on known spectroscopic databases (Fan, 1996) and literature 64

information on phytosiderophores (Ma and Nomoto, 1993). 65

4

However, for totally unknown compounds or those not represented in the databases, the above 66

approach is not always sufficient to deduce the complete structure(s). To supply the necessary 67

information, we have now added high-resolution mass spectrometry to the previous methods to obtain 68

molecular formula information. This greatly facilitated the determination of the complete structures of 69

various exudate components not in the databases. We also demonstrated that root exudate profiles 70

were highly dependent on plant species and genotypes of the same species as well as growth conditions. 71

Moreover, we observed a significant attenuating effect of Cd treatment on root exudate profiles and yet 72

an enhancement of transition metal uptake for both barley and wheat plants. 73

74

RESULTS AND DISCUSSION 75

76

Exudate Analysis 77

Roots exudates of wheat, barely, and rice plants were analyzed by a combination of NMR and 78

GC-MS methods as described previously (Fan et al., 1997b). Using this approach, comprehensive 79

profiles of the components were obtained directly from crude exudates without sample fractionation. 80

Illustrated in Figure 1 are results of 2-D 1H total correlation spectroscopy (TOCSY), 1H-13C 81

heteronuclear single quantum coherence spectroscopy (HSQC), and 1H-13C heteronuclear multiple 82

bond correlation spectroscopy (HMBC) for the root exudate of the wheat amphiploid (AgCS) under 83

the low Fe-high Cd (↓Fe - ↑Cd) treatment. From 1H TOCSY analysis, covalent correlations between 84

pairs of protons that are up to 4 bonds apart were traced, while the HSQC and HMBC experiments 85

revealed carbon coupling to its directly attached protons and to protons 2-4 bonds away, respectively. 86

Two dominant components of the AgCS exudate were identified from these spectra, i.e. 2’-87

deoxymugineic acid (2’-DMA) and MES (the buffer included in the plant growth medium but not during 88

exudate collection). The phytosiderophore, 2’-DMA, has been reported in common wheat (Triticum 89

aestivum L.) root exudates (Murakami et al., 1989), but its exudation by the AgCS hybrid has not 90

been characterized previously. As for MES, it is thought to be an impermeant buffer (but see below) 91

and is not expected to be present in root exudates[DC1]. 92

2’-DMA was assigned based on the 1H and 13C chemical shifts, proton covalent (scalar) 93

connectivity acquired from the TOCSY spectrum (Fig. 1A), the proton to carbon covalent connectivity 94

obtained from the HSQC (Fig. 1B) and HMBC spectra (Fig. 1C), as well as molecular ion data 95

acquired from electrospray MS (ES-MS). All proton connectivities expected for 2’-DMA (H-2 to H-96

3 to H-4, H-1’ to H-2’ to H-3’, and H-1” to H-2” to H-3”) were present in Fig. 1A, so were the one-97

5

bond carbon to proton correlations in Fig. 1B. In addition, scalar correlations were evident in the 98

HMBC spectrum (Fig. 1C) for the three carboxylate carbons to their closest protons (C-1 to H-2, C-99

4’ to H-3’, and C-4” to H-3”) and between protonated carbons and their neighboring protons that are 100

2 (e.g. C-2” to H-3”), 3 (e.g. C-1” to H-3”), and 4 bonds (e.g. C-1’ to H-4) apart. Moreover, two 101

mass ions (m/z 304 and 326) were evident in the full scan ES-MS spectrum (Fig. 2) of the same 102

exudate, which corresponded to the molecular ion (M) and M+Na of 2’-DMA. Taken together, these 103

NMR and MS data provided unequivocal identification of 2’-DMA in crude AgCS root exudates. 104

The presence of MES was deduced from the NMR (Fig. 1A-C), ES-MS (Fig. 2), and high-105

resolution fast atom bombardment (FAB) mass spectral data (data not shown). As shown in Fig. 2, the 106

ES-MS spectrum of the crude AgCS exudate was abundant in the mass ions of 196, 218, and 413, in 107

addition to 304 and 326. The m/z 218 and 413 ions were related to the m/z 196 ion based on their 108

daughter ion patterns (data not shown), and could correspond to the Na adduct of the 196 monomer 109

and dimer, respectively. From the FAB data of m/z 196, a molecular formula of C6H14O4NS was 110

inferred with a mass error of 5.8 ppm. This information greatly assisted NMR in deducing the MES 111

structure as described below. 112

The 1H TOCSY data (Fig. 1A) revealed two sets of –CH2–CH2– groups with one set (A) having 113

nonequivalent and the other (B) equivalent geminal protons. The integrated area of set A resonances 114

was roughly twice that of set B resonances. In accordance with the molecular formula, set A protons 115

should have originated from two chemically equivalent –CH2–CH2– groups while set B protons were 116

derived from one –CH2–CH2– group. These three –CH2–CH2– groups accounted for all six of the 117

carbons allowed in the molecular formula. In addition, the chemical shifts of both set A and set B 118

protons indicated that they were neighboring to heteroatoms such as N, O, or S. No other carbons 119

including those from the carboxylates were present in the structure based on the analysis of 1-D 13C 120

NMR and 2-D HMBC (Fig. 1C and data not shown), which is consistent with the molecular formula. 121

Moreover, no exchangeable (primary or secondary) amines were present based on the 1H TOCSY 122

spectrum recorded in H2O instead of D2O (data not shown), suggesting that the N is in a tertiary amine. 123

A quarternary amine was ruled out based on the required number (14) of protons in the molecular 124

formula. Altogether, the above structural clues allowed the MES structure to be deduced ab initio. 125

Further proof for this assignment was obtained from the 1H NMR analysis of the MES standard (data 126

not shown), which matched exactly with that acquired from the AgCS exudate. Thus, using the 127

combined NMR and MS approach, molecular structures of exudate components were identified from 128

crude exudates without the need for a priori knowledge or any sample fractionation. 129

6

Using a similar approach, the exudate profile for the Chinese spring wheat (CS, a parent of AgCS) 130

was examined. An example analysis of the CS exudates by TOCSY is shown in Figure 3. Assigned 131

components included Ile, Leu, Val, lactate (lac), Ala, acetate, 2’-DMA (peaks 1-9), malate, citrate, 132

glycinebetaine (GB), Gly, α- and β-glucose (Glc), and α-trehalose (see (Fan, 1996) for NMR 133

parameters for these compounds). Except for the sugars, these assignments were verified by the 1H-134 13C HSQC analysis (data not shown) and in part by the GC-MS analysis of the same exudate (see 135

below). In addition to components yielding sharp resonances, there were those whose cross-peaks 136

were relatively broad (indicated by arrows). The identities of these components are currently unknown, 137

although they are likely to be of macromolecular origin (e.g. proteins), as indicated by their chemical 138

shifts and broad peak widths. A polypeptide that appears to be related to Al resistance has been 139

reported in wheat root exudates (Basu et al., 1999) 140

141

Comparison of Root Exudate Profiles from Different Plant Species, Genotypes, and Growth 142

Conditions 143

Altogether, the above NMR spectroscopic analysis provided comprehensive profiles of exudate 144

components (both known and unknown to our database) excreted by barley, wheat, and rice plants, as 145

illustrated in Figure 4. By comparing these 1H NMR spectra, it is clear that exudate profiles varied with 146

plant species and genotypes as well as treatment conditions. For example, the barley exudate differed 147

from wheat exudates (CS and AgCS), most notably in the dominant mugineic acids present (3-epi-148

OHMA+MA for barley and 2’-DMA for wheat). A significant difference in the relative quantity of GB, 149

Ala, γ-aminobutyrate (GAB), Gly, Glu, and acetate was also noted, although the difference in acetate 150

may be complicated by a differential loss through volatilization during the freeze-drying process. The 151

rice exudate profile was different again from those of barley and wheat, in that except for Gly, the amino 152

acids and mugineic acids were present only at very low concentration. In addition, an unknown polyol 153

component was abundant in the rice exudate but absent in barley or wheat exudates. 154

Except for the exudation of 2’-DMA, the two wheat genotypes (CS vs. AgCS) exhibited rather 155

different exudate profiles (Fig. 4). The AgCS exudate was dominated by the medium buffer MES and 156

contained relatively little amino and organic acids, while the CS exudate was abundant in both amino 157

and organic acids. In addition, unknown component(s) represented by a group of resonances from 1.1-158

1.6 ppm in the AgCS exudate were absent in the CS exudate. The difference in the MES content was 159

presumably a result of the different treatment conditions, with MES included in the AgCS but absent in 160

CS growth media. However, it is unlikely that the abundance of MES in the AgCS exudate was purely 161

7

a result of contamination from the growth medium. This is because extensive rinsing of roots was 162

performed before exudate collection. Moreover, an identical MES treatment was used to culture barley, 163

yet MES was much less abundant in the barley exudate (Fig. 4). Instead, it is plausible that MES was 164

taken up by AgCS roots and then excreted along with other exudate components. The possible effects 165

of MES on the physiology of plants grown in buffered hydroponic media are unknown, but this finding 166

suggests that this question should be studied in more detail for experimental protocols that use MES and 167

other buffers to maintain the pH of plant growth media. 168

The same exudates from Fig. 4 were also analyzed by GC-MS and the results are shown in Table 169

1, along with those from the NMR spectroscopic analysis. The GC-MS analysis not only confirmed 170

many of the NMR assigned components described above, but also provided absolute quantification of 171

these components plus those that were below NMR detection limits. At the same time, the NMR 172

analysis (with the aid of GC-MS analysis) provided absolute quantification for components (e.g. 173

mugineic acids) that were not readily analyzed by GC-MS. Thus, the two approaches complemented 174

each other in providing richer information than that achievable by each method alone. 175

In terms of total amounts of identified exudate components on a per gram root weight basis, the 176

wheat hybrid AgCS ranked the highest (although a major portion was attributed to MES), followed by 177

barley, wheat CS, and rice. It is also notable that many quantitative differences in the exudate 178

components of the four plant types were evident from the GC-MS profiles, which is consistent with the 179

NMR characterization. Moreover, the abundance of the mugineic acid phytosiderophores in barley and 180

wheat exudates was noted. In contrast, the exudation of these phytosiderophores by Fe-deficient rice 181

roots was not apparent, which may be related to the NMR detection limit plus a much lower capacity of 182

rice roots in excreting these compounds (Kawai et al., 1988). This lower capacity for mugineic acid 183

exudation by rice roots has been correlated with their reduced ability to tolerate Fe deficiency (Mori et 184

al., 1991; Kawai et al., 1988; Marschner et al., 1986). 185

186

Effect of Combined Cd and Fe Deficiency Treatment on Root Exudate Profiles of Barley and 187

Wheat Plants 188

Utilizing the above approach, we examined the chemical profiles of both barley and AgCS wheat 189

exudates collected under a combination of Cd and Fe deficiency treatments. The effects of these 190

treatments on plant growth, metal uptake, total Fe-mobilizing capacity, and other physiological 191

parameters have been reported elsewhere (Shenker et al., submitted to Journal of Environmental 192

Quality, JEQ). Here, we compared the effect of these treatments on root metal content with that on 193

8

exudate composition, as shown in Figures 5 and 6, for barley and wheat, respectively. Although many 194

more exudate components have been analyzed (see Figure 4), the comparison is illustrated for abundant 195

metal ion ligands (MIL), since they are more likely to influence metal mobilization and uptake by plant 196

roots. The stability constants of these MIL with different metal ions vary considerably, and their actual 197

ability to mobilize various metal ions will depend on the pH and ionic composition of the soil solution, as 198

well as kinetic factors involved in the dissolution of soil minerals. 199

Under Fe deficient conditions (↓Fe - ↑Cd and ↓Fe - ↓Cd), the increased exudation of the 3-epi-200

OHMA, MA, 2’-DMA, and malate in barley (Figure 5 and Table 2) and 2’-DMA in wheat exudates 201

(Figure 6) was evident, as compared to Fe sufficient conditions (↑ Fe - ↑Cd and ↑Fe - ↓Cd). This 202

increase was in turn related to the elevated production of Fe mobilizing substances (FeMS) by barley 203

(data not shown) and wheat (Figure 6) under Fe deficiency. These results are consistent with the role of 204

these MIL in Fe acquisition. The exudation of other abundant MIL (i.e. lactate, Ala, GAB, and GB for 205

barley; MES for wheat) did not exhibit a clear response to Fe deficiency (Figure 5, Fig. 6, and Table 206

2), which suggests that they are not specifically involved in Fe acquisition. However, these MIL may 207

still function generally in chelating Fe for uptake.[DC2] 208

Accompanying the increased exudation of the phytosiderophores was the elevated content of Zn 209

and Cu in barley roots (Figure 5), Cu in wheat roots (Figures 6) as well as Zn, Cu, and Mn in both 210

shoots (Shenker et al., submitted to JEQ). This finding suggests that the MAs [DC3]may be important in 211

the acquisition of Zn, Cu, and Mn, in addition to Fe. It is also noteworthy that the exudation pattern of 212

malate was similar to that of the MAs by barley roots (Fig. 5). This result could be accounted by the 213

exudation of malate and MAs via a common anion channel. However, other abundant anions in root 214

tissues including citrate and malate (unpublished results) did not exhibit such exudation pattern, which 215

argues against a systemic leakage through such a channel. In fact, citrate was not exuded by barley or 216

wheat roots to any significant extent (data not shown). Regardless of the exudation mechanism(s), it is 217

possible that the exuded malate could serve a similar role as MAs in mediating the observed 218

accumulation of transition metals in barley and wheat tissues. 219

As for Cd acquisition, there was no statistically significant correlation of the above MIL to the Cd 220

content of roots (Figures 5 and 6) or shoots (data not shown) of barley and wheat. This result suggests 221

that, Cd acquisition may involve a mechanism different from that of transition metal acquisition. It is also 222

interesting to note that there appeared to be a weak correlation between MES concentration in wheat 223

exudates and root Cd content (Figure 6). It is possible that the presence of MES may interact with Cd 224

to affect the latter uptake into wheat. Cd accumulation could weaken membrane integrity or increase 225

9

membrane permeability via lipid peroxidation (Goyer, 1997; Stohs and Bagchi, 1995), which allows 226

Cd-bound MES to enter into root cells and later to be released to the collection solution down the 227

concentration gradient. This hypothesis is consistent with the higher Cd uptake and MES release by 228

AgCS wheat roots than by barley roots (cf. Figures 5 and 6; Table 1). For barley, the exudation of 229

Ser, Thr, and Glu appeared to be related to the high Cd treatment under Fe sufficient conditions (Figure 230

5 and Table 2). It is possible that these MIL may be involved in Cd acquisition[DC4]. 231

In conclusion, the combination of multinuclear and 2-D NMR with GC-MS and high-resolution 232

MS provided de novo identification of a number of components directly in crude root exudates of 233

different plant types. This approach was applied to examine the role of exudate MIL in the acquisition 234

of Cd and transition metals by barley and wheat. The results suggest that enhanced exudation of 235

mugineic acids and malate may be involved in transition metal but not Cd acquisition. Therefore, the 236

acquisition mechanism appears to differ between the essential and toxic pollutant metals. 237

238

EXPERIMENTAL 239

240

Plant Growth 241

The plant varieties used were Barley (Hordeum vulgare L., cv. CM-72), wheat (Triticum 242

aestivum L., cv. AgCS, a hybrid of Chinese spring wheat and Lophopyrum. sp) supplied by Dr. J. 243

Dvorak (Rommel and Jenkins, 1959), Chinese spring wheat (CS, a parent of AgCS), and rice (Oryza 244

sativa, L., cv M201). Barley and AgCS wheat seedlings (eight per treatment) were grown for 22 d in 245

nutrient solutions containing CdCl2 buffered by an excess of EDTA and HEDTA (Fan et al., 1997b; 246

Shenker et al., submitted JEQ) to produce Fe-sufficient (pFe = 17.2) and deficient (pFe = 18.2-18.4) 247

conditions in combination with free Cd activity (pCd) of 8.8 or 10.0. Other free metal activities were 248

pZn = 10.1, pCu =13.5; and pMn = 7.3 (for complete plant growth protocol refer to Shenker et al., 249

submitted JEQ). The four treatments are designated as: 1) low Fe/high Cd (↓Fe - ↑Cd); 2) high 250

Fe/high Cd (↑Fe - ↑Cd); 3) low Fe/low Cd (↓Fe - ↓Cd); and 4) high Fe/low Cd (↑Fe - ↓Cd). 251

Morpholinoethane sulfonate (MES) at 2 mM was also included to buffer the pH of the nutrient solution 252

at 6.2±0.2. Chinese spring (CS) wheat and rice (cv. M201) seedlings (four per treatment) were grown 253

similarly for a month except that the nutrient solution employed was half-strength modified Hoagland 254

solution (Epstein, 1972) containing full-strength (50 µM) Fe(III)-EDTA, 25 µM EDTA, and 10 µM 255

CdSO4 in the absence of MES (Fan et al., 2000). The calculated pMs in this nutrient solution were 256

pFe = 17.6, pMn = 7.3, pCu = 12.8, pZn = 9.9, pNi = 12.1, and pCd = 8.9 (Shenker et al., submitted 257

10

to JEQ). The pH of the growth media was maintained at 6.0±0.3 by daily titration with 0.1 M HCl or 258

KOH while the nutrient solution was renewed frequently to avoid nutrient depletion. 259

260

Root Exudate Collection 261

Roots were soaked in 1 mM Ca(NO3)2 for 10-20 min and rinsed extensively with doubly 262

deionized (DD) H2O to remove organic and inorganic constituents of the nutrient solution, including 263

metabolites, exudates, chelates, mineral elements, and buffer ions before exudate collection. Root 264

exudates were collected 2-3 times each in 150-200 ml DDH2O during peak diurnal release of 265

phytosiderophores (2-4 h after light onset) of the last 3-6 growth days. The collected solutions were 266

immediately filtered through glass microfiber filters (GF/C, Whatman, Maidstone, England) to exclude 267

root fragments and much of the microbial population, pooled, and lyophilized for storage. Prior to 268

NMR studies, the exudates were resuspended in sterile water, filter sterilized, and passed through 269

Chelex 100 resin (BioRad, Hercules, CA) to remove paramagnetic ions, as described previously (Fan 270

et al., 1997b). 271

272

NMR Analysis 273

The Chelex 100-treated exudates of barley and AgCS were acidified to pH 2, lyophilized, and 274

redissolved in D2O, whereas the CS and rice exudates were not acidified (pH values ranging from 7.0-275

7.8). NMR spectra were recorded at 25 °C on a Varian (Palo Alto, CA) 11.75 Tesla UnityPlus (all 1H 276

experiments) and a Bruker AM-400 spectrometer (1-D 13C analysis). The following 2-D 1H 277

experiments were conducted: 1H total correlation (TOCSY), double-quantum filtered correlation 278

(DQF-COSY), 1H-13C heteronuclear single quantum coherence (HSQC), and 1H-13C heteronuclear 279

multiple bond correlation (HMBC). Selected samples were dissolved in DD H2O containing 10% 280

D2O+0.9M HCl for Watergate TOCSY measurement (Piotto et al., 1992) at 5 °C which optimized the 281

detection of exchangeable protons (Fan et al., 1997a). Detailed acquisition parameters are described in 282

respective figure legends. The 1-D NMR spectra were apodized using an exponential function yielding 283

0.5-2 Hz line broadening, and zerofilled to 64 or 128 K points before Fourier transformation. The 2-D 284

NMR data were zerofilled or linear-predicted to 8 K x 2 K points and apodized with a mild Gaussian 285

function before 2-D Fourier transformation. 286

287

11

Mass spectrometry 288

Nanospray mass spectra were acquired on an LCQ “classic” quadrupole ion trap mass 289

spectrometer (ThermoQuest, Austin, TX) equipped with a nanospray source (Protana, Odense, 290

Denmark) operated at a spray voltage of 800V and a capillary temperature of 150°C. Lyophilized root 291

exudates were dissolved in 50% methanol, 1% formic acid and 2ul transferred to an Econo12 292

nanospray needle (New Objective Inc, Cambridge, MA). MS2 and MS3 analyses were performed at a 293

collision energy of 30% and an isolation width of 4 Da. 294

Accurate mass of exudate components were measured on a Finnigan 2001 Newstar FT/MS 295

system (ThermoQuest, Madison WI) at 3 .0 Tesla. The spectrometer was externally calibrated with 296

PMMA 640. The lyophilized sample was delivered to the mass spectrometer by flow injection at 5 ml / 297

min using a syringe pump (Cole Palmer, Vernon Hills, IL) to an Analytica API external source operating 298

at a capillary voltage of 5.5 kV. 299

300

GC-MS Analysis 301

All exudates were acidified in 0.1 N HCl before an aliquot was lyophilized and derivatized with N-302

methyl-N-[tert-butyldimethylsilyl]trifluoroacetamide (MTBSTFA) before analysis on a Varian 3400 gas 303

chromatograph interfaced to a Finnegan ITD 806 mass spectrometer (Fan et al., 1997b). 304

305

Exudate Quantification 306

All components of exudate collection solutions were quantified by their GC-MS response as 307

compared to that of the standard with known concentration, except for the mugineic acids (MAs), 308

glycinebetaine (GB), acetate, and MES (see Fig. 4.). The latter were quantified from their 1H NMR 309

responses by comparison with the methyl resonance of alanine at 1.47 ppm (either endogenous or 310

added) after corrected for signal saturation and number of protons. The concentration of the 311

endogenous Ala was determined from GC-MS analysis (Fan , 1997b). In addition, the FeMS data 312

were adapted from Shenker et al. (submitted to JEQ), which were acquired using the Cu-CAS (chrome 313

azurol S) colorimetric assay (Shenker et al., 1995). This assay measured the total capacity of Fe 314

mobilization in crude exudates. 315

316

Acknowledgements. 317

This work was supported by U.S. Department of Energy grant numbers DE-FG07-96ER20255, 318

U.S. E.P.A. grant #R825960010, and U.S. E.P.A. funded (#R819658) Center for Ecological Health 319

12

Research at Univ. of California-Davis. We thank Dr. Steven Howell of NIMR for recording the 320

nanospray ES-MS spectra. NMR spectra were recorded at the MRC Biomedical NMR Centre, Mill 321

Hill. 322

323

13

Table 1. GC-MS and NMR quantification of wheat and barley root exudates under low Fe plus Cd 323

treatment. 324

325

µmole/gm root dry wt.

Component Barley1 Wheat CS1 Wheat AgCs2 Rice1

lactate 3.51 1.70 –3 3.23

pyruvate 6.95 ND3 – ND

Ala 2.08 0.13 – 0.01

Gly 0.28 0.07 – 0.52

Val 0.32 0.07 – ND

Leu 0.16 0.07 – ND

Ile 0.08 0.05 – ND

succinate 0.26 0.03 – ND

GAB 1.22 ND – ND

Pro 0.17 0.01 – ND

fumarate 0.23 0.02 – ND

PO4-containing 0.10 0.02 – 1.24

Ser 0.38 0.05 – 0.06

Thr 0.36 ND – ND

malate 0.46 0.49 – 0.38

Phe 0.06 0.01 – ND

Asp 0.22 0.02 – 0.01

Glu 0.25 ND – ND

Asn 0.21 ND – ND

Gln 0.17 ND – ND

3-epi-OHMA 8.19 ND – ND

2'-DMA 0.87 5.64 25.37 ND

MA 2.13 ND ND ND

GB 1.54 0.22 ND ND

MES 2.43 ND 209.18 ND

Acetate ND 11.60 ND ND

14

Total 32.63 20.19 234.55 5.46 1Quantified by GC-MS except for 3-epi-OHMA, 2’-DMA. MA, GB, MES, and acetate, which were 326

analyzed by NMR using existing lactate as an internal standard. 327 2AgCs quantified by NMR with the addition of 1 mM Ala as an internal standard. 328 3ND, not detected; –, not determined. 329

330

15

Table 2. Interactive effect of Cd and Fe deficiency on barley exudate profiles1 330

331

Compound ↓Fe - ↑Cd S.D. ↑ Fe - ↑Cd S.D. ↓Fe - ↓Cd S.D. ↑Fe - ↓Cd S.D. ______________________________________________________________________________________________________________________

lactate 2.88 0.48 2.83 1.07 3.15 1.13 2.69 0.10

pyruvate 7.59 1.74 5.72 0.69 6.06 0.91 5.98 1.70

alanine 1.63 0.45 2.95 0.98 2.20 1.44 1.80 0.62

glycine 0.24 0.07 0.38 0.12 0.27 0.14 0.23 0.08

valine 0.28 0.06 0.30 0.11 0.33 0.14 0.21 0.06

leucine 0.15 0.02 0.17 0.04 0.16 0.05 0.12 0.03

succinate 0.28 0.06 0.21 0.06 0.29 0.13 0.15 0.04

GAB 1.03 0.19 1.35 0.36 1.16 0.61 0.74 0.19

fumarate 0.27 0.08 0.28 0.05 0.27 0.13 0.24 0.02

serine 0.36 0.03 0.49 0.14 0.35 0.16 0.31 0.09

threonine 0.38 0.09 0.53 0.07 0.32 0.15 0.31 0.10

malate 0.34 0.08 0.20 0.08 0.48 0.53 0.09 0.05

aspartate 0.19 0.05 0.27 0.11 0.14 0.05 0.11 0.05

glutamate 0.19 0.05 0.35 0.16 0.22 0.10 0.19 0.08

asparagine 0.17 0.03 0.19 0.06 0.15 0.08 0.13 0.05

glutamine 0.11 0.09 0.14 0.05 0.23 0.16 0.16 0.05

αglycerol-3-P 0.20 0.05 0.16 0.02 0.17 0.07 0.12 0.05

3-epi-OH-

MA

5.25 2.14 1.15 0.77 7.97 4.06 0.96 0.22

2'-DMA 0.49 0.29 0.00 0.01 0.62 0.36 0.01 0.01

MA 0.79 0.92 0.09 0.16 1.02 0.35 0.04 0.07

GB 1.14 0.32 1.56 0.56 1.52 1.00 0.82 0.34

MES 1.28 0.77 1.44 1.04 2.28 1.44 3.24 2.04

Total 25.94 21.45 30.36 19.13 ______________________________________________________________________________________________________________________

1Units in µmole/g root dry wt. 332

333

16

REFERENCES 333

Basu, U., Good, A. G., Aung, T., Slaski, J. J., Basu, A., Briggs, K. G.. and Taylor, G. J. 1999. A 23-334

kDa, root exudate polypeptide co-segregates with aluminum resistance in Triticum aestivum. 335

Physiologia Plantarum 106: 53-61. 336

Einhellig, F. A., Rasmussen, J. A.,, Hejl, A. M., and Souza, I. F. 1993. Effects of Root Exudate 337

Sorgoleone on Photosynthesis. Journal of Chemical Ecology 19: 369-375. 338

Epstein, E. 1972. Mineral Nutrition of Plants: Principles and Perspectives. New York:John Wiley & 339

Sons. 340

Fan, T. W.-M. 1996. Metabolite profiling by one- and two-dimensional NMR analysis of complex 341

mixtures. Progress in Nuclear Magnetic Resonance Spectroscopy 28: 161-219. 342

Fan, T. W. M., Higashi, R. M.,, Frenkiel, T. A., and Lane, A. N. 1997a. Anaerobic nitrate and 343

ammonium metabolism in flood-tolerant rice coleoptiles. Journal of Experimental Botany 48: 1655-344

1666. 345

Fan, T.W.-M., Pedler, J., Lane, A.N., Crowley, D., and Higashi, R.M. 1997b. Comprehensive 346

analysis of organic ligands in whole root exudates using nuclear magnetic resonance and gas 347

chromatography-mass spectrometry. Analytical Biochemistry 251: 57-68. 348

Fan, T.W.-M., Baraud, F., Higashi, R.M., eds. Eller, P.G. and Heineman, W.R. 2000. Genotypic 349

influence on metal ion mobilization and sequestration via metal ion ligand production by wheat. 350

In: Nuclear Site Remediation. American Chemical Society Symposium Series, vol. 778, pp. 351

417-431, Washington, D.C. 352

Grayston, S. J., Campbell, C. D.,, Vaughan, D., and Jones, D. 1995. Influence of root exudate 353

heterogeneity on microbial diversity in the rhizosphere. Journal of Experimental Botany 46: 27. 354

Goyer,R.A. 1997. Toxic and essential metal interactions. Annu. Rev. Nutr. 17: 37–50. 355

Hodge, A. and Millard, P. 1998. Effect of elevated CO-2 on carbon partitioning and exudate release 356

from Plantago lanceolata seedlings. Physiologia Plantarum 103: 280-286. 357

Hopkins, B. G., Whitney, D. A.,, Lamond, R. E., and Jolley, V. D. 1998. Phytosiderophore release 358

by sorghum, wheat, and corn under zinc deficiency. Journal of Plant Nutrition 21: 2623-2637. 359

Janczarek, M., Urbanik-Sypniewska, T., and Skorupska, A.. 1997. Effect of authentic flavonoids and 360

the exudate of clover roots on growth rate and inducing ability of nod genes of Rhizobium 361

leguminosarum bv. trifolii. Microbiological Research 152: 93-98. 362

Kawai, S., Takagi, S., and Sato, Y. 1988. Mugineic acid-family phytosiderophores in root-secretions 363

of barley, corn and sorghum varieties. Journal of Plant Nutrition 11: 633-642. 364

17

Ma, J. F. and Nomoto, K.. 1993. Two Related Biosynthetic Pathways of Mugineic Acid in 365

Gramineous Plants. Plant Physiology (Rockville) 102: 373-378. 366

Marschner, H., V. Römheld and M. Kissel. 1986. Journal of Plant Nutrition 9: 695-713. 367

Mori, S., Nishizawa, N., Hayashi, H., Chino, M., Yoshimura, E., and Ishihara, J. 1991. Why are young 368

rice plants highly susceptible to iron-deficiency? Plant and Soil 130: 143-156. 369

Murakami, T., Ise, K., Hayakawa, M., Kamei, S., and Takagi, S. Chemical Letters 1989, 2137-2140, 370

1989. 371

Oades, J. M. 1978. Mucilages at the root surface. Journal of Soil Science 29: 1-16. 372

Piotto, M., Saudek, V. and Sklenar, V.. 1992. Gradient-Tailored Excitation For Single-Quantum Nmr 373

Spectroscopy of Aqueous Solutions. Journal of Biomolecular Nmr 2: 661-665. 374

Rommel, R., and Jenkins, B. C.. 1959 Amphiploids in Triticeae produced at the University of 375

Manitoba from March 1958 to December 1959. Wheat Inf. Serv. 9-10: 23. 376

Römheld V. 1991. The Role of Phytosiderophores in Acquisition of Iron and Other Micronutrients in 377

Graminaceous Species - an Ecological Approach. Plant and Soil, 130:127-134. 378

Samtsevich, S. A. 1965. Active excretion of plant roots and their significance. Soviet Pl. Physiol. 12: 379

731-740. 380

Sauerbeck, D., Nonnen, S. and Allard, J.L. 1981. Consumption and turnover of photosynthates in the 381

rhizosphere depending on plant species and growth conditions. Landw. Forschung Sonderheft 382

37:207-216. 383

Scheidemann, P. and Wetzel, A. 1997. Identification and characterization of flavonoids in the root 384

exudate of Robinia pseudoacacia. Trees (Berlin) 11: 316-321. 385

Shenker, M., Hadar, Y., and Chen, Y. 1995. Rapid method for accurate determination of colorless 386

siderophores and synthetic chelates. Soil Science Society of America J. 59: 1612-1618. 387

Shenker, M., Fan, T.W.M., and Crowley, D. E. 2000. Phytosiderophores influence on cadmium 388

uptake by wheat and barley Plants. Journal of Environmental Quality, submitted. 389

Stohs, S.J. and Bagchi, D. 1995. Oxidative mechanisms in the toxicity of metal ions. Free Radical 390

Biology in Medicine 18: 321–336. 391

Whipps J.M. 1990. Carbon economy. pp. 59-97 in: J.M. Lynch, The Rhizosphere. John Wiley & 392

Sons. Chichester, England. 393

Yang, C.H. and Crowley, D.E. 2000. Rhizosphere microbial community structure in relation to 394

root location and plant iron nutritional status. Applied and Environmental Microbiology, 66: 395

345-351. 396

18

Zhang F., V. Römheld, and Marschner, H. 1991. Diurnal rhythm of release of phytosiderophores and 397

uptake rate of zinc in iron-deficient wheat. Soil Science and Plant Nutrition 37:671-678. 398

399

19

FIGURE LEGENDS 399

400

Figure 1. 2-D 1H-TOCSY, 1H-13C HSQC, and 1H-13C HMBC spectra of crude AgCS wheat 401

exudate. 402

The TOCSY spectrum (a) was acquired with 2.5 s interpulse delay, 5 kHz spectral width in both 403

dimensions, 8.2 kHz B1 field strength, 47.3 ms mixing time, and 0.41s and 0.072 s acquisition time in 404

F2 and F1, respectively. The HSQC spectrum (b) was obtained with 2.5 s interpulse delay, 5 kHz 405

spectral width in F2 and 25 kHz in F1, and 0.147 s and 0.022 s acquisition time in F2 and F1, 406

respectively. The HMBC spectrum (c) was recorded with the same parameters as the HSQC 407

spectrum except for 0.499 s acquisition time in F2. DMA: 2’-deoxymugineic acid (right structure); CA, 408

CB: A and B carbons of morpholinoethane sulfonate (MES, left structure), respectively. The pH of the 409

exudate sample was 2.5. 410

411

Figure 2. Full scan electrospray mass spectrum of crude AgCS wheat exudate. 412

The ES-MS spectrum was acquired in 50% methanol plus 1% formic acid from a LC-Q mass 413

spectrometer 414

415

Figure 3. 2-D 1H-TOCSY spectrum of Chinese Spring wheat root exudate. 416

The TOCSY spectrum was acquired using 1.5 s interpulse delay, 5 K Hz spectral width in both 417

dimensions, 8.54 kHz B1 field strength, 47.3 ms mixing time, and 0.41 s and 0.051 s acquisition time in 418

F2 and F1, respectively. Peaks 1-9: H-2’, H-2”, H-3, H-1”, H-1’, H-3’, H-4, H-3”, and H-2 of 2’-419

DMA, respectively; lac: lactate, GB: glycinebetaine, α,β-Glc: α,β-glucose. The pH of the exudate 420

sample was 7.5. 421

422

Figure 4. 1-D 1H NMR spectra of barley, wheat (CS and AgCS), and rice root exudates. 423

The 1-D spectra were acquired using 2 s acquisition time, 2 (rice) or 5 s (the rest) interpulse delay, 424

5000 Hz spectral width, and 512 (rice) or 128 (the rest) transients. The abbreviations are as those in 425

Figures 1 and 5. The pH values of barley and AgCS root exudates were 2.2 and 2.5, respectively 426

while those of rice and CS root exudates were 7.5. 427

428

Figure 5. Relation of barley root exudate components to root metal contents. 429

20

The root exudate components were analyzed as described in Materials and Methods while the root 430

metal content was acquired from Shenker et al., submitted to JEQ. 3-epi-OHMA, 3-epi-431

hydroxymugineic acid; 2’-DMA, 2’-deoxymugineic acid; MA Sum, 3-epi-OHMA+2’-DMA+ MA. 432

Two-way ANOVA test or Duncan’s multiple range analysis of variance from the GLM statistical 433

package of SAS was performed on the data. Different letters (italicized letters for exudate components 434

and normal letters for root metal content) indicate statistical difference at a p < 0.05 significance level. 435

436

Figure 6. Relation of wheat root exudate components to root metal contents. 437

The root exudate components were analyzed as described in Materials and Methods while the root 438

metal content and concentration of Fe mobilizing substance (FeMS) were acquired from Shenker et al., 439

submitted to JEQ. 2’-DMA, 2’-deoxymugineic acid; MES, morpholinoethane sulfonate. Statistical 440

analysis was performed as in Figure 5. 441

Page: 4 [DC1] Do we have a reference for this? Perhaps one of Parkers papers on chelator buffering. Otherwise not important unless the reviewer ask for more documentation. Page: 8 [DC2] This is highly speculative since lactate is a very weak ligand. Nor am I aware that alanine and GAB have high affinity for iron as compared to protons or other ions such as sodium and calcium. This statement could be controversial to those who use Geochem to speciate soil solutions. Eg Parker would strongly disagree with their importance. The only way to tell would be to do some empirical test and use them as soil extractants. Page: 8 [DC3] Again, the role of malate in metal uptake is speculative and there is no evidence that this compound would be effective. An article by Jones and Darrah suggest that malate would be of little importance: See: DL Jones, and PR Darrah, Role of root derived organic acids in the mobilization of nutrients from the rhizosphere. Plant and Soil 166:247-257, 1994. Page: 9 [DC4] This is an interesting idea, but is there any prior literature suggesting that amino acids bind Cd or are involved in its uptake?

N

OH O OH O

NH

OH O

OH23 4

1'

2'3'

1"

2"3"

3"-DMA3-DMA

2"-DMA2'-DMA

1'-D

MA

2-DMA

4-D

MA

1"-DMA

3'-DMAN OH2C

CH2S-O

O

O

A

B

A

AB

5.0 4.0 3.0 2.04.5 3.5 2.5

1H Chemical Shift (ppm)

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Fig. 1A

2"-DMA2'-DMA

3-DMA

1'-DMA

4-DMA3'-DMA

1"-DMA

N

OH O OH O

NH

OH O

OH23 4

1'

2'3'

1"

2"3"

3"-DMA

2-DMA

1'-DMA 3-DMA2"-DMA

2-D

MA3"-DMA

4-DMA2'-DMA

3'-DMA

1"-DMA

70 65 60 55 50 45 40 35 30 25 20

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

13C Chemical Shift (ppm)

1H C

hem

ical Sh

ift (pp

m)

Fig. 1B

N OH2C

CH2S-O

O

O

A

B

A

CA CB

CA

1'-DMA 3-DMA2"-DMA

2-D

MA3"-DMA

4-DMA2'-DMA

3'-DMA

1"-DMA

3"-DMA

2-DMA

4-DMA

2"-DMA

1"-DMA

3-DMA

2'-DMA

3'-DMA

1'-DMA

70 65 60 55 50 45 40 35 30 25 2013C Chemical Shift (ppm)

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1H C

hem

ical Sh

ift (pp

m)

Fig. 1C

N

OH O OH O

NH

OH O

OH23 4

1'

2'3'

1"

2"3"

4"4'1

N OH2C

CH2S-O

O

O

A

B

A

CA CB

CA

2'-DMA

2'-DMA+NaMES

MES+Na

100

90

80

70

60

50

40

30

20

10

0

M/Z

Rel

ativ

e A

bu

nd

ance

100 150 200 250 300 350 400 450 500

MES dimer+Na

Fig. 2

1.01.52.02.53.03.54.04.55.0


1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1.0

Ala

lac

acetate

Valmalate

citrate

GB

βGlcαGlc

αTrehalose

123

45

6

7

89

Gly

Leu Ile

Fig. 3

1H Chemical Shift (ppm)5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

AgCS Wheat Exudate

2'-DMA2'-DMA 2'-DMA

N OH2C

CH2S-O

O

O

A

B

A

MES

MES

Unknown

CS Wheat Exudate

acetate

Gly

glycinebetaine

lactateAla2'-DMA

2'-DMA2'-DMA

Valmalate

succinate

IleLeu

2'-DMA

Barley Exudate

lactateAla

glycinebetaine

GAB

GAB

Val

succinate

MAs

malate

Thr

2'-DMA

MA

epi-OHMAMAs

Glu

malate

R = H, 2'-DMAR = OH, MA

COO-

CC

NC

NH

COO-

CC

COO-

OHH

R'

HHH H

H HH H

1

23

4

1'

2'

3'

1"

2"

3"CH R

R = R’= OH,

3- epi-OH-MA

MES

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0


Rice Exudate

lactate

Gly

acetate

Fig. 4

Mn vs Alanine

Fe / Cd Fe / Cd Fe / Cd Fe / Cd

bab

b

a

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0

100

200

300

400

500alanine

Mn

a

a

aa

Cd vs Malate

0.0

0.2

0.4

0.6

0

50

100

150

200

250malate

Cd

Fe vs 3-epi-OHMA

0.0

2.0

4.0

6.0

8.0

10.0

0

20

40

60

80

1003-epi-OH-MA

Fe

Fe vs MA

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0

20

40

60

80

100MA

Fe


a

b b

a

b b

a

a

10

20

30

0

Fe vs Lactate

0.0

0.8

1.6

2.4

3.2

4.0

0

20

40

60

80

100lactate

Fe

Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd

Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd Fe / Cd


Cd vs Threonine

0.0

0.2

0.4

0.6

0

50

100

150

200

250threonine

Cda

a

b b

b

a

bb


mo

le E

xud

ate/

g r

oo

t d

ry w

t.

g M

etal

/g r

oo

t d

ry w

t.

b

a

b

a

aa

b b

a

b b

a

a

a

a

a

a

a

b

a

b b

a

a

Fig. 5

Zn vs 3-epi-OHMA

0.0

2.0

4.0

6.0

8.0 3-epi-OH-MA

Zn

ab

b

a

ab

a

b b

a

Cu vs MA Sum

0

20

40

60

80

100

120MA sum

Cu

a

bb

a

b b

0.0

2.0

4.0

6.0

8.0

10.0 a

a

Fe vs 2'-DMA

Zn vs 2'-DMA

Cu vs 2'-DMA120

Cd vs 2'-DMA


Mn vs MESMES

Mn


mo

le E

xud

ate/

g r

oo

t d

ry w

t.

g M

etal

/g r

oo

t d

ry w

t.




Fe vs FeMS

0

6

12

18

24

30

0

20

40

60

80

100Total FeMS

Fe


a

b

c

a

bb

a

b

a

b

a

b

a

b

a

b

a

b

a

b

0

40

80

120

160

200

240

0

100

200

300

400

500

a

b

b

a

bb

a

b

a

b

0

6

12

18

24

30

0

20

40

60

80

1002'-DMA

Fe

c

a

c

b

0

6

12

18

24

30

0

8

16

242'-DMA

Zna

b

a

b

c

bab

a

0

6

12

18

24

30

0

10

20

30

2'-DMA

Cua

ab

0

6

12

18

24

30

0

100

200

300

400

500

6002'-DMA

Cd

a a

b b

Fig. 6

Chemical profiling of gramineous exudates using high -resolution … › em52 › eprints ›...

Documents

Transcript of Chemical profiling of gramineous exudates using high -resolution … › em52 › eprints ›...