Drought stress increases the production of 5-hydroxynorvaline in two C4 grasses

Article abstract of DOI:10.1016/j.phytochem.2009.03.001

Plants produce various compounds in response to water deficit. Here, the presence and identification of a drought-inducible non-protein amino acid in the leaves of two C₄ grasses is first reported. The soluble amino acids extracted from the lea

Full text of DOI:10.1016/j.phytochem.2009.03.001

Phytochemistry 70 (2009) 664–671
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Drought stress increases the production of 5-hydroxynorvaline in two C4 grasses
b
b
b
b
Ana E. Carmo-Silva a,b, , Alfred J. Keys , Michael H. Beale , Jane L. Ward , John M. Baker ,
*
Nathaniel D. Hawkins b, Maria Celeste Arrabaça a, Martin A.J. Parry b
a Centro de Engenharia Biológica and Departamento de Biologia Vegetal, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
b Department of Plant Sciences, Centre for Crop Genetic Improvement, Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, England, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 October 2008
Received in revised form 16 February 2009
Available online 7 April 2009
Plants produce various compounds in response to water deficit. Here, the presence and identification of a
drought-inducible non-protein amino acid in the leaves of two C4 grasses is first reported. The soluble
amino acids extracted from the leaves of three different species were measured by high-performance
liquid chromatography of derivatives formed with o-phthaldialdehyde and b-mercaptoethanol. One
amino acid that increased in amount with drought stress had a retention time not corresponding to
any common amino acid. Its identity was determined by metabolite profiling, using 1H NMR and
GC–MS. This unusual amino acid was present in the dehydrated leaves of Cynodon dactylon (L.) Pers.
and Zoysia japonica Steudel, but was absent from Paspalum dilatatum Poir. Its identity as 2-amino-5-
hydroxypentanoic acid (5-hydroxynorvaline, 5-HNV) was confirmed by synthesis and co-chromatogra-
phy of synthetic and naturally occurring compounds. The amount of 5-HNV in leaves of the more drought
tolerant C4 grasses, C. dactylon and Z. japonica, increased with increasing water deficit; therefore, any
benefits from this unusual non-protein amino acid for drought resistance should be further explored.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Cynodon dactylon
Paspalum dilatatum
Zoysia japonica
Poaceae
C4 grasses
Metabolite profiling
Water deficit
Amino acid
2-Amino-5-hydroxypentanoic acid (5-
hydroxynorvaline, 5-HNV)
1. Introduction
its relative Arabidopsis thaliana, stress-inducible compounds were
present in higher amounts in the absence of stress and increased
Water deficit is one of the most important factors affecting
plant growth, development and survival. In particular in Mediter-
ranean climates, which are characterized by long hot and dry peri-
ods, water availability is the major limitation to plant productivity
(Turner, 2004). To exacerbate the problem, aridity is expected to
increase in many areas of the globe (Petit et al., 1999). Thus, the
efficient management of water resources by agricultural and recre-
ational systems is a priority. Therefore, there is increasing pressure
to improve the efficiency of water use by plants, namely through
the use of species and varieties better adapted to drought
conditions.
Plant adaptability to unfavourable environments may involve a
number of morphological, physiological, biochemical and molecu-
lar adjustments. Some features associated with plant resistance to
drought are constitutive rather than adaptive and result from the
selection of traits conferring better fitness to arid environments
(Chaves et al., 2003). Plant species may adapt to extreme environ-
ments through the acquisition of novel biosynthetic capacities. For
instance, in the extremophile Thellungiela halophila, compared to
to a greater extent under salt stress (Gong et al., 2005).
Metabolic adjustments in response to the adverse environment
conditions may result in production of different types of organic
compounds (Shulaev et al., 2008). Metabolome and transcriptome
data, namely the identification of metabolites and genes that show
increased or decreased amounts and/or expression under drought,
should be considered when defining targets for genetic engineer-
ing of drought tolerance (Seki et al., 2007). In forage and turf
grasses, the molecular basis for stress tolerance remains largely
unknown (Zhang et al., 2006).
Warm-season C4 grasses perform better than their C3 counter-
parts at high temperatures and irradiance levels (Brown, 1999).
Their efficient assimilation of CO2 in combination with lower tran-
spiration rates results in high water use efficiency (Edwards et al.,
1985). In warm dry summer conditions, C4 grasses produce higher
yields than C3 grasses (Gherbin et al., 2007), but better drought
resistance in C4 relative to C3 grasses is not always observed
(Ripley et al., 2007).
Three classical variants of C4 photosynthesis are named
after the main decarboxylating enzyme: NADP-malic enzyme
(NADP-ME), NAD-malic enzyme (NAD-ME) and phosphoenolpyru-
vate carboxykinase (PEPCK). The bio-geographical distribution of
C4 species demonstrates that NADP-ME species are positively
correlated with annual rainfall, whereas NAD-ME species
* Corresponding author. Present address: USDA-ARS, US Arid-Land Agricultural
Research Center, 21881 N. Cardon Lane, Maricopa, AZ 85238, USA. Tel.: +1 520 316
6364; fax: +1 520 316 6330.
E-mail address: elizabete.carmosilva@gmail.com (A.E. Carmo-Silva).
0031-9422/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2009.03.001

A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
665
predominate in arid zones (e.g., Hattersley, 1992; Cabido et al.,
2008). Enhanced water use efficiency in NAD-ME species (Ghan-
noum et al., 2002) may contribute to explain this distribution.
However, as observed by Taub (2000), the distribution of the grass
subfamilies Panicoideae and Chloridoideae is also strongly corre-
lated with the precipitation gradients. Thus, characteristics other
than the biochemistry of photosynthesis may be responsible for
the geographical distributions, and may reflect divergent patterns
associated with the multiple origins of C4 grasses (Kellogg, 1999).
The free amino acids contained in the leaves were investigated
in three different C4 grasses exposed to drought stress: Paspalum
dilatatum Poir. (NADP-ME), Cynodon dactylon (L.) Pers. (NAD-ME)
and Zoysia japonica Steudel (PEPCK). Dallisgrass (P. dilatatum), ber-
mudagrass (C. dactylon) and zoysiagrass (Z. japonica) are perennial
species used for forage and turfgrass purposes throughout the
world and especially in tropical and subtropical regions (Brown,
1999). P. dilatatum belongs to the subfamily Panicoideae, while C.
dactylon and Z. japonica belong to the subfamily Chloridoideae
(Watson and Dallwitz, 1992). Previous studies under rapidly-
imposed (Carmo-Silva et al., 2007) and slowly-induced water
deficit (Carmo-Silva et al., 2008) suggested better drought
tolerance for C. dactylon and Z. japonica than for P. dilatatum.
Gradually-induced drought conditions caused changes in amounts
of many of the readily identified amino acids (Carmo-Silva et al.,
2008, 2009), but one compound that increased with leaf dehydra-
tion was not immediately identified. The characteristics of this
unusual component and its ultimate identification by metabolite
profiling are described.
2. Results and discussion
This investigation arose from more extensive studies of the ef-
fects of drought on leaves and carbon metabolism in C4 grasses.
Soluble amino acids were measured to provide an indication of
the effects of drought on photorespiration by P. dilatatum, C.
dactylon and Z. japonica. The results showed that photorespiration
was slow even in drought-stressed leaves (Carmo-Silva et al.,
2008), but the analysis also demonstrated increases in many amino
acids as a consequence of drought, including a component that was
Fig. 1. Reversed-phase high-performance liquid chromatograms of amino acids present in the leaves from well-watered (a) and drought-stressed (b) plants of Cynodon
dactylon (solid traces). Co-chromatography of each sample with 40 mM 2-amino-5-hydroxypentanoic acid (5-hydroxynorvaline 1, 5-HNV) resulted in co-elution with the
unknown (X) present in the drought-stressed sample (dashed traces). Chromatography of 20 mM 5-HNV (1) alone is also shown (dotted traces).

666
A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
Table 1
amino acid that increased with drought in C. dactylon and Z. japon-
ica but not in P. dilatatum.
Retention times (RT) for various amino compounds as OPA derivatives in reversed-
phase HPLC.
Chromatograms of o-phthaldialdehyde (OPA) derivatives
showed the presence of an unusual compound in the drought-
stressed leaves of C. dactylon (Fig. 1), whereas the amount present
in control leaves was very low or negligible. Leaves of Z. japonica
from both drought-stressed and well-watered plants contained
this compound, but it was not detected in P. dilatatum (chromato-
grams not shown). This unknown component, with a retention
time of 19 min, accounted for up to 7% of the total amino com-
pounds measured through high-performance liquid chromatogra-
phy (HPLC). Besides protein amino acids, many other amino
compounds have been reported to be present in the leaves of
grasses and some of these vary in amount upon exposure to chang-
ing environment conditions (Jones, 1985). An OPA adduct can only
be formed with substances having primary amine groups (Stoney-
Simons and Johnson, 1978). Table 1 shows the retention times of
compounds analysed in attempts to identify the unknown compo-
nent, shown as ‘‘X”.
Since X had at first only been characterized as an OPA derivative,
its detection by other methods of analysis involved comparison of
components in leaf extracts from drought-stressed and well-wa-
tered plants of C. dactylon. After passage of the leaf acid extracts
through a Dowex-50 column, bound components were eluted with
0.05 M NH4OH. HPLC analysis showed that the component of inter-
est appeared in fractions containing the main amino acids. The
purification procedure was applied to extracts from leaves of both
well-watered and drought-stressed plants of C. dactylon.
Amino compound
RT (min)
GSH + GSSG
Aspartate
Glutamate
Asparagine
Aminoadipic acid
Serine
Methioninesulphoximine
Glutamine
Homoserine
2-Aminopimelic acid
Histidine
Citruline
Glycine
2.8
4.0
8.2
10.3
13.0
13.2
13.9
14.8
16.7
17.0
17.3
17.5
17.8
19.0
19.5
19.7
19.8
20.6
20.9
21.4
21.5
22.6
23.9
24.8
25.0
26.3
27.0
27.7
28.2
29.9
30.4
35.4
X
Ornithine
Taurine
b-Alanine
Alanine
2-Aminoisobutyric acid
c
-Aminobutyric acid
3-Aminoisobutyric acid
ACC
Ethanolamine
a-Aminobutyric acid
Ammonia
Tryptophan
Methionine
Valine
Phenylalanine
Isoleucine
Leucine
1H NMR spectroscopic analysis (Fig. 2) indicated the presence of
a new compound in the drought-stressed plant sample. Complex
multiplets were observed in the 1D spectrum at d 1.64 and d
1.91, which were not present in the 1H NMR spectrum of well-wa-
tered plants. A [1H–1H] COSY experiment indicated correlations be-
tween these peaks and suggested further correlations to peaks at d
3.64 and d 3.72. The signal at d 3.72 was characteristic of a proton
on C2 of an amino acid, while a spin-system of three bond cou-
plings in the COSY spectrum (d 3.72–1.91, d 1.91–1.64 and d
1.64–3.64) indicated a chain of four hydrogen-bearing carbon
Putrescine
not a common protein amino acid. Several amino acids, and espe-
cially proline, increased by more than tenfold in the leaves of the
C4 grasses, particularly C. dactylon and Z. japonica, in response to
gradually-induced drought stress (Carmo-Silva et al., 2009). It be-
came a reasonable objective therefore to identify the non-protein
Fig. 2. Sections of 1H NMR spectra and [1H–1H] COSY correlations of leaves from well-watered (grey traces) and drought-stressed (black traces) plants of Cynodon dactylon.

A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
667
atoms, with the chemical shift of the terminal methylene (d 3.64)
being suggestive of an electronegative substituent such as a hydro-
xyl group.
GC–MS analysis (Fig. 3) of the trimethylsilyl (TMS) derivatives
of components from extracts of drought-stressed and well-wa-
tered leaf samples indicated an additional component with a
retention time of 7.35 min in the drought-stressed material. Anal-
ysis of its mass spectrum indicated an M+ of 349 with fragments
5-hydroxynorvaline 1 (5-HNV). Comparison of the mass spectra
of endogenous 5-HNV 1 to that synthesized in the laboratory
showed that there was good agreement between the two spectra.
Thus, by a combination of the structural information from 1H
NMR (Fig. 2) and GC–MS (Fig. 3) analyses, the unusual amino acid
was identified as 2-amino-5-hydroxypentanoic acid (5-hydroxy-
norvaline 1, 5-HNV) (Fig. 4). Co-chromatography of the synthetic
and the naturally occurring compound (Fig. 1) further confirmed
its identity.
consistent with those expected from
a TMS derivative of
Fig. 3. GC–MS analysis of trimethyl silylated derivatives of components in leaf extracts from well-watered (grey trace) and drought-stressed (black trace) plants of Cynodon
dactylon (a) and comparison of mass spectra for the derivatives of synthetic 2-amino-5-hydroxypentanoic acid (5-hydroxynorvaline 1, 5-HNV) and the endogenous
compound (b). Structures giving rise to fragments with m/z 142, 232 and 349 are shown.

668
A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
seems to be no evidence for a metabolic process directly involving
5-HNV (1) in either the synthesis or degradation of proline (2)
and the conversion of 5-HNV (1) to proline (2) in vitro requires
harsh conditions. Gaudry (1951) reported the synthesis of DL-pro-
line (2) from DL-5-HNV (1) by prolonged heating with HCl. Subse-
quently 5-HNV (1) was identified as a product of HCl hydrolysis of
proteins that had been chemically modified. Thomas et al. (1983)
observed the production of 5-HNV (1) through borohydride
reduction of a thiol ester of a glutamate residue present in the
transmembrane human C3 protein, followed by HCl hydrolysis
of the modified protein. Depending on the conditions, proline
(2) was also formed through the formation of 5-chloronorvaline
(3) as an intermediate. Jennings and Anderson (1987) also reacted
exposed carboxyl groups in a transmembrane protein with Wood-
ward’s reagent K, and reduced the adducts formed with borohy-
dride; hydrolysis with HCl released 5-HNV (1) formed from a
glutamate residue and this 5-HNV (1) was partly converted to
proline (2) under the strong acidic conditions. Oxidation of sev-
eral proteins by metal catalysis in the presence of O2, followed
by reduction with borohydride and HCl hydrolysis, produced 5-
HNV (1) derived from proline (2) and arginine residues (Amici
et al., 1989). The presence of 5-HNV (1) in protein hydrolysates
made after reduction with borohydride is a good indicator of prior
oxidation by active oxygen species (Ayala and Cutler, 1996). One
possible advantage of proline (2) accumulation in plants under
stress is the effective scavenging of reactive oxygen species both
by the free amino acid and by proline (2) residues in proteins
(Matysik et al., 2002).
Conserved plant responses under adverse environmental condi-
tions may involve synthesis of compounds of primary metabolism,
while specific divergent responses are more likely to involve prod-
ucts of secondary metabolism (Sanchez et al., 2008). The drought-
induced accumulation of proline in C4 grasses (Jones, 1985) was
observed in P. dilatatum, C. dactylon and Z. japonica, with a similar
variation of this amino acid with leaf dehydration in all three
grasses (Carmo-Silva et al., 2009). Conversely, there is a clear dis-
tinction of the response of 5-HNV (1) in the three C4 species
(Fig. 5). Previous studies with these grasses suggested that C. dact-
ylon and Z. japonica were better adapted to conditions of water
scarcity than P. dilatatum, in agreement with the geographical dis-
tribution of grasses of the C4 subtype NADP-ME (Hattersley, 1992;
Cabido et al., 2008) and the subfamily Panicoideae (Taub, 2000) in
areas with good rainfalls. Z. japonica shows constitutive character-
istics of adaptation to xeric environments that enable it to resist
drought conditions (Carmo-Silva et al., 2009). C. dactylon responds
to decreased water availability by quickly buffering the adverse ef-
fects of the stress event, and thus keeping an active photosynthetic
metabolism, through efficient stomatal control that delays water
loss (Carmo-Silva et al., 2008). The presence of 5-HNV (1) in leaves
of Z. japonica before exposure to water deficit, together with the
drought-inducible increase in the production of 5-HNV (1) both
in leaves of Z. japonica and C. dactylon, suggests a possible link with
an adaptive drought stress response. Neither the biosynthetic
pathway of 5-HNV (1), nor its subsequent metabolism, is known
in plants. The presence and drought-induced increase of 5-HNV
(1) in leaves of C4 grasses with greater drought resistance needs
further investigation in order to understand if this response is com-
mon to other grass species and if it plays an important role in stress
defensive mechanisms.
Fig. 4. Structures of the 2-amino-5-hydroxypentanoic acid (5-hydroxynorvaline, 5-
HNV) and related compounds mentioned in the text: 5-hydroxynorvaline (1),
proline (2), 5-chloronorvaline (3) and glutamic c-semialdehyde (4).
y = 0.006967 x - 0.0000701 x2
(s.e. 0.000507; 0.0000055)
0.25
0.20
0.15
0.10
0.05
0.00
y = 0.2330 - 0.001206 x
(s.e. 0.0169; 0.000213)
40
60
80
100
RWC (%)
Fig. 5. Variation of the content in 5-hydroxynorvaline (1) (5-HNV) with the relative
water content (RWC) in the leaves of Paspalum dilatatum (black diamonds), Cynodon
dactylon (grey squares, solid line) and Zoysia japonica (open triangles, dashed line).
Each data point corresponds to one sample, with 24 samples per species. Regression
lines were fitted when the RWC effect was significant (P < 0.05;
s
R
2 = 0.898,
2 = 0.03151, d.f. = 68, P < 0.001).
The non-protein amino acid5-HNV (1) was present in all leaves of
Z. japonica and in leaves from drought-stressed plants of C. dactylon,
whereas no significant (P > 0.05) amounts were found in P. dilatatum
(Fig. 5). The content of 5-HNV (1) increased with leaf dehydration in
both C. dactylon and Z. japonica, but the increase was much greater in
the former species, attaining values similar to those found in Z.
japonica at ca. 65% RWC. To the best of our knowledge, the presence
of 5-HNV (1) in plant leaves has not previously been reported,
although it is a known constituent of legume seeds (Thompson
et al., 1964; Dunnill and Fowden, 1967). Some molecular responses
commonly observed during the maturation and desiccation of seeds
may also be observed in vegetative tissues upon exposure of plants
to drought conditions (Bray, 1993). Preliminary analyses of leaves
from well-watered and drought-stressed plants of both wheat (C3)
and maize (C4, NADP-ME) showed no significant amounts of 5-
HNV (1) to be present. 5-HNV (1), however, accumulated in cultures
of pyridoxine auxotrophs of Escherichia coli B (Hill et al., 1993), and
an enzyme from Neurospora crassa catalysed the reversible conver-
sion of 5-HNV 1 to glutamic
c-semialdehyde 4 (Yura and Vogel,
3. Concluding remarks
1959). Hill et al. (1993) discussed some conflicting evidence for
metabolism of 5-HNV (1) in Streptomyces sp.
A non-protein amino acid present in the leaves of two grasses
was identified as 5-hydroxynorvaline (1). This compound seems
not to have been reported previously in leaf tissue. The amounts
The structural relationship of 5-HNV (1) to precursors in the
biosynthesis of proline (2) and ornithine have made it of special
interest in the past (Thompson et al., 1964). However, there

A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
669
present were increased by water deficit in two C4 grasses that were
previously shown to be more drought tolerant.
40 min (16,000g) at 4 °C. Supernatants were filtered using 0.2
syringe filters into 2 mL HPLC autosampler vials. Standard solu-
tions of -amino-n-butyric acid, and other amino compounds were
lm
a
prepared in 0.1 M HCl and diluted as needed for introduction into
the column from the autosampler in 0.017 M HCl (final concentra-
tion). Amino acids in leaf samples were quantified by reference to
standards using Waters Millenium32 software (Waters, Milford,
MA, USA).
4. Experimental
4.1. Plant material
The C4 grasses P. dilatatum Poir. cv. Raki, Cynodon dactylon (L.)
Pers. var. Shangri-Lá and Zoysia japonica Steudel ‘Jacklin Sunrise
Brand’ (produced by Jacklin Seed Company, Post Falls, ID, USA)
were grown from seeds in pots with peat-free compost in a glass-
house as previously described (Carmo-Silva et al., 2008). Artificial
light was provided whenever the natural light was below a photo-
4.4. Purification and concentration of the amino acid fraction from leaf
acid extracts
Samples of leaf acid extracts (1 mL) were applied to Dowex-50
NHþ4 columns (0.6 mL). These were eluted individually with 0.05 M
NH4OH, with 1 mL fractions collected. Fractions containing amino
acids were combined, and solutions were freeze-dried.
synthetic photon flux density (PPFD) of 500 l
mol mꢀ2 sꢀ1 during a
16 h photoperiod. Temperature was maintained at ca. 25 °C during
the day and 18 °C during the night. All plants were well-watered
until the beginning of the drought stress treatment. Pots were
placed according to a split-plot design and water deficit was
imposed by ceasing to provide water to the ‘stress’ pots. Young
fully expanded leaves of five-week old plants of P. dilatatum and
C. dactylon and nine-week old plants of Z. japonica were collected
in the growth environment 4–5 h after the beginning of the
photoperiod. Three control and five non-watered pots were used
per species per day for three consecutive days, corresponding to
8–12 days without watering (Carmo-Silva et al., 2008), making a
total of 24 samples per species (nine control and 15 non-watered
pots). A second leaf sample was collected from each pot for deter-
mination of the leaf relative water content (RWC) (Catsky, 1960).
1
4.5. H NMR analyses of a concentrated amino acid fraction from leaf
acid extracts
1H NMR spectra were recorded at 600.05 MHz with 64k com-
plex data points over a sweep width of 12 ppm on a Bruker Avance
600 spectrometer (Bruker BioSpin, Coventry, UK) at a temperature
of 300 K (ca. 28 °C). The freeze-dried amino acid fractions were dis-
solved in CD3OD in D2O (1:4, v/v) and referenced to 0.05% w/v
2,2,3,3-d4-sodium 3-(trimethylsilyl) propionate (d4-TSP) at d 0.00.
The residual HOD peak was eliminated by a pre-saturation pulse
during the 5 s relaxation delay. Parts of the [1H] spectrum of the
new amino acid could be discerned by comparison of the spectra
of drought-stressed and well-watered C. dactylon samples (multi-
plets at d 1.64 and d 1.91). The remainder of the signals of the
new amino acid (d 3.64 and d 3.72) was observed in a [1H–1H] COSY
spectrum of the mixture, that was recorded with 128 t1 increments
of 2048 data points. The spectrum was acquired with 32 scans over
a sweep width of 9 ppm; the residual HOD peak was eliminated by
a pre-saturation pulse during the 1.5 s relaxation delay.
4.2. High-performance liquid chromatography of o-phthaldialdehyde
derivatives
The preparation of acid extracts of leaves, derivatisation with o-
pthaldialdehyde (OPA) in the presence of 2-mercaptoethanol, and
HPLC were carried out essentially as described by Noctor and Foyer
(1998). Analysis used a Waters Alliance 2695 Separation Module, a
474 Scanning Fluorescence Detector and a Waters Symmetry C18
4.6. GC–MS analyses of TMS-amino acids
3.5
autosampler was set to mix and pre-incubate 10
15 L of OPA reagent for 2 min at 25 °C immediately before inject-
lm 4.6 ꢁ 150 mm column (Waters, Milford, MA, USA). The
lL of sample with
A 50 lL sample solution was transferred to a 1.1 mL high recov-
ery vial and evaporated to dryness in vacuo. To the residue ob-
l
ing this mixture onto the column. The OPA reagent (40 mM OPA,
0.28 M 2-mercaptoethanol, 0.45 M H3BO3 adjusted to pH 9.2 with
NaOH) was prepared ca. 12 h before each run and used within
the following three days.
A linear elution gradient was formed using solution A (80%
50 mM NaOAc pH 5.9, 19% MeOH, 1% THF) and solution B (80%
MeOH, 20% 50 mM NaOAc pH 5.9) and was initiated with 100% A
for the first minute and continued with increasing B to 10% from
6 to 10 min, to 45% from 16 to 20 min, to 100% from 32 to
40 min and then decreasing to 0% from 41 to 46 min, at a flow rate
of 0.8 mL minꢀ1. The column temperature was 30 °C. Separations
by similar methods are described by Hunt (1991) and Brückner
and Westhauser (2003).
tained, was added methoxyamine hydrochloride in anhydrous
pyridine (50
90 min with continuous agitation at 1400 rpm in a Thermomixer
Comfort (Eppendorf AG, Cambridge, UK). A volume of 70 L N-
l
L, 16 mg mLꢀ1). The sample was heated at 30 °C for
l
methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) + 1% trichlo-
romethylsilane (TMCS) was then added and the sample was heated
for a further 30 min at 37 °C with continuous agitation (1400 rpm).
GC–MS analysis was carried out using a Pegasus III time of flight
mass spectrometer (Leco Corp., St. Joseph, MI, USA) coupled to an
Agilent 6890N gas chromatograph system (Agilent Technologies,
Palo Alto, CA, USA) fitted with an MPS2 autosampler (Gerstel,
GmbH, Mülheim an der Ruhr, Germany) and a DB-5 ms capillary
column (15 m ꢁ 0.18 mm ID ꢁ 0.18
guard column). An aliquot (0.5 L) of each liquid sample was in-
jected using a splitless injection technique with an inlet tempera-
ture of 280 °C and helium gas carrier flow of 1.4 mL minꢀ1
lm d.f. with 5 m integrated
l
4.3. Acid extraction of leaves
,
Leaf samples (50–200 mg fr. wt.) were placed in liquid N2 and
stored at ꢀ80 °C. Each sample was ground to a fine powder in
liq. N2 and mixed with 0.1 M HCl (1.4 mL). Each mixture was
ground further during thawing to produce a fine suspension. The
homogenate so obtained was centrifuged for 10 min (16,000g) at
constant flow. The oven was set to an initial temperature of 70 °C
for 2 min, then ramped at 17 °C minꢀ1 to 350 °C and held for
1.5 min. The GC interface and source temperatures were 310 and
245 °C, respectively, and EI+ (electron ionization) mass spectra
were acquired from 40 to 800 amu at 70 eV from 3.75 to 20 min
with an acquisition rate of 20 spectra sꢀ1. Spectra of known amino
acids were assigned by reference to an in-house spectral library
and the NIST library. The novel amino acid TMSi derivative gave
an EI spectrum with m/z (%) 349 [0.6, M+], 334 (0.4, M–CH3), 232
4 °C. Samples (100
lL) of each supernatant were then mixed with
120 -amino-n-butyric acid (100 l
l
M
a
L) as an internal standard
and 1 mL of H2O to give an HCl final concentration of 0.017 M. Mix-
tures were stored at ꢀ20 °C, then later thawed and centrifuged for

670
A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
Ayala, A., Cutler, R.G., 1996. The utilization of 5-hydroxyl-2-amino valeric acid as a
specific marker of oxidized arginine and proline residues in proteins. Free
Radical Biol. Med. 21, 65–80.
(24, M–CO2TMSi), 147(914), 142(53), 100(9), 75(22), 73(100), 70
(28), 59(9) and 45(24).
Bray, E.A., 1993. Molecular responses to water deficit. Plant Physiol. 103, 1035–
1040.
Brown, R.H., 1999. Agronomic implications of C4 photosynthesis. In: Sage, R.F.,
Monson, R.K. (Eds.), C4 Plant Biology. Academic Press, New York, pp. 473–507.
4.7. Chemical synthesis
5-Hydroxynorvaline (1) was synthesized from methyl glutamic
acid by adaption of the method of Thompson et al. (1964). Methyl
glutamic acid (1 g) was added in small portions under dry N2,
with stirring, to a 2 M solution of LiBH4 in THF (Sigma–Aldrich,
Gillingham, UK; 12.5 mL). The solution was stirred overnight, and
then heated under N2 until reflux began. This being maintained
for 6 h. After cooling, the reaction mixture was poured into ice-cold
MeOH–H2O (60:40, v/v) and stirred for 40 min. The solution was
acidified to pH 5.0 with glacial HOAc and then concentrated in
vacuo to approximately 10 mL. Examination by TLC (nBuOH–
HOAc–H2O, 4:1:1; ninhydrin detection) showed the presence of
two products. Aliquots (1 mL) of this solution were applied to cat-
ion exchange resin cartridges (10 ꢁ 200 mg, SCX, Extract-Clean,
Alltech Co., Ltd., Tokyo, Japan) washed with H2O with the ‘neutral’
flow-through collected each time. The target compound (lower TLC
spot) was contained in the flow-through with the eluent for each
separation combined (ca. 20 mL). The compound responsible for
the upper TLC spot was retained on these cartridges and then
eluted with 2 M NH4OH. Final purification of 5-hydroxynorvaline
(1) (lower TLC spot) was achieved by elution of a 1 mL (from
20 mL) portion through another SCX cartridge, where it was re-
tained and subsequently eluted with 2 M NH4OH. Evaporation of
the solution gave 5-hydroxynorvaline 1 (6 mg) containing 10%
proline 2 as an impurity, as estimated by NMR spectroscopic and
GC–MS analyses. HREI–MS [(TMSi)3 derivative] m/z 349.1933,
[M+, 0.6%] (calc. for C14H35NO3Si3, 349.1925): m/z (%) 334 (0.4,
M–CH3), 232 (24, M–CO2TMSi), 147(914), 142(53), 100(9),
75(22), 73(100), 70(28), 59(9) and 45(24); HREI–MS [(tBuMe2Si)3
derivative] m/z 475.3336, [M+] (calc. for C23H53NO3Si3, 475.3333).
1H NMR (D2O:CD3OD, 4:1): d 1.64 (2H, m, 4-H2), 1.91 (2H, m,
3-H2), 3.64 (2H, dt, J = 6.3 and 1.3 Hz, 5-H2) and 3.72 (1H, dd,
J = 6.7 and 5.4 Hz, 2-H). This compound was identical (GC–MS
and NMR) to the compound identified in the natural amino acid
extract.
Brückner, H., Westhauser, T., 2003. Chromatographic determination of
amino acids in plants. Amino Acids 24, 43–55.
Cabido, M., Pons, E., Cantero, J.J., Lewis, J.P., Anton, A., 2008. Photosynthetic pathway
L- and D-
variation among C4 grasses along
Biogeogr. 35, 131–140.
a precipitation gradient in Argentina. J.
Carmo-Silva, A.E., Soares, A.S., Marques da Silva, J., Bernardes da Silva, A., Keys, A.J.,
Arrabaça, M.C., 2007. Photosynthetic responses of three C4 grasses of different
metabolic subtypes to water deficit. Funct. Plant Biol. 34, 204–213.
Carmo-Silva, A.E., Powers, S.J., Keys, A.J., Arrabaça, M.C., Parry, M.A.J., 2008.
Photorespiration in C4 grasses remains slow under drought conditions. Plant
Cell Environ. 31, 925–940.
Carmo-Silva, A.E., Francisco, A., Powers, S.J., Keys, A.J., Ascensão, L., Parry, M.A.J.,
Arrabaça, M.C., 2009. Grasses of different C4 subtypes reveal leaf traits related to
drought tolerance in their natural habitats: changes in structure, water
potential and amino acid content. Am. J. Bot., in press.
Catsky, J., 1960. Determination of water deficit in discs cut out from leaf blades. Biol.
Plant. 2, 76–77.
Chaves, M.M., Maroco, J.P., Pereira, J.S., 2003. Understanding plant responses to
drought – from genes to the whole plant. Funct. Plant Biol. 30, 239–264.
Dunnill, P.M., Fowden, L., 1967. The amino acids of the genus Astragalus.
Phytochemistry 6, 1659–1663.
Edwards, G.E., Ku, M.S.B., Monson, R.K., 1985. C4 photosynthesis and its regulation.
In: Barber, J., Baker, N.R. (Eds.), Photosynthetic Mechanisms and the
Environment, vol. 6. Elsevier Science Publishers B.V. (Biomedical Division),
Amsterdam, pp. 287–327.
Gaudry, R., 1951. The synthesis of amino acids from 2,3-dihydrofuran DL-ornithine,
DL-proline, and DL-a-amino-d-hydroxyvaleric acid. Can. J. Chem. 29, 544–551.
Ghannoum, O., von Caemmerer, S., Conroy, J.P., 2002. The effect of drought on plant
water use efficiency of nine NAD-ME and nine NADP-ME Australian C4 grasses.
Funct. Plant Biol. 29, 1337–1348.
Gherbin, P., De Franchi, A.S., Monteleone, M., Rivelli, A.R., 2007. Adaptability and
productivity of some warm-season pasture species in
environment. Grass Forage Sci. 62, 78–86.
a Mediterranean
Gong, Q., Li, P., Ma, S., Rupassara, S.I., Bohnert, H.J., 2005. Salinity stress adaptation
competence in the extremophile Thellungiella halophila in comparison with its
relative Arabidopsis thaliana. Plant J. 44, 826–839.
Hattersley, P.W., 1992. C4 photosynthetic pathway variation in grasses (Poaceae):
its significance for arid and semi-arid lands. In: Chapman, G.P. (Ed.), Desertified
Grasslands: Their biology and Management. Linnean Society Symposium Series,
No. 13. Academic Press, London, pp. 181–212.
Hill, R.E., White, R.L., Smith, K.C., 1993. The identification of 5-hydroxy-L-norvaline
in cultures of pyridoxine auxotrophs of Escherichia coli B. J. Nat. Prod. 56, 1246–
1254.
Hunt, S., 1991. Non-protein amino acids. In: Rogers, L.J. (Ed.), Methods in Plant
Biochemistry
– Amino Acids, Proteins and Nucleic Acids. Academic Press,
London, pp. 1–52.
4.8. Statistical analysis
Jennings, M.L., Anderson, M.P., 1987. Chemical modification and labeling of
glutamate residues at the stilbenedisulfonate site of human red blood cell
band 3 protein. J. Biol. Chem. 262, 1691–1697.
Jones, C.A., 1985. C4 Grasses and Cereals: Growth, Development and Stress
Response. John Wiley & Sons, New York.
Kellogg, E.A., 1999. Phylogenetic aspects of the evolution of C4 photosynthesis. In:
Sage, R.F., Monson, R.K. (Eds.), C4 Plant Biology. Academic Press, New York, pp.
411–444.
Matysik, J., Alia, Bhalu, B., Mohanty, P., 2002. Molecular mechanisms of quenching
of reactive oxygen species by proline under stress in plants. Curr. Sci. 82, 525–
532.
Regression analysis was applied to the variation of the content
of 5-HNV with RWC in the leaves of the C4 grasses, using GenStatÒ
9.2, 2005 (Lawes Agricultural Trust, Rothamsted Research,
Harpenden, UK). The non-significantly different parameters
(t-tests, P < 0.05) of the fitted model (F-test, P < 0.05) were amal-
gamated in order to attain parsimony. The residuals were checked
and found to conform to the assumptions of the analysis.
Noctor, G., Foyer, C.H., 1998. Simultaneous measurement of foliar glutathione,
gamma-glutamylcysteine, and amino acids by high-performance liquid
chromatography: comparison with two other assay methods for glutathione.
Anal. Biochem. 264, 98–110.
Acknowledgements
Petit, J.R., Jouzel, J., Raynaud, D., Barkov, N.I., Barnola, J.M., Basile, I., Bender, M.,
Chappellaz, J., Davis, M., Delaygue, G., Delmotte, M., Kotlyakov, V.M., Legrand,
M., Lipenkov, V.Y., Lorius, C., Pepin, L., Ritz, C., Saltzmanl, E., Stievenard, M.,
1999. Climate and atmospheric history of the past 420,000 years from the
Vostok ice core, Antarctica. Nature 399, 429–436.
Ripley, B.S., Gilbert, M.E., Ibrahim, D.G., Osborne, C.P., 2007. Drought constraints on
C4 photosynthesis: stomatal and metabolic limitations in C3 and C4 subspecies
of Alloteropsis semialata. J. Exp. Bot. 58, 1351–1363.
A.E. Carmo-Silva acknowledges Fundação para a Ciência e a Tecn-
ologia, Portugal, for financial support (Ph.D. Grant SFRH/BD/13730/
2003). Rothamsted Research is a Grant-aided Institute of the Bio-
technology and Biological Sciences Research Council, UK. The
authors thank AgResearch, Margot Forde Forage Germplasm Cen-
tre, New Zealand, and Alípio Dias e Irmão, Lda, Portugal, for provid-
ing grass seeds; and Dr. Samuel Dufour, Rothamsted Research, for
help and advice on Analytical Chemistry.
Sanchez, D.H., Siahpoosh, M.R., Roessner, U., Udvardi, M., Kopka, J., 2008. Plant
metabolomics reveals conserved and divergent metabolic responses to salinity.
Physiol. Plant. 132, 209–219.
Seki, M., Umezawa, T., Urano, K., Shinozaki, K., 2007. Regulatory metabolic networks
in drought stress responses. Curr. Opin. Plant Biol. 10, 296–302.
Shulaev, V., Cortes, D., Miller, G., Mittler, R., 2008. Metabolomics for plant stress
response. Physiol. Plant. 132, 199–208.
References
Amici, A., Levine, R.L., Tsai, L., Stadtman, E.R., 1989. Conversion of amino acid
residues in proteins and amino acid homopolymers to carbonyl derivatives by
metal-catalyzed oxidation reactions. J. Biol. Chem. 264, 3341–3346.
Stoney-Simons Jr., S., Johnson, D.F., 1978. Reaction of o-phthalaldehyde and thiols
with primary amines: fluorescence properties of 1-alkyl (and aryl) thio-2-
alkylisoindoles. Anal. Biochem. 90, 705–725.

A.E. Carmo-Silva et al. / Phytochemistry 70 (2009) 664–671
671
Taub, D.R., 2000. Climate and the US distribution of C4 grass subfamilies
and decarboxylation variants of C4 photosynthesis. Am. J. Bot. 87,
1211–1215.
Turner, N.C., 2004. Sustainable production of crops and pastures under drought in a
Mediterranean environment. Ann. Appl. Biol. 144, 139–147.
Watson, L., Dallwitz, M.J., 1992. The Grass Genera of the World. CAB International,
Wallingford.
Thomas, M.L., Davidson, F.F., Tack, B.F., 1983. Reduction of the b-Cys-c-Glu thiol
ester bond of human C3 with sodium borohydride. J. Biol. Chem. 258, 13580–
13586.
Yura, T., Vogel, H.J., 1959. An x-hydroxy-a-amino acid dehydrogenase of
Neurospora crassa: partial purification and some properties. J. Biol. Chem. 234,
339–342.
Zhang, Y., Mian, M.A.R., Bouton, J.H., 2006. Recent molecular and genomic studies on
stress tolerance of forage and turf grasses. Crop Sci. 46, 497–511.
Thompson, J.F., Morris, C.J., Hunt, G.E., 1964. The identification of
hydroxynorvaleric acid and -homoserine in Jack bean seeds (Canavalia
ensiformis). J. Biol. Chem. 239, 1122–1125.
L-a-amino-d-
L