Efficiency of different methods of extraction and purification of cytokinins

Article abstract of DOI:10.1016/j.phytochem.2006.03.010

The increasing use of advanced methods, such as mass spectrometry, for the determination of cytokinins has raised special requirements for the extraction and purification of this class of plant hormones. Extraction of Arabidopsis thaliana plants with three different solvents, [80% (v/v) MeOH, Bieleski's MCF-7, and modified Bieleski's] provided similar yields of most analyzed cytokinins determined by high-performance liquid chromatography-tandem mass spectrometry (HPLC/MS/MS). However, the extraction with a modified Bieleski's solvent (MeOH-HCO₂H-H₂O [15:1:4, v/v/v]) gave the highest responses of deuterated cytokinins (used as test compounds) in plant extracts as compared to the responses of pure deuterated standards (relative internal standard response, RISR). Purification of cytokinins using Oasis MCX sorbent with reversed-phase and cation-exchange characteristics, in comparison to the DEAE Sephadex RP-C₁₈ method, provided higher levels of zeatin riboside monophosphate and similar levels of cytokinin bases, ribosides and glucosides. Using this method the content of UV-absorbing contaminates was decreased by about 90% and the RISR values of all tested cytokinin standards but riboside monophosphates were increased about two-fold. The former method provided preparations more suitable for HPLC/MS/MS analysis with respect to simplicity and sample purity.

Full text of DOI:10.1016/j.phytochem.2006.03.010

PHYTOCHEMISTRY
Phytochemistry 67 (2006) 1151–1159
www.elsevier.com/locate/phytochem
Efficiency of different methods of extraction and
purification of cytokinins
a
a
a
a
Kl a´ ra Hoyerov a´ , Alena Gaudinov a´ , Ji rˇ ´ı Malbeck , Petre I. Dobrev ,
b
a
a
a,
*
ˇ
Tom a´ sˇ Koc a´ bek , Blanka Solcov a´ , Alena Tr a´ vn ´ı cˇ kov a´ , Miroslav Kam ´ı nek
a
Institute of Experimental Botany, Academy of Sciences of the Czech Republic, Rozvojov a´ 135, CZ-165 02 Praha 6, Czech Republic
ˇ
Institute of Plant Molecular Biology, Academy of Sciences of the Czech Republic, Brani sˇ ovsk a´ 31, CZ-370 05 Cesk e´ Bud eˇ jovice, Czech Republic
b
Received 2 December 2005; received in revised form 6 March 2006
Available online 5 May 2006
Abstract
The increasing use of advanced methods, such as mass spectrometry, for the determination of cytokinins has raised special require-
ments for the extraction and purification of this class of plant hormones. Extraction of Arabidopsis thaliana plants with three different
solvents, [80% (v/v) MeOH, Bieleski’s MCF-7, and modified Bieleski’s] provided similar yields of most analyzed cytokinins determined
by high-performance liquid chromatography-tandem mass spectrometry (HPLC/MS/MS). However, the extraction with a modified Biel-
2 2
eski’s solvent (MeOH–HCO H–H O [15:1:4, v/v/v]) gave the highest responses of deuterated cytokinins (used as test compounds) in
plant extracts as compared to the responses of pure deuterated standards (relative internal standard response, RISR). Purification of
cytokinins using Oasis MCX sorbent with reversed-phase and cation-exchange characteristics, in comparison to the DEAE Sephadex
RP-C18 method, provided higher levels of zeatin riboside monophosphate and similar levels of cytokinin bases, ribosides and glucosides.
Using this method the content of UV-absorbing contaminates was decreased by about 90% and the RISR values of all tested cytokinin
standards but riboside monophosphates were increased about two-fold. The former method provided preparations more suitable for
HPLC/MS/MS analysis with respect to simplicity and sample purity.
ꢀ
2006 Elsevier Ltd. All rights reserved.
Keywords: Arabidopsis thaliana; Cytokinin extraction; Cytokinin purification; Plant hormones; Solid-phase extraction; HPLC/MS/MS
1
. Introduction
biosynthesis. More than 40 natural cytokinins have been
identified in plant tissues (Za zˇ ´ı malov a´ et al., 1999). The
structures of those cytokinins analyzed in the present
report are shown in Fig. 1. Their occurrence in minute
Cytokinins are a class of plant hormones that in coop-
eration with auxin play unique role in the control of
developmental processes in plants such as cell division
and differentiation, formation and growth of roots and
shoots, apical dominance and senescence. Natural cytoki-
nins are 6-N-substituted purine derivatives. Those which
occur in plants as free bases are supposed to be the
biologically active compounds. Glycosidic conjugates of
cytokinins are transport, storage or inactivated forms of
cytokinins; while cytokinin riboside phosphates predomi-
nantly represent the primary products of cytokinin
ꢀ
8
amounts in non-transformed plants (<10 M), in the
presence of structurally related compounds and enzymes
catalyzing metabolic conversions and the degradation of
cytokinins (Mok and Mok, 2001), complicates both their
purification and determination (Horgan and Scott, 1987;
Jones et al., 1996). Increasing use of mass spectrometry
for the detection and determination of a wide spectrum
of cytokinins requires both the operational as well as
dependable extraction and purification techniques which
prevent metabolic conversions during extraction, and also
provide samples of sufficient purity for mass
spectrometry.
*
Corresponding author. Tel.: +420 2 20390445; fax: +420 2 20390446.
E-mail address: kaminek@ueb.cas.cz (M. Kam ´ı nek).
0
031-9422/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2006.03.010

1152
K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
Fig. 1. Structures, names and abbreviations of cytokinins investigated.
A wide array of solvents, particularly MeOH, EtOH,
Anion-exchange column chromatography provides an
efficient step for the separation of cytokinin riboside phos-
phates from cytokinin bases, ribosides and glucosides.
DEAE-cellulose and DEAE-Sephadex have been success-
fully used for this purpose. Cytokinin riboside phosphates,
applied in neutral aqueous solution, are retained on such
sorbents and can be eluted with an ammonium salt
(ꢁ1 M) or 6% formic acid. Cytokinin bases, ribosides
and glucosides, which are not bound to anion-exchange
sorbents, can be retained either on a reversed-phase (RP)-
C18 or on a cation-exchange column, attached in tandem
with the anion-exchange column (MacDonald and Morris,
1985; Redig et al., 1996; Prinsen et al., 1995).
New solid-phase extraction (SPE) sorbents bearing both
the RP and cation-exchange characteristics greatly increase
and simplify the purification of cytokinins and allow sepa-
ration of different plant hormones (auxin, abscisic acid,
and cytokinins) on a single column, after their simulta-
neous extraction. Moreover, using MCX RP cation-
exchange sorbent, cytokinin riboside monophosphates
were separated from cytokinin bases, ribosides and gluco-
sides at high recoveries by simple step elution, with solvents
of increasing pH and concentration of MeOH (Dobrev and
Kam ´ı nek, 2002). Another method using mixed mode
reversed-phase anion-exchange SPE prior to two-dimen-
sional HPLC was recently developed for purification of
auxin and abscisic acid (Dobrev et al., 2005).
perchloric acid and their mixtures, have been used for the
extraction of cytokinins (Laloue et al., 1974; Horgan and
Scott, 1987; Crouch and Van Staden, 1992). Attention has
been paid to the inactivation of phosphatases catalyzing
the hydrolysis of cytokinin riboside phosphates during
extraction. Bieleski (1964) developed MCF-7 solvent con-
sisting of MeOH–CHCl –HCO H–H O (12:5:1:2, v/v/v/v)
3
2
2
that was found to inactivate phosphatases in various plant
materials. This solvent has also been widely used for the
extraction of cytokinins.
As was shown in the model experiments of Horgan
and Scott (1987), the extraction of soybean callus tissue
with Bieleski’s solvent minimizes hydrolysis of AMP.
However, preservation of cytokinin riboside phosphates
during extraction of the plant material with this solvent
has not been directly tested yet. Analyzing soybean callus
1
4
tissue after feeding with [ C]benzyladenine, which was
metabolically converted in vivo to the corresponding
riboside monophosphate, Crouch and Van Staden
(
1992) found no substantial differences in the proportions
of riboside monophosphate and the corresponding
dephosphorylated product (riboside), when freeze-dried
tissue was extracted with either 80% (v/v) EtOH or with
Bieleski’s solvent. Moreover, the presence of CHCl3 in
the Bieleski’s solvent was reported to increase the extrac-
tion of lipophilic material that complicates further purifi-
cation (Laloue et al., 1974; Horgan and Scott, 1987). The
present availability of labeled cytokinin riboside mono-
phosphates may allow accurate comparisons of different
solvents, with respect to protection of cytokinin riboside
phosphates.
Purification of cytokinins using immunoaffinity chroma-
tography allows very powerful purification of most cytoki-
nins (MacDonald and Morris, 1985; Va nˇ kov a´ et al., 1998;
Corbesier et al., 2003). Antibodies of low specificity allow-
ing binding of a wide spectrum of cytokinins have been

K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
1153
successfully used for this purpose. However, even these
antibodies do not bind cytokinin N-7- and O-glucosides
tested, but riboside monophosphates that were determined
after enzymatic hydrolysis as corresponding ribosides. As
described in Section 4.6. the hydrolysate was purified on
Sep-Pak Plus C18 column where, evidently, contaminants
affecting RISR were removed. The results indicate that the
solvents differ in their extraction efficiency of compounds
other than cytokinins, which are not sufficiently removed
during purification and may influence the sensitivity, accu-
racy and reliability of the determination of cytokinins by
HPLC/MS/MS.
Differences in RISR values, between the extractions with
Bieleski’s and modified Bieleski’s solvent, show that CHCl3
in the Bieleski’s solvent significantly enhances the extrac-
tion of compounds, decreasing the RISR of all tested cyt-
okinins but Z where the difference was not statistically
significant. The fact that differences in RISR values in plant
material extracted with 80% MeOH and modified Bieleski’s
solvent are statistically significant only for few cytokinins
(iP, iPR and iP7G) supports such opinion. The RISR val-
ues exceeding 1.0 (extraction of iPR with 80% MeOH and
modified Bieleski’s) indicate an increase of the signal due to
the matrix effect.
(
Banowetz, 1994; Jordi et al., 2000) that need to be ana-
lyzed separately (Corbesier et al., 2003). Cytokinin O-glu-
cosides washed out from immunoaffinity columns can be
purified on the same type of column after their hydrolysis
by b-glucosidase (Werner et al., 2003; Nov a´ k et al., 2003).
The identification and determination of cytokinins by
high-performance liquid chromatography-mass spectrome-
try (HPLC/MS/MS) is increasingly used in plant hormone
research. In addition to the more rigorous identification of
different cytokinins, these methods also allow the simulta-
neous quantification of a very wide spectrum of cytokinins
in a single run. For these and other reasons, they are
becoming a standard in the analysis of cytokinins in plant
materials (Redig et al., 1996; Prinsen et al., 1998; Nord-
str o¨ m et al., 2004). In contrast to immunological methods
(
RIA and ELISA), mass spectrometry is also suitable for
the determination of cytokinin N7- and O-glucosides which
are not well recognized by antibodies raised against the
corresponding cytokinin ribosides (compare S a´ enz et al.,
2
003 and Nov a´ k et al., 2003). Because of the high selectiv-
ity of mass spectrometry, different cytokinins can be
detected and determined in partially purified samples.
The aim of this paper is to compare the efficiency of
different methods of cytokinin extraction and purifica-
tion, with respect to both the prevention of hydrolysis
of the cytokinin riboside phosphates, as well as sufficient
sample purity for cytokinin determination using HPLC/
MS/MS.
Several authors used Bieleski’s solvent without CHCl3
to reduce extraction of contaminants, avoid the problems
of handling and disposal of CHCl -containing extracts
3
(Wang et al., 1995; Jones et al., 1996; Lighfoot et al.,
1997). However, the effect of this on the extraction and
determination of cytokinins has not been assessed. Accord-
ing to the results presented here, the modified Bieleski’s sol-
vent appeared to be the most suitable of the tested solvents
for the extraction of cytokinins, prior to their determina-
tion using HPLC/MS/MS.
2
. Results and discussion
Cytokinin riboside phosphates can be hydrolyzed to the
corresponding ribosides by phosphatases during plant tis-
sue extraction (Horgan and Scott, 1987; Crouch and Van
Staden, 1992; Redig et al., 1996). Conversions of ZRMP
and iPRMP to the corresponding ribosides during extrac-
tion with 80% MeOH, Bieleski’s MCF-7 and modified Biel-
eski’s solvents were tested. Two hundred picomol of
The comparison of efficiencies of the different extraction
and purification methods is based on the determination of
cytokinin levels and relative internal standard responses
(
RISR) of deuterated cytokinins, expressing the ratio of
HPLC/MS/MS chromatogram peak areas of deuterated
cytokinins recorded in plants extracts with those of the cor-
responding pure deuterated standards. The RISR charac-
terizes both the losses during the extraction and
purification, as well as the matrix effects of the mass spec-
trometric analysis.
2
2
[ H ]ZRMP and [ H ]iPRMP were added separately to
5
6
the frozen powder of plant material at the beginning of
the extraction, and the deuterated products were deter-
mined using HPLC/MS/MS after extraction and purifica-
tion by mixed-mode-SPE. Both riboside monophosphates
were hydrolyzed almost exclusively to the corresponding
ribosides, and further hydrolysis of the ribosides to the cor-
responding bases was very low (61.5%, results not shown).
Among all of the tested solvents, iPRMP appeared to be
much more sensitive to hydrolysis than ZRMP undergoing
nearly 60% conversion during extraction with 80% (v/v)
MeOH, as compared to 12% and 18% following extraction
with Bieleski’s and modified Bieleski’s solvents, respec-
tively. Hydrolysis of ZRMP was also the highest in 80%
(v/v) MeOH (13%), as compared to Bieleski’s and modified
Bieleski’s solvents (both 3%) (Fig. 3). Correspondingly,
Horgan and Scott (1987) reported that AMP as a model
riboside monophosphate was also hydrolyzed to adenosine
2
.1. Testing of different extraction solvents
In comparing three different solvents (a) 80% (v/v)
MeOH, (b) Bieleski’s MCF-7, and (c) modified Bieleski’s,
the deuterated cytokinins were added at the beginning of
the extraction, and the extracts were purified using the
mixed-mode-SPE as described in Section 4.3. As shown
in Fig. 2A, the tested solvents gave similar yields of ana-
lyzed cytokinins. Small differences in levels of endogenous
cytokinins confirm the efficacy of the internal standard cal-
ibration method. However, the modified Bieleski’s solvent
provided the highest RISR values for almost all cytokinins

1154
K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
Fig. 2. The effect of three different extraction solvents on cytokinin levels (A) and RISRs, representing the ratio of HPLC/MS/MS responses of deuterated
cytokinins from plant extracts to the responses of pure cytokinin standards (B). The plant material was extracted as described in Section 4.2, and the
cytokinins were purified as specified in Section 4.4. Results are averages of the analyses of three independent samples. Vertical bars indicate SD.
more readily during extraction of soybean callus tissue with
80% MeOH than with Bieleski’s solvent.
Evidently, the activity of phosphatases is sufficiently
suppressed when plant tissue is powdered in liquid nitrogen
and extracted at ꢀ20 ꢁC with diluted MeOH acidified with
HCO H (modified Bieleski’s solvent). Moreover, according
2
to Bieleski (1964), extraction in acid solvents at ꢀ25 ꢁC
speeded-up the deactivation of phosphatases. Unfortu-
nately, he did not test the extraction efficiency of acid sol-
vents lacking CHCl . As shown in Fig. 3, the omission of
3
CHCl from the Bieleski’s MCF-7 solvent in the modified
3
Bieleski’s solvent did not affect hydrolysis of ZRMP, and
only slightly increased hydrolysis of iPRMP.
Fig. 3. Conversion of cytokinin (CK) riboside monophosphates
2
2
2.2. Comparison of purification methods
([ H
5 6
]ZRMP and [ H ]iPRMP) to corresponding CK ribosides following
extraction of Arabidopsis plants with different solvents. The plant material
2
2
5 6
was supplied with 200 pmol of [ H ]ZRMP and [ H ]iPRMP and
Two different purification procedures, (a) DEAE Sepha-
extracted as described in Section 4.2. Cytokinins were purified as specified
in Section 4.2. Results are expressed as a percentage of the cytokinin
riboside monophosphates converted to the corresponding ribosides, and
represent averages of the analyses of three independent plant samples.
Vertical bars indicate SD.
dex RP-C18 and (b) mixed-mode-SPE were compared,
using extracts of plant material with modified Bieleski’s
solvent. Both methods provided similar yields of analyzed
cytokinins but iPR and ZRMP where the latter method

K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
1155
provided significantly higher yields (Fig. 4A). The two
methods differed significantly in RISR of cytokinins that
were almost twice as high in the case of mixed-mode-
SPE, as compared to the DEAE Sephadex RP-C18
due to the different recoveries of the deuterated standards.
Moreover, the recoveries may be concentration-dependent.
To determine the recoveries of cytokinins achieved by the
two purification procedures, independent of mass spectro-
metric analysis, plant extracts corresponding to 1 g of plant
(
Fig. 4B). The ratios of RISRs, recorded by the mixed-
mode-SPE and DEAE Sephadex RP-C18 varied between
.25 (iP) and 2.3 (Z9G). Values of RISRs for some cytoki-
3
3
material were supplied with [2- H]Z and [2- H]ZR, repre-
senting cytokinin base and riboside, just before their appli-
cation onto DEAE Sephadex and dual-mode SPE columns.
So as to know if recovery is affected by cytokinin concen-
tration, a separate set of plant extracts was supplied with
40 pmol of the corresponding cold cytokinins, in addition
1
nins exceeded 1.0, with a maximum of 1.1 (iP7G), most
probably due to the increase of the signal by a matrix effect.
These results indicate that the DEAE Sephadex RP-C18
method was much less efficient in the removal of those con-
taminants suppressing RISRs. This view is supported by
the comparison of total ion current (TIC) of the MS anal-
ysis. TIC values were approximately twice as high during
the whole interval of elution of cytokinins when plant sam-
ples were purified by DEAE Sephadex RP-C , as com-
3
to those labeled [2- H]. This concentration exceeds the
endogenous concentrations of Z and ZR in non-trans-
formed plants (Corbesier et al., 2003; Werner et al., 2003;
3
Ananieva et al., 2004). The recoveries of both [ H]Z and
3
[ H]ZR were very high (P90%) and statistically significant
1
8
pared to mixed-mode-SPE. Moreover, extracts purified
by the DEAE Sephadex RP-C18 exhibited a much greater
variation of TIC profiles of the samples from parallel
extractions, than those purified by mixed-mode-SPE
differences were found neither between the two purification
procedures nor between samples with and without the
addition of cold cytokinins. However, the differences
(1.4–2.9%) were not statistically significant, showing that
within the tested concentration range recoveries are not
dependent on cytokinin concentration (results not shown).
This strongly indicates that differences in the levels of
tested deuterated cytokinins in plant extracts, purified by
(
Fig. 5).
It is possible that the differences in RISRs of the tested
cytokinins, following their purification using the DEAE
Sephadex RP-C18 and dual-mode-SPE, are predominantly
(
A)
(
B)
Fig. 4. The effect of purification of Arabidopsis plant extracts with DEAE Sephadex RP-C18 and mixed-mode-SPE on cytokinin levels (A) and RISRs,
representing the ratio of HPLC/MS/MS responses of deuterated cytokinins from plant extract to the responses of pure cytokinin standards (B). The plant
material was extracted with modified Bieleski’s solvent, as described in Section 4.2, and the cytokinins were purified, as specified in Sections 4.3 and 4.4.
Results are averages of the analysis of three independent plant samples. Vertical bars indicate SD.

1156
K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
(
A)
(
B)
Fig. 5. MS total ion current profiles (TIC) of Arabidopsis plant extracts purified with DEAE Sephadex RP-C18 (A) and mixed-mode-SPE (B), expressed in
arbitrary units. The plant material was extracted with modified Bieleski’s solvent, as described in Section 4.2, and the cytokinins were purified as specified
in Sections 4.3 and 4.4. Each curve corresponds to one separate analysis. Cytokinins were eluted at retention time from 7 min (Z7G) to 27 min (iPR) as
marked by bold lines on the time scales.
the two methods, are caused by the different efficiency of
removal of those contaminants interfering with MS analy-
sis, rather than by their differential recoveries.
ability for the purification of plant extracts prior to the
determination of cytokinins by methods other than
HPLC/MS/MS, such as ELISA and RIA.
Purification of cytokinins using dual-mode-SPE is
apparently extremely efficient in the removal of UV
absorbing contaminants, decreasing the content of UV
absorbing material by 90%, as compared to DEAE Sepha-
dex RP-C18 (Fig. 6). This demonstrates its potential suit-
3
. Conclusion
The modified Bieleski’s, as compared to two other tested
extraction solvents, sufficiently suppressed dephosphoryla-
tion of cytokinin riboside monophosphates and reduced
the extraction of compounds decreasing the RISRs of
tested deuterated cytokinins. This solvent, lacking CHCl3,
is easier and safer to handle and appeared to be the most
suitable for extraction of cytokinins. Purification of cytoki-
nins using mixed-mode-SPE, as compared to DEAE
Sephadex RP-C18 method, was powerful in the removal
UV absorbing contaminants providing preparations exhib-
iting high RISRs of deuterated counterparts of natural cyt-
okinins. This method was found simpler, faster and more
operational. It also allows more complex plant hormone
analysis by providing a partially purified fraction contain-
ing auxin and abscisic acid.
(
A)
(
B)
4. Experimental
4.1. Plant material and chemicals
Fig. 6. HPLC UV traces of Arabidopsis plant extracts, purified with
DEAE Sephadex RP-C18 (A) and mixed-mode-SPE (B). The plant
material was extracted with modified Bieleski’s solvent, as described in
Section 4.2, and the cytokinins were purified, as specified in Sections 4.3
and 4.4. No internal standards were added.
Arabidopsis thaliana (ecotype Columbia) plants were
grown in a greenhouse under 16 h light/8 h dark at
3 ꢁC/20 ꢁC daylight cycle. Leaf rosettes of 55-day old
plants were frozen in liquid nitrogen and stored at ꢀ80 ꢁC.
2

K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
1157
ꢀ
Deuterium-labeled cytokinins were purchased from
(10 ml, HCO form) connected in tandem with a Sep-Pak
3
3
3
Apex (Honiton, Devon, UK). [2- H]Z and [2- H]ZR (both
Plus C . The column was washed with 20 ml of distilled
1
8
ꢀ
1
0
.59 TBq mmol ) were obtained from the Isotope Labora-
water. At this point, the Sep-Pak Plus C18 column (con-
taining cytokinin bases, ribosides and glucosides) and the
DEAE Sephadex column (with entrapped cytokinin ribo-
side phosphates) were disconnected and each column was
separately washed with 10 ml water. Cytokinins from
Sep-Pak Plus C18 were eluted with 5 ml of 80% (v/v)
MeOH and solvent was evaporated under vacuum at
40 ꢁC. The DEAE Sephadex column was connected with
a new Sep-Pak Plus C18 column, and the CK riboside
phosphates were eluted from the former column into Sep-
Pak Plus C18 with 10 ml 1 M NH HCO . Finally, cytoki-
tory of the Institute of Experimental Botany, Prague,
Czech Republic. EtOH of spectrophotometric grade,
MeOH, HCO H, NH OH, NH HCO , all of analytical
2
4
4
3
grade, were supplied by Lachema a.s. (Neratovice, Czech
Republic). Calf intestine alkaline phosphatase (4U/mg)
was a product of Sigma–Aldrich (St. Louis, USA). Sep-
Pak Plus C18 cartridges (containing 400 mg sorbent of
C18 phase bonded on silica gel matrix), and Oasis MCX
columns (containing 150 mg sorbent of a sulfonated
ꢀ
1
copolymer, capacity 1 mequiv g
,
particle diameter
4
3
3
0 lm) were obtained from Waters (Milford, MA, USA).
nin riboside phosphates were eluted from the Sep-Pak
Plus C18 with 5 ml of 80% (v/v) MeOH. The solution
was evaporated under vacuum at 40 ꢁC to water phase,
and the riboside phosphates were hydrolyzed to the corre-
sponding ribosides, as described in Section 4.6.
ꢀ1
DEAE Sephadex A-25, capacity 3.5 mequiv g and parti-
cle size 40–120 lm was product of Pharmacia (Uppsala,
Sweden). Syringes (20 ml) with frits (porosity 20 lm) were
purchased from Isolute SPE Accesories (Alltech, UK). The
SPE procedures were performed using a vacuum manifold
equipped with vacuum control valve that allowed the con-
current processing of 12 samples (Supelco’s Visiprep,
Sigma, St. Louis, USA).
4.4. Purification of cytokinins mixed-mode-SPE
The method of Dobrev and Kam ´ı nek (2002) was used.
Plant extracts were diluted with 5 ml of 1 M formic acid
and applied on an Oasis MCX column. The column was
washed with 5 ml 1 M formic acid, and plant hormones
auxin (IAA) and abscisic acid (ABA) were eluted with
5 ml MeOH. Subsequently, the cytokinin riboside phos-
phates were eluted with 5 ml 0.35 M ammonia in water,
and hydrolyzed by alkaline phosphatase, as described in
Section 4.6. Cytokinin bases, ribosides and glucosides were
eluted in the next step with 5 ml 0.35 M ammonia in 60%
(v/v) MeOH. Solvents were evaporated at 40 ꢁC under
vacuum.
4
.2. Extraction of cytokinins
Three different extraction solvents: (a) 80% (v/v) MeOH,
(
b) Bieleski’s MCF-7 solvent (MeOH–CHCl –HCO H–
3 2
H O 12:5:1:2, v/v/v) and (c) modified Bieleski’s solvent
2
(
MeOH–HCO H–H O; 15/1/4, v/v/v) were compared. Ara-
2 2
bidopsis leaf rosettes were frozen by dipping in liquid nitro-
gen and then homogenized to a powder with a pestle in a
ceramic mortar in liquid nitrogen. One gram plant powder
aliquots were distributed to 50 ml polypropylene centrifuge
tubes. Ten millilitre aliquots of cold (ꢀ20 ꢁC) tested sol-
vents were immediately added together with a mixture of
the following deuterium-labeled cytokinin standards
4.5. Preservation of cytokinin riboside phosphates
(
50 pmol each in total volume of 50 ll of 50% MeOH,
Preservation of cytokinin riboside monophosphates dur-
ing extraction was tested using three different solvents: (a)
80% (v/v) MeOH, (b) Bieleski’s MCF-7 and (c) modified
Bieleski’s solvent. Arabidopsis leaf rosettes were frozen
and extracted as described in Section 4.2. Two hundred
2
2
2
2
2
v/v): [ H ]iP), [ H ]Z, [ H ]DHZ), [ H ]iPR, [ H ]ZR,
6
5
3
6
5
2
2
2
2
2
2
2
[
[
[
H ]DHZR, [ H ]iP7G, [ H ]iP9G, [ H ]Z7G, [ H ]Z9G,
3 6 6 5 5
H ]ZOG, [ H ]ZROG, [ H ]iPRMP, [ H ]ZRMP and
2
2
2
5
5
6
5
H ]DHZRMP. After overnight extraction at ꢀ20 ꢁC sol-
3
2
2
ids were removed by centrifugation (13,000g, 20 min,
ꢁC) and re-extracted with 5 ml of corresponding extrac-
pmol of [ H ]ZRMP and [ H ]iPRMP were added to 1 g ali-
5 6
4
quots of frozen plant powder at the beginning of extraction,
instead of the mixture of deuterated cytokinin standards.
tion solvent by vortexing for 30 s, followed with standing
at ꢀ20 ꢁC for 1 h. Pooled extracts were passed through
Sep-Pak Plus C to remove pigments and lipids and evap-
orated under vacuum at 40 ꢁC near to dryness prior
purification.
2
2
Extracted [ H ]ZRMP and [ H ]iPRMP, as well as their
5
6
ribosides and bases released during extraction, were purified
using mixed-mode-SPE (as described in Section 4.4), and
determined by HPLC/MS/MS using the standard addition
method. The corresponding deuterated ribosides were the
only clearly quantified products. The results were expressed
as the percentage of deuterated riboside phosphates con-
verted to the corresponding deuterated ribosides.
1
8
4
C18
.3. Purification of cytokinins using DEAE Sephadex RP-
The method was adapted according to MacDonald and
Morris (1985) and Redig et al. (1996). Plant extracts con-
taining deuterated cytokinin standards were diluted with
4.6. Hydrolysis of cytokinin riboside monophosphates
1
0 ml water and alkalized with ammonium to pH 7.2–7.5.
Cytokinin riboside monophosphates were determined as
cytokinin ribosides after their dephosphorylation, prior to
The solution was applied to a column of DEAE Sephadex

1158
K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
HPLC/MS/MS analysis. The MCX column fractions con-
taining cytokinin riboside monophosphates were evapo-
rated to water phase and supplied with 2 M CH CO NH
4
Table 1
MS/MS transitions for cytokinin quantitation
Cytokinin
Parent ion
Product ion
3
2
to final 100 mM concentration. After addition of alkaline
phosphatase (0.3 U), the solution was incubated at 37 ꢁC
for 1.5 h, and then neutralized with formic acid to pH 5–
ZxG-d5 (x = 7,9,O)
ZROG-d5
trans-Z-d5
ZR-d5
IPxG-d6 (x = 7,9)
IPR-d6
387
519
225
357
372
342
210
225
387
207
225
210
210
137 + 148
6
. The solution was passed through a Sep-Pak Plus C18
column and entrapped cytokinin ribosides were eluted with
5
ml 80% (v/v) MeOH.
IP-d6
ZxG (x = 7,9,O)
382
514
220
352
366
336
204
220
382
202
220
204
204
136 + 148
3
.7. Recoveries of [ H] cytokinin standards and content of
4
contaminants
ZROG
trans-Z
ZR
IPxG (x = 7,9)
IPR
To determine how the purification procedures affect
cytokinin recovery, independent of the cytokinin extraction
and MS analysis, Arabidopsis leaf rosettes were powdered
in liquid nitrogen and extracted with modified Bieleski’s
solvent under conditions specified in Section 4.2 without
the addition of deuterated cytokinin standards. Prior to
purification of the cytokinins using DEAE Sephadex RP-
C18 and mixed-mode-SPE (as described in Sections 4.3
IP
Dry samples of purified cytokinins were dissolved in
20 ll 50% (v/v) acetonitrile. Solutions were diluted by the
addition of 80 ll H O and passed through Micro-Spin cen-
2
trifuge nylon filters (0.2 lm, Alltech, Deerfield, IL, USA).
The filtrate aliquots of 5 ll were injected onto a C18 HPLC
column (AQUA 250 mm · 2.0 mm, 5 lm, Phenomenex,
Torrance, CA, USA) and eluted at a flow rate of
3
3
and 4.4, respectively), 4 kBq of [2- H]Z and [2- H]ZR were
separately added to extract aliquots corresponding to 1 g of
plant material. To determine the potential effect of cytoki-
nin concentration on recovery, a separate set of extract ali-
quots was supplied with 40 pmol of cold cytokinins (Z and
ꢀ1
0.2 ml min using a linear gradient of acetonitrile (B) in
.0005% (v/v) acetic acid in water (A): 10% B for 5 min,
0
3
to 17% B in 10 min; then to 50% B in 11 min; and finally
increased to 90% B and maintained for 5 min. The column
was equilibrated by the starting composition of the mobile
phase for 20 min before each analytical run. MS data were
collected in positive MS/MS product ion mode for quanti-
fication and the TIC was recorded in a MS full scan (50–
2000 amu) mode.
Endogenous natural cytokinins were determined by the
same HPLC/MS/MS method using the internal standard
calibration and the corresponding deuterated cytokinins
as internal standards. The measured masses (m/z) are
shown in Table 1. The relative responses of internal stan-
dards (RISR) were used to characterize the effectiveness
of extraction and purification procedures (recovery) as well
as the effectiveness in removal of residual contaminants
affecting the MS signal (matrix effect). RISRs are expressed
as the ratio of MS/MS responses of deuterated standards
ZR), in addition to [ H] cytokinins. The radioactive com-
pounds were separated by HPLC using a series 200 auto-
sampler and quaternary pump (both Perkin–Elmer,
Wellesley, MA, USA),
a
Luna C18 (2) column
(
150 · 4.6 mm, 3 lm, Phenomenex, Torrance, CA, USA)
maintained at 35 ꢁC, and two detectors coupled in series:
a 235 C diode array detector (Perkin–Elmer, Wellesley,
MA, USA), and a Ramona 2000 flow-through radioactiv-
ity detector (Raytest, Straubenhardt, Germany). Two sol-
vents (A: 40 mM CH CO H adjusted with NH OH to
3
2
4
pH 4 and B: CH CN/CH OH, 1/1, v/v) were used at a flow
3
3
ꢀ
1
rate of 0.6 ml min with linear gradients of 10–15% B in
min, 15–20% B in 9 min, 20–34% B in 0.1 min, 34–45%
B in 7.9 min, 45–100% B in 2 min, 100% B for 2 min, and
00–10% B in 2 min. The column eluate was monitored
2
1
at 270 nm on the diode array detector and, after on-line
mixing with three volumes of liquid scintillation cocktail
Flo-Scint III (Packard BioScience, Meriden, CT, USA),
on the Ramona 2000 radioactivity detector. The radioac-
tive analytes were identified on the basis of coincidence
of their retention times with authentic standards.
(used as test compounds) added to plant samples at the
beginning of extraction and responses of corresponding
pure deuterated standards.
4
.8. HPLC/MS/MS analysis
Acknowledgements
The analyses were carried out on a HPLC/MS system,
The authors wish to thank Dr. J. Hanu sˇ for synthesis of
radiolabelled Z and ZR, Dr. L. Kohout, Institute of Or-
ganic Chemistry and Biochemistry, Prague, Czech Repub-
lic, and Dr. C. Brinegar, Department of Biological
Sciences, San Jose State University, San Jose, CA, USA,
for his critical reading of the manuscript. Funding of this
research by the Grant Agency of the Academy of Sciences,
consisting of a PAL HTS autosampler (CTC Analytics,
Zwingen, Switzerland) and quaternary HPLC pump
Rheos 2000 (Flux Instruments, Basel, Switzerland) con-
nected to an Ion-Trap mass spectrometer LCQ (Finnigan,
San Jose, CA, USA) equipped with an electrospray inter-
face (ESI).

K. Hoyerov a´ et al. / Phytochemistry 67 (2006) 1151–1159
1159
Czech Republic (project A600380507) by the Grand
Agency of the Czech Republic (project 206/02/0967) is
greatly appreciated.
response. In: Dashek, W.V. (Ed.), Methods in Plant Biochemistry and
Molecular Biology. CRC Press, Boca Raton, FL, pp. 133–152.
MacDonald, E.M.S., Morris, R.O., 1985. Isolation of cytokinins by
immunoaffinity chromatography and analysis by high-performance
liquid chromatography radioimmunoassay. Meth. Enzymol. 110, 347–
3
58.
References
Mok, D.W.S., Mok, M.C., 2001. Cytokinin metabolism and action. Annu.
Rev. Plant Physiol. Plant Mol. Biol. 52, 89–118.
˚
Ananieva, K., Malbeck, J., Kam ´ı nek, M., van Staden, J., 2004. Changes in
endogenous cytokinin levels in cotyledons of Cucurbita pepo (zucchini)
during natural and dark-induced senescence. Physiol. Plant. 122, 133–
Nordstr o¨ m, A., Tarkowski, P., Tarkowska, D., Dolezal, K., Astot, C.,
Sandberg, G., Moritz, T., 2004. Derivatization for LC-electrospray
ionization-MS: A tool for improving reversed-phase separation and
ESI responses of bases, ribosides, and intact nucleotides. Anal. Chem.
76, 2869–2877.
142.
Banowetz, G.M., 1994. Immunoanalysis of cytokinins. In: Mok, D.W.S.,
Mok, M.C. (Eds.), Cytokinins: Chemistry, Activity and Function.
CRC Press, Boca Raton, FL, pp. 305–316.
Bieleski, R.L., 1964. The problem of halting enzyme action when
extracting plant tissues. Anal. Biochem. 9, 431–442.
Nov a´ k, O., Tarkowski, P., Tarkowska, D., Dolezal, K., Lenobel, R.,
Strnad, M., 2003. Quantitative analysis of cytokinins in plants by
liquid chromatography-single-quadrupole mass spectrometry. Anal.
Chim. Acta 480, 207–218.
Corbesier, L., Prinsen, E., Jacqmard, A., Lejeune, P., Van Onckelen, H.,
P e´ rilleux, C., Bernier, G., 2003. Cytokinin levels in leaves and leaf
exudate and shoot apical meristems of Arabidopsis thaliana during
floral transition. J. Exp. Bot. 54, 2511–2517.
Crouch, N.R., Van Staden, J., 1992. Extraction of cytokinin nucleotides
with Bieleski solvents: Is the inconvenience warranted? J. Plant
Physiol. 140, 378–380.
Dobrev, P.I., Kam ´ı nek, M., 2002. Fast and efficient separation of
cytokinins from auxin and abscisic acid and their purification
using mixed-mode solid-phase extraction. J. Chromatogr. A 950,
Prinsen, E., Redig, P., Strnad, M., Galis, I., van Dongen, W., van
Onckelen, H., 1995. Quantifying phytohormones in transformed
plants. In: Gartland, K.M.A., Davey, M.R. (Eds.), Agrobacterium
Protocols, Methods in Molecular Biology, vol. 44. Humana Press,
Clifton, UK, pp. 245–261.
Prinsen, E., Van Dongen, W., Esmans, E.L., Van Onckelen, H.A., 1998.
Micro and capillary liquid chromatography-tandem mass spectrome-
try: a new dimension in phytohormone research. J. Chromatogr. A
826, 25–37.
Redig, P., Schm u¨ lling, T., Van Onckelen, H., 1996. Analysis of cytokinin
metabolism in ipt transgenic tobacco by liquid chromatography-
tandem mass spectrometry. Plant Physiol. 112, 141–148.
S a´ enz, L., Jones, H., Oropeza, C., Vl a´ cˇ il, D., Strnad, M., 2003.
Endogenous isoprenoid and aromatic cytokinins in different plant
parts of Cocos nucifera (L.). Plant Growth Regul. 39, 205–215.
Va nˇ kov a´ , R., Gaudinov a´ , A., S u¨ ssenbekov a´ , H., Dobrev, P.I., Strnad, M.,
Hol ´ı k, J., Lenfeld, J., 1998. Comparison of oriented and random
antibody immobilization in immunoaffinity chromatography of cyt-
okinins. J. Chromatogr. A 811, 77–84.
Wang, J., Letham, D.S., Taverner, E., Badenoch-Jones, J., Hocart, C.H.,
1995. A procedure for quantification of cytokinins as free bases
involving scintillation proximity immunoassay. Physiol. Plant. 95, 91–
98.
Werner, T., Hanu sˇ , J., Holub, J., Schm u¨ lling, T., vanOnckelen, H.,
Strnad, M., 2003. New cytokinin metabolites in IPT transgenic
Arabidopsis thaliana plants. Physiol. Plant. 118, 127–137.
Za zˇ ´ı malov a´ , E., Kam ´ı nek, M., B rˇ ezinov a´ , A., Motyka, V., 1999. Control
of cytokinin biosynthesis and metabolism. In: Libbenga, K.R., Hall,
M.A., Hooykaas, P. (Eds.), Biochemistry and Molecular Biology of
Plant Hormones, Ser. New Comprehensive Biochemistry. Elsevier,
Amsterdam, pp. 141–160.
21–29.
Dobrev, P.I., Havl ´ı cˇ ek, L., V a´ gner, M., Malbeck, J., Kam ´ı nek, M., 2005.
Purification of plant hormones auxin and abscisic acid using solid
phase extraction and two-dimensional high-performance liquid chro-
matography. J. Chromatogr. A 1075, 159–166.
Horgan, R., Scott, I.M., 1987. Cytokinins. In: Rivier, L., Crozier, A.
(Eds.), The Principles and Practice of Plant Hormone Analysis.
Academic Press, London, pp. 304–365.
Jones, L.H., Mart ´ı nkov a´ , H., Strnad, M., Hanke, D.E., 1996. Occur-
rence of aromatic cytokinins in oil palm. J. Plant Growth Regul. 15,
39–49.
Jordi, W., Schapendonk, A., Davelaar, E., Stopen, G.M., Pot, C.S., De
Visser, R., VanRhijn, J.A., Gan, S., Amasino, R.M., 2000. Increased
cytokinin levels in transgenic PSAG12-IPT tobacco plants have large
direct and indirect effects on leaf senescence, photosynthesis and N
partitioning. Plant Cell Environ. 23, 279–289.
Laloue, M., Terrine, C., Gawer, M., 1974. Cytokinins: formation of the
0
nucleoside-5 -triphosphate in tobacco and Acer cells. FEBS Lett. 46,
45–50.
Lighfoot, D.A., McDaniel, K.L., Ellis, J.K., Hammerton, H., Nicander,
B., 1997. Methods of analysis of cytokinin content, metabolism, and