CLIN.CHEM.21/9, 1245-1252 (1975)
An Analytical System for Rapid Separation
of Tissue Nucleotides at Low Pressures on
Conventional Anion Exchangers
Joseph X. Khym
An analytical anion-exchange
procedure has been de-
veloped for the rapid separation of acid-soluble nucleotides (the so-caHed “free” or tissue nucleotides). It permits assay at low pressures (275-415 kPa; 40-60 psi)
in less than 1 h on 10-cm columns of Aminex resins
(conventional styrene-type anion exchangers) with alkaline citrate solutions as the eluent. Separation variables
have been investigated, to determine optimum conditions for the routine analysis of samples containing tissue nucleotides. Also described here is a simple solvent-extraction
procedure
for removing
HCIO4 or
CCI3CO2H quantitatively from cell extracts that contain
acid-soluble nucleotides: they are removed from aque-
ous acid solutions with a water-insoluble
solved in a water-immiscible
solvent.
amine dis-
AddItIonal Keyphrases: quantitative separation of acid-soluble nucleotides
#{149}
rapid chromatographic separation of tissue nucleotides at low pressures #{149}
analysis of “free” nucleotides #{149}removal of trichloroacetic or perchloric acids
by simple extraction
Various
available
liquid-chromatographic
procedures
are
for isolating
and assaying
acid-soluble
nucleotides (1). In some of the newer procedures,
the
nucleotides
are separated
on ion-exchange
celluloses
(2) and dextrans
(3) or by reversed-phase
systems (4,
5). However, because of the recent trend toward both
speed
and resolving
power in the development
of
chromatographic
systems (6-8),
and because of the
particular
demands for speed, sensitivity,
and quantitation in the routine determinations
of cellular nucleotide
pools,
pellicular
ion-exchange
(9, 10) systems
have been used more often than other chromatographic procedures. Thus, since 1967 when pellicular
exchangers
were first introduced
(9) conventional
resin anion exchangers have been used less and less
in chromatographic
separations
of tissue nucleotides
(1, 9, 10). However, in a recent preliminary
communication, Khym (11) demonstrated
that on the citrate
form of Aminex
A-27 (a conventional
styrene-type
anion exchanger,
particle
size 12-15 m)
with alkaBiology
Division,
Oak Ridge National
Laboratory,
Tenn. 37830.
Research
sponsored
by the U. S. Atomic Energy
under contract with Union Carbide Corp.
Received
Mar. 20, 1975; accepted April 8, 1975.
Oak Ridge,
line sodium
citrate
as the eluent, tissue
may be separated
in about 1 h on columns
nucleotides
only 10 cm
in height. The citrate system compared
very favorably with similar analyses carried out with pellicular
anion exchangers,
which have the disadvantage
of
lower capacity
than conventional
resins. Furthermore, because the A-27 columns are short, they can
be operated at lower pressures (275-415 kPa; 40-60
psi) than are necessary
for the pellicular
ion-exchange columns (up to 34.5 MPa, or 5000 psi). This
gives the added advantage
of being able to operate
with low-cost chromatographic
equipment
(e.g., glass
columns).
The present work details the influence of various
separation variables associated with the low-pressure
liquid chromatography
of tissue nucleotides
on the
citrate form of Aminex exchangers.
A rapid, simple
solvent-extraction
procedure is also described for the
quantitative
removal of residual precipitating
acid
from cell extracts
containing
acid-soluble
nucleotides.
Materials and Methods
Reagents
Sodium citrate (Na3C6H5O7.2H20),
citric acid, and
CC13CO2H were purchased
from the Mallinckrodt
Chemical Works; sodium azide and sodium hydroxide solution (1 mol/liter) from Fisher Scientific Co.;
71% HC1O4 solution from J. T. Baker Chemical Co.;
and Alamine 336 (tricaprylyl tertiary amine), a General Mills product, from McKerson Corp., Minneapolis, Minn. 55408. Freon-TF’
(CC12FCCIF2, bp 47 #{176}C)
was obtained from E. I. du Pont de Nemours and Co.,
Inc. All reagents
were of analytical
grade. “A-grade”
ribo- and deoxyribonucleoside
-triphosphates
as well as riboides and purine and pyrimidine
from Calbiochem.
5’-mono-,
-di-, and
and deoxyribonucleosbases were obtained
Aminex A-27 (bead diameter, -‘13 tim) and Aminex A-28 (bead diameter,
‘-‘9 m) styrene/divinylbenzene anion exchangers were bought from Bio-Rad
Laboratories.
Commission
1
Nonstandard
roethane;
abbreviations
and ADP-Rib,
adenosine
used:
Freon-TF,
trichlorotrifluo-
5’-diphosphoribose.
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
1245
Equipment
.
equipment.
The following
acsold by Chromatronix,
Inc. but
Chromatographic
cessories-originally
now obtainable
from Laboratory
era Beach, Fla. 33404-were
investigations:
jacketed
Data Control,
used throughout
glass columns
.
.
Rivi-
these
(0.63 X 33 cm),
adjustable
internal
septum-type
injecvarious
lengths
of Y16-in.
Table 1. Values for Molar Absorptivities,
Determined
at 254 nm for Nucleoside Phosphates
lfl Citrate Buffer (pH 8.2)a
.
Compound
CTP
tors (Model
164A1 1), and
(o.d.) Teflon
tubing
with Cheminert
connector
fittings. The off-column
injectors
were modified
by
blocking the original inlet port and boring a new port
about 2.5 cm up toward the septum holder. New sep.
turn holders
were fabricated
with needle guides of
UMP
to better
accommodate
#{128}354
AMP
CDP
interchangeable
top and bottom
bed support plungers, off-column
smaller diameter,
injection syringes.
Compound
#{128}354
CMP
6.0
UDP
7.9
13.0
ADP
13.1
ATP
12.8
GMP
12.4
GDP
12.7
GTP
UTP
#{176}Identical
values
for
254
were
obtained
with either
the 19- or 50-
M’ cells when placed in a ISCO type 6 optical unit. Determinations
were made on 0.02, 0.1. and 1.0 absorbance range scales.
Hamilton
Pumps
and gauges.
The flow rate of eluent
through a column and its component
parts was controlled with a Milton-Roy Minipump (0-1000 psi, 460
ml/h maximum flow rate). Gauges (Durogauge, 0-200
psi) with an AISI 316 tube and socket were obtained
from Dresser
Industrial
Valve, Stratford,
Conn.
eluent solutions contained
3 X 10
mol of sodium
azide per liter, to prevent bacterial growth in both
the citrate eluents and on the citrate form of the Aminex resins. Both the dilute (0.025 mol/liter) and the
concentrated
(0.5 mol/liter)
eluents were prepared
from 1 mol/liter sodium citrate solution, which was
06497.
Ultraviolet
detector.
column was monitored
filtered
analyzer
ity)
work (11), the
with a UA-2 ISCO ultraviolet
(single cell, 10-mm path length, 50-#{128}l
capac-
(Instrumentation
In the initial
Specialties
Company,
Lincoln,
Neb. 68505). In the present work, monitoring
was
done with a more sensitive UA-5 ISCO instrument
(dual cell, 10-mm path length, l9-il capacity). External recorders (e.g., Beckman 10-in, recorder) were attached to the ISCO’s when peak areas were measured
quantitatively,
Each ultraviolet
detector was calibrated
independently by determining
extinction
coefficients as follows: Single authentic
compounds,
dissolved
at a
known concentration
in citrate buffer of pH 8.2, were
allowed to flow directly through the measuring cell of
the ultraviolet analyzer. Because the optical cells had
a 10 mm pathlength,
the extinction coefficients were
calculated from the equation:
E254
=
concentration
absorbance
at 254 nm
of authentic
compound
(jimol/ml)
The results of one such calibration
are shown
in
Table 1. Measurements
on the UA-5 ISCO were
made in the absorbance
range 0.02-1.0; deviation
was not noted,
To prevent
bubble formation
in the measuring
cells of the optical units, about 3 m of small-bore (0.3
mm)
Teflon
tubing
(Chromatronix,
part
No.
T063012) was affixed to the exit side of the cell. Such
tubing adds about 100 kPa (15-20 psi) to the operating pressure
of the chromatographic
system.
.
Elution
Preparation
1246
CLINICAL
of sodium
CHEMISTRY,
citrate
eluents.
Vol. 21, No. 9,
1975
a 0.45-tim
Millipore
membrane
after
protected
solutions
from
atmospheric
will decrease
Gradient
apparatus.
C02,
the
pH of citrate
markedly.
The
eluent,
generated
as a
convex gradient (25 to 500 mmol of citrate per liter,
pH 8.2), was delivered to the column by the apparatus shown in Figure 1. The advantage of this design is
that, at the completion of a run, the column is easily
equilibrated
with dilute buffer in anticipation
of the
next run by means of a three-way stopcock. Furthermore, as seen in Figure 1, this design allows the apparatus to be detached and cleaned during the 15- to
20-mm equilibration.
The gradient employed does not impair the reproducibility of the analyses or cause the resin bed to
swell or shrink noticeably. Depending on the amount
and nature of sample impurities,
about 50 consecutive determinations
can be performed on a single column, after which usually only the first few millimeters have to be replaced.
(1)
from linearity
through
its preparation.
pH was adjusted
by adding small
amounts of either citric acid (1 mol/liter) or NaOH (1
mol/liter), or both, to prepared citrate eluents. If not
All citrate
Column Preparation
Before packing in a column, 3-5 ml of Aminex
A-27 or A-28 was consecutively
treated with the following solutions: one part of concentrated
HC1 plus
two parts of ethanol
(20 mm); 1 mol/liter
HC1 (4 h); 1
mol/liter NaOH (1 h); 1 mol/liter citric acid (until the
filtrate was acid); 0.5 mol/liter sodium citrate, pH 8.3
(until
the filtrate
ins and
the
was free of Cli.
various
wash
Slurries
solutions
of the res-
were
filtered
through
15-ml sintered-glass
funnels (ASTM, “mediurn” porosity
Pyrex No. 36060). After the last treatment, a concentrated
slurry of resin and sodium citrate (pH 8.2, 25 mmol/liter),
consisting
of 1 part
resin and about 0.5 part citrate
eluent,
was prepared.
GAANT
water-insoluble
organic solvent (Freon-TF).
The extractions are done as follows. Equal volumes of tissue-extract
samples
and amine/Freon
solutions
(about 0.5 mol of Alamine 336 per liter) are mixed
gently
for 3-4 mm in a screw-cap
container.
The
phases may be separated
in a few minutes by lowspeed centrifugation.
Extraction
of aqueous HC1O4
SMbp
solutions
yields
ly separated
two clear
organic
phases,
from a single aqueous
both
clean-
layer; extraction
of aqueous
CC13CO2H solutions2
yields clear phases
for both the organic
and aqueous
layers. In either
case nucleotides,
nucleosides,
and purine and pyrimi-
dine bases remain quantitatively
in the aqueous (top)
phase, which now has a pH of 4-5. Aliquots
of the
BAR
Na-citrate, pH 8.2
‘To 0.025 not/I
3xtO4
not/I
NON3
RESERVOIR
Fig. 1. Gradient apparatus for 10-cm Aminex column
aqueous
phase may be place.d directly on the Aminex
columns
from the aqueous
phase or first be concentrated
by vacuum
distillation.
Citrate
eluent
at pH
8.2 is usually added to extracted
samples
so that the
final citrate concentration
is 25 mmol/liter.
Fresh preparations
of Alamine/Freon
solutions
are
clear and colorless and impart negligible
amounts
of
ultraviolet-absorbing
material
to aqueous
phases
during
extractions.
Alamine
may
the preparations
To operate at lower pressures,
the middle disk
(0.006 in. mean pore diameter)
contained in the adjustable bed support plungers is removed, and the
two outer disks (0.034 in.) are put back in place. With
the column at room temperature
and the bottom suport plunger in place, the concentrated
resin slurry is
added to a column with a disposable pipet. After inserting
the top plunger, dilute citrate eluent is
pumped through the column until most of the resin
has settled. The pumping is stopped, and more resin
is added to the column until the bed height remains
stationary
at 10 cm. At this point the column temperature
is brought
to 70 #{176}C
by means
of a Haake
Model
F constant-temperature
circulating-water
bath, and 0.5 mol/liter
citrate
eluent
is pumped
through the column for about 2 h. (The pressure decreases noticeably when the column is operated at 70
#{176}C).
After equilibrating,
with dilute citrate eluent,
the column is ready for use.
After packing, a new column often shows improved
resolution after a few column runs have been made.
In some instances,
adjacent pairs such as UTP and
GDP will not separate
at all initially, but will do so in
subsequent
determinations.
Removal
of Perchloric
or Trichioroacetic
Acids
Acid-soluble nucleotides
are usually
extracted
from tissues with either CC13CO2H (about 0.3 mol/
liter) or HC1O4 (0.3 to 0.5 mol/liter);
because these
acids interfere with most subsequent
chromatographic assays,
they
customarily
are removed
from samples
before analysis.
The procedure
used in these studies
involves neutralizing
the acid solutions
with a water-insoluble
amine
(Alamine
336, mol wt 392) contained
in a
Upon
standing
crystallize
from
for several
the Freon
days,
solvent
and
may turn yellow.
Column Operation
The columns
of 0.6 ml/min
these
pressure
(40-50
pumped
are routinely
operated
and at a temperature
conditions,
Aminex
at a flow rate
of 70 #{176}C.
Under
A-27 columns
develop
a
drop across the resin bed of 275-345
kPa
psi) when
dilute
citrate
eluent
is being
through
the system. 4An increase of about
140 kPa (20 psi) occurs as the concentration
of citrate
eluent approaches the limit concentration
of 0.5 mol/
1. Under the same conditions of flow rate and temper-
ature,
Aminex
A-28 columns
operate
at pressures
140-170 kPa (20-25 psi) higher than those of equivalent columns of A-27. As noted in the legend informa-
tion, operating pressures vary considerably
with either flow rate or temperature.
Small sample volumes (10-200 gil) were injected
with Hamilton syringes without interrupting
the flow
of eluent. Larger amounts (up to 1.0 ml) were applied
to the column with the pumps momentarily
off.
Identification
and Quantitation
In these studies identifications
were made solely
on the basis of the retention time (or volume to peak)
of a given
compound.
Elution
positions
were
estab-
lished with authentic
compounds
(either ribo- or
deoxyribonucleoside
5’-mono-, -di-, or -triphosphates
or related compounds)
under a specific set of column
operating conditions. The reference compounds
(1 to
50 nmol), dissolved in citrate eluent (25 mmol/liter),
were applied to the column singly or in various simple mixtures until elution positions were established
beyond any doubt. For example, once the elution
2
The
ultraviolet-absorbing
with CCI3CO2H
material
is also quantitatively
CLINICAL
CHEMISTRY,
arising
removed
from
or associated
by the extraction.
Vol. 21, No. 9, 1975
1247
order of, say, the nucleoside monophosphates
was established,
two nucleoside diphosphates
were added
to a mixture of the monophosphates
and the analyses
were repeated. If there was any doubt about elution
position, concentrations
of the compounds
in ques-
tion were changed in the mixtures
under
study.
Peaks were quantitated
by peak-height
and peakwidth measurement
(8, 12) according to the equation:
nmol
=
HA#{149}W#{149}lO64
(2)
254
where HA is the net peak height (in absorbancy
units) measured at 254 nm, Wmi is the peak width (in
volume units) measured
at half peak height, it2M is
the extinction coefficient of a particular compound as
previously defined, and 1064 is an equation constant.
Results and Discussion
Removal of Acid from Tissue Extracts
Even
used
though
to extract
CCI3CO2H
and
nucleotides
from
HC1O4 are widely
tissues,
there
the choice depending
upon the origin of the kind of
tissue being analyzed
(16-18).
Nonetheless,
by the
Alamine/Freon
procedure
given here, acid-soluble
nucleotide material, once extracted from tissue fragments, quantitatively
remains in the aqueous phase,
while both CC13CO2H and HC1O4 (and, presumably,
any other strong acid) are partitioned
quantitatively
to the organic phase. This was observed by two different
sets of experiments.
In one, separate
mixtures
of either ribonucleoside
5’-mono-, or -di-, or -triphosphates were dissolved in cold CC13CO2H (0.3 mol/
liter) or HC1O4 (0.4 mol/liter)
and shaken with an
volume
of 0.5 mol/liter
Alamine
in Freon
sol-
vent. In each case, upon separation of the phases, the
aqueous layer had a pH of 4-5 and contained
all of
the added nucleotides,
as ascertained
spectrophotometrically.
that losses
In the other experiment,
it was observed
were slight, if they occurred
at all, when
mixtures of nucleoside phosphates
in either acid were
extracted
with Alamine/Freon
solvent, concentrated
to dryness in a micro rotary evaporator, and then dissolved in 15 mmol/liter
citrate eluent and separated
on the citrate form of Aminex A-27. A typical experiment of this type is described
in Figure 2.
Removal
of acids with amines
dissolved
in water-
immiscible
solvents
is not new; the system was first
used by Smith and Page (19) to remove mineral acids
from protein hydrolysates.
Although simple and convenient
(as compared
with removal of the C104
anion as its potassium salt or of CCI3CO2H by repeated extractions
with ether), this technique
has not
been used routinely to remove acid extractants
from
tissue preparations.
The choice of amine and (or) sol1248
CUNICAL CHEMISTRY. Vol. 21, No. 9, 1975
0
20
40
Tt
tonI
Fig. 2. Recovery of ribonucleoside
phosphates after removal
of 0.3 mol/Iiter CCI3CO2Hwith Alamine/Freon soluhon
A 50-ILl aliquot of a mixture of rlbonucleoside phosphates, each at a concentration of about 0.1 ILmol/mI, was assayed
on a 10-cm A-27 column (Panel
A) at 70 #{176}C
and a flow rate of 0.6 mI/mm. 1 ml of this same rlbonucleoside
phosphate mixture was diluted to 10 ml with cold 0.3 mol/liter CCbCO2H and
the sample extracted with an equal volume of Alamine/Freon solution. From
the resulting aqueous phase, a 5-mI allquot was concentrated to dryness and
the residue dissolved In 0.5 ml of 25 mmol/llter citrate eluent. A 50-M1 aliquot
was assayed (Panel B) on the same column and under the same conditions
used In Panel A. In each case, chart speed was 15 cm/h and full-scale deflection was 0.1 absorbancy units
has
been no agreement as to the best method for quantitative extraction
(13-15).
Not only cold acids, but
other solvents (13-15) are often used for extractions,
equal
0
vent is arbitrary
(e.g., see references
17, 19, 20) as
long as the chosen solvent system carries out the intended task. The choices of reagents
here were based
on the facts that Alamine 336 is obtainable
in a very
pure state, is very insoluble in water, and does not
yield ultraviolet-absorbing
impurities,
and that
Freon-TF
and water are virtually insoluble in each
other (21).
Analytical Variables Affecting
Separations on the Column
Nucleotide
Effect of temperature.
The optimum temperature
for separating
nucleoside phosphates
on the citrate
form of A-27 (or A-28) at pH 8.2 is 70 #{176}C
(Figure 3).
At temperatures
in the range of 50-60 #{176}C
AMP/CDP,
GMP/UDP,
and ATP/GTP
pairs do not separate,
while at 75 #{176}C
AMP/CDP
separate cleanly but UTP
tends to overlap GDP. AMP and CDP may be completely separated at 70 #{176}C
by having a gradient delay
of 2-3 mm-that
is, after sample injection, dilute citrate eluent is pumped momentarily
a few minuteslonger before the gradient is started (or, alternatively, the exit tube from the gradient device to the column is made correspondingly
longer).
Because 70 #{176}C
is the temperature
used, it is essential to determine the stability of the nucleoside phosphates at this temperature.
The results of this study
are shown in Figure 4. Sodium azde was not added to
stock preparations
of nucleoside phosphates,
nor was
it present during the 2-h heating period as described
in Figure 4. Sodium azide solution, even at a concentration of 0.3 mmol/liter,
degrades nucleoside phosphates slowly over a period of several days. Thus,
even though it was present in all citrate eluents, it
0.2
MW’
NJCLEOTIDE
MXTIJ1E 64 WATER BATh
AT 7l’C FOR 2 IPE. lIEN
ON COLIAWI
75.C
JJfrI!WIAWGW
70
0.1
cM’
LEt’
A
CW
‘C
POJCLEOTCE MXTLWE
P
UTP
A
AOR
0.1
AlP
40
iea
GTP
0.l
A mixture of ribonucleoside phosphates, dissolved in 25 mmol/llter citrate eluent, was analyzed directly on a 10-cm A-27 column and again after heating
the mixture at 71 #{176}C
for 2 h. Identical aliquots (75 ILl) were Injected into the
column (operated at 70 #{176}C
and a flow rate of 0.6 mI/mm)
ATP
p
-
0
4
40
I.0
so
FLOW RATE’O.6W,/,n,n
OPERAING
0
PRESSURES’ 70-95
TiME (minI
0.4
0.2
phosphates
dissolved
a,
Si
in
psI
II4LAA
Oi034OO
buffers.
Effect
of pH. The optimum
pH range for separating nucleoside phosphates
on the citrate form of either A-27 or A-28 is between 8.0 and 8.3 (these are
room temperature
values; the actual pH values of citrate eluent at 70 #{176}C
were not determined).
Because
of slight differences
in different
batches of resins,
FL0WRATEO.3mI/,n.,
08
adjacent nucleoside phosphate peaks.
Between pH 8.0 and 8.3, the retention time for undine and guanosine
nucleoside
phosphate
increases
markedly (and more noticeably at pH’s higher than
8.3) as the pH is increased.
This is because of increased ionization
of the acidic groups in the base
moieties of the uridine and guanosine phosphate
derivatives.
Thus, if certain adjacent
pairs such as
UDP/CTP
or UTP/GDP
are not completely resolved,
an appropriate
increase or decrease in pH should increase the resolution. Adenosine and cytidine nucleoside phosphates
do not have acidic groups or their
base moieties that ionize in the range between pH’s
8.0 to 8.3, hence elution positions remain the same
for these derivatives with small changes in pH of the
citrate eluent. Thus the resolution between AMP and
CDP cannot be improved by pH changes, but is remedied by controlling
citrate concentrations
as mentioned previously.
Since
1967, speed
have been the main incentives
and efficiency
for using pellicular
ex-
GM’
r
0.6
OPERATiNG
PRESSURES’
ir
A
35-50
GDP
A
psI
ATP
0.4
JL IIfiIJAJLJTP
0.2
small adjustments
in pH values within the generally
optimal
range sometimes
improve
the resolution
of
of analysis.
1,0 ,,,I/onn
PRESSURES. ISO-ITO
0.6
The same mixture of ribonucleoside phosphates,
each at a concentration
of
about 0.1 ILmoI/ml, was assayed on a 10-cm A-27 column at a flow rate of
0.6 mI/mm and at the temperaturesshown. In each case a 5O-ILlaliquot was
used
to nucleoside
FLOW RATE’
OPERATING
-
0.8
Fig. 3. Effect of temperature on the resolution and retention
time of nucleoside phosphate peaks
was not added
80
low)
Fig. 4. Stability of nucleoside phosphates at 70 #{176}C
in citrate
buffer (pH 8.2)
50 ‘C
Speed
ON COLI.NN
7/pj
Lap
I
citrate
DRECTLY
GM’
ATP GTP
ras
(oW,)
Fig. 5. The same mixture of ribonucleoside phosphates, each
a
at a concentration of about 1.0 izmol/ml,
was assayed on
10-cm A-28 column at the flow rates and pressures shown
A 30-IL1 allquot was used in this case and the chart speed of the external recorder was 0.1 in/mm and full scale deflection was 1.0 absorbancy units
changers
rather
of exchangers
for
However, as shown
in Figure 5, with analytical resolution (8) such analyses can be completed
in about 40 mm on columns of
conventional
resin exchangers that are only 10 cm in
tissue
nucleotide
height.
Also,
than
assay
as seen
other
types
(8-10,
22).
in Figure
5, there
is complete
resolution of adjacent peaks when the flow rate is decreased from 1 to 0.3 ml/min; at this slower flow rate
an analysis takes about 2.5 h to complete, but there is
the advantage
of low operating pressures (only 240340 kPa, or 35-50 psi).
If greater speeds are important in chromatographic
assays of tissue nucleotide
samples,
5-cm columns
of
A-27 or A-28 may be used to carry out nucleotide
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
1249
AMP
Table 2. Recovery of AMP and ATP from Mixtures
Analyzed on a 10-cm Column of Aminex
A-27 Citrate
Added
Found#{176}
Added
Foundb
5.2
6.4
5.2
6.3
ATP
nmol
Compound
52
64
AMP
ATP
a Determination made
bOeterminat ion made
51
62
with use of 1.0 absorbance
with use of 0.1 scale.
z
scale.
C
S
separations in about 25 mm. On the smaller columns,
resolutions
are still sufficient, but there is considerable overlapping of some adjacent peaks.
Quantitative
04
AlP
02
Measurements
70
Samples were examined routinely on A-27 or A-28
columns at a flow rate of 0.6 ml/min and a temperature of 70 #{176}C.
In these determinations,
external recorders
attached
to the
column-monitoring
systems
were operated at 12 in./h. To test for quantitative
recovery under these conditions,
AMP/ATP
mixtures
of known concentration
were examined. The results
of these assays (Table 2) were calculated by means of
equation 2. Peak-height
and peak-width
were measured from chromatograms
typified by the one shown
in Figure 6, which also shows the retention times for
all 12 common nucleoside phosphates
run under the
same conditions
as the AMP/ATP
mixtures.
Practical
this method
of analyzing
tides is being used to evaluate
extracting
“free”
nucleotides
preparations.3 Table
tissue
different
from
3 shows
nucleo-
methods
Escherichia
AMP/ATP mixtures of known concentrations
were assayed on a 10-cm A-27
column under conditions given in the text. Data from this chromatogram
were
used in equation 2 to calculate the recoverles that are shown in Table 2
0.50
chemicals
for
C”
uOP
re-
of practical interest
for purity. An example
analysis
is testing
of this is
of four commercial
COP
sponding
monophosphate
from 17 to 22%.
in a proportion
GOP
ranging
0
0
10
20
30
TIME
40
50
60
(minI
Fig. 7. Analysis of a ribonucleoside diphosphate preparation
Chromatography
of Compounds
Ribonucleoside Phosphates
Solid COP, UDP, ADP, and Q)P were dissolved In 25 mmol/liter citrate eluent
at a concentration
of about 1 mmol/llter. A 15-IL’ aliquot ylelded the chromatogram
shown when the mixture was assayed on a 10-cm A-27 column
(operated
at 70 #{176}C
and at a flow rate of 0.6 mI/mIn). Chart speed was 0.1 In.i
mm
Related to
Under the column operating conditions described
here, NAD and NMN elute before UMP and, if present in concentrations
about equal to that of UMP,
tend to overlap this peak. NADP elutes in the same
position as AMP. When assayed singly, NADH gives
with a small
0.25
ribonu-
cleoside diphosphate
preparations.
Purposely, preparations several years old were chosen for this test.
The diphosphates
were dissolved in dilute citrate eluent, and the resulting mixture was assayed on an
Aminex A-27 column. The chromatogram
(Figure 7)
showed that each diphosphate
contained
its corre-
a chromatogram
S
-
‘‘...
the simultaneous
breakthrough
peak
These experiments
carried out with the collaboration
Schenley
of the Biophysics
and Cell Physiology
Section
Ridge National Laboratory.
1250
Fig. 6. Elution positions of AMP and ATP in simple and complex mixtures: Quantitative determination
of AMP/ATP
mixtures
coli
some preliminary
Another application
reagent
80
(onnI
Applications
At present,
sults.
TRE
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
and
of R. L.
at Oak
another small peak with a retention time greater than
GTP. Other similar compounds
such as ADP-Rib,
FAD, etc., have not been investigated.
Mixtures of deoxyribonucleoside
phosphates
yield
chromatograms
that are similar to those obtained for
the ribo-denivatives.
However, as seen in Figure 8,
dAMP and dCDP as well as dCTP and dTDP do not
separate, and dADP and dTTP separate only partially. Other sets of conditions are being investigated
for
Table 3. “Free”
Nucleotides in Escherichia
co/i Ceilsa after Various Treatments
Solution
Treatmentb
UMP
AMP
CDP
GMP
Hot H20
Sonicated
UV-irradiated,
son icated
Hot H20
Hot H20
UV H20
3
3
1
16
8
3
4
6
2
5
5
6
3
2
2
14
12
7
3
T
1
6
10
4
extractedC
ADP
GDP
ATP
GTP
4
4
T
6
4
1
Td
4
1
T
3
2
1
2
T
2
2
T
4
5
4
3
1
T
4
T
T
2
T
T
UDP
MmoI/liter
a About 1O cells/mi of extractant.
b Cold CCI3CO2H was added to samples after
C CMP, CTP, UTP not measured.
dT = trace.
treatment.
None was added to the experiment
0.2
described
by the next-to-last
row of data.
04
I
101
JjjJjaawy
0:
o
20
60
80
Fig. 9. Separation of 5’-mononucleotides
Tve (.nI
under isocratic elu-
Fig. 8. Separation of deoxyribonucleoside phosphates
tion conditions
deoxyderivatives
were dissolved at a concentration
of about 0.1
mmol/llter in 25 mmol/llter citrate eluent. A SO-M’allquot was analyzed on a
10-cm column of A-27 (operated at 70 #{176}C
and a flow rate of about 0.64 ml/
mm)
To determine plate numbers (see Table 4) a mixture of these mononucleotides was separated on a 10-cm A-27 column at a constant citrate eluent
concentration
(25 mmol/liter, pH 8.2). The column was operated at 70 #{176}C
and a flow of about 0.6 mI/mm. Sorbed material was about 30 mmol each of
CMP, UMP, AMP, and GMP, contained in 30 MIof the eluent. Chart speed was
0.5 in/mm; full-scale deflection was 0.5 absorbance
unit
These
the chromatography
of deoxyribonucleoside
phosphates. This work, as well as the separation of bases,
nucleosides,
and the 2’- and 3’-mononucleoside
phosphates on the citrate form of Aminex exchangers, will
be reported elsewhere.
Efficiency of Separation and Comparisons
Pellicular Ion-Exchange Systems
(T)
=
=
L/N
tR
is the retention
time
when gradient
elution
of N and HETP
()
where ‘mi is the retention
time of a peak expressed
in volume units and Wmi is the same quantity given
for equation 2. Because N is proportional
to column
length (L), a more useful measure of column efficiency may be calculated from the height equivalent to a
theoretical plate (HETP), which is defined by:
I]ETP
where
However,
to
Column efficiency is expressed quantitatively
in
terms of the number of theoretical
plates (N) found
for a column (see references 8, 9, 10, or 12). N may be
calculated by (9, 23):
N
(5)
(4)
Another measure of efficiency is the plates per second (N/s) encountered
by a peak as it passes through
a column; this is obtained from (9, 24, 25):
by equations
for a peak,
in seconds.
is used, calculations
3 and 4, respectively,
are not valid, because the mass distribution
coefficient of a solute will not remain constant throughout
a column when the eluate is changing concentration.
Thus, to obtain some idea of efficiency values for nucleotide separations
with the citrate/Aminex
procedure, the four common 5’-ribonucleoside
monophosphates were chromatographed
at constant eluate concentration
(Figure 9). Accordingly,
efficiency values
were calculated for the separation shown in Figure 9
by the use of equations 3 to 5; these values, given in
Table 4, compare favorably, and in most instances
are better than similar values found for pellicular
ion-exchange
systems (9, 10, 24, 25). Efficiency can
be improved further by using A-28 rather than A-27
columns. This has been judged qualitatively
by noting that, when compared under identical conditions,
nucleoside
phosphate
peaks have narrower
band
widths on A-28 columns.
In some instances, pellicular exchangers
have an
advantage over conventional
resins in tissue nucleotide work. One advantage is that separations
can be
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
1251
Table 4. Efficiency
Values for the Separation
of 5’-Ribonucleotides on a 10-cm Column
of Aminex A-27 (Citrate Form) with Citrate
Eluent (pH 8.2, 25 mmol/liter)
Compound
CMP
UMP
AMP
GMP
Plate no.
510
851
760
1350
Plate
height
(mm)
0.20
0.12
0.13
0.07
Plates/s
13.1
11.6
6.1
5.8
at ambient
temperatures
with the newer
pellicular
ion-exchangers
(22). As Figure 2 shows,
this is not important
for the common ribonucleoside
phosphates,
but for more labile compounds
(e.g.,
NAD) hydrolytic
decompositions
would be of concern. Another advantage
of the pellicular analytical
performed
system
is that
very sensitive
detectors
(e.g., less than
0.02 absorbancy
units full scale) can be used without
serious baseline noise and drift (9). In the present
work, considerable
baseline drift occurred on the 0.02
absorbance
scale (presumably
because of the steep
gradient used), especially toward the very end of a
chromatogram.
Nevertheless,
because the capacities
of conventional
resins are several hundred fold greater than those of the pellicular resins, larger sample
volumes may be injected into the A-27 or A-28 columns used here. Thus, less-sensitive
absorbancy
ranges are not always needed.
References
versed-phase
(1974).
Biochem.
system.
Biophys.
Res. Commun.
6. Done, J. N., Kennedy,
uid chromatography.
G. J., and Knox, J. H., Revolution
Nature 237, 77 (1972).
7. Taggart,
W. P., The key words
Res. 16, 76 (1974).
of HPLC
58, 605
in liq-
are high performance.
md.
J. X., Analytical
Ion-Exchange
Procedures in Chemistry and Biology. Prentice-Hall,
Inc., Englewood
Cliffs, N. J., 1974.
8. Khym,
9. Horvath,
C., High-performance
ion-exchange
chromatography
with narrow-bore
columns: Rapid analysis of nucleic acid constituents at the subnanomole
level. In Methods of Biochemical
Analysis, 21, D. Glick, Ed., Interscience,
New York, N. V., 1973, p 79.
10. Brown, P. R., High-Pressure
Liquid Chromatography.
WileyInterscience,
New York, N. Y., 1973.
11. Khym, J. X., Rapid separation
of tissue nucleotides
on conventional anion-exchange
resins at low pressures.
J. Chromatogr. 97,
277 (1974).
12. Singhal,
R. P., and Cohn, W. E., Analytical
separation
of nucleosides by anion-exchange
chromatography:
Influence of pH, solvents, temperature,
concentration,
and flow rate. Anal. Biochem.
45, 585 (1972).
13. Sadkkonen,
Chromatographic
J. J.,
Chromatography
of free nucleotides.
In
6, M., Lederer, Ed., Elsevier,
Amster-
Reviews
dam, 1964, pp 55-58.
14. Mandel, P., Free nucleotides
Res. Mol. Biol. 3, 304 (1964).
in animal
tissues.
15. Gray, H. J., Nucleotides
nucleotide
metabolism.
and
Prog. Nucleic
Methods
Cancer Res. 3, 243 (1967).
16. Bagnara,
A. S., and Finch, L. R., Quantitative
estimation
of intracellular
nucleoside
triphosphates
coli. Anal. Biochem. 45,24 (1972).
17. Macnicol, P. K., Analysis of adenine nucleotides
intermediates
in mature and senescent
leaf tissue.
45,624 (1972).
extraction
and
of Escherichia
and metabolic
Anal. Biochem.
1$, Dhople, A. M., and Hanks, J. H., Quantitative
extraction
of
adenosine
triphosphate
from cultivable
and host-grown
microbes:
Calculation
of adenosine
triphosphate
pools. Appi. Microbiol. 26,
399 (1973).
19. Smith,
long-chain
(1948).
L. E., and Page, J. E., The acid-binding
properties
of
aliphatic
amines.
J. Chem. Soc. Chem. md. 67, 48
1. Rajcsanyi, P. M., Csillag, M., and Kriskovics,
E., Separation
of
nucleic acid constituents
by liquid chromatography.
Sep. Purif.
Methods 3, 167 (1974).
20. Warner, A. H., and Finamore,
during brine shrimp
embryogenesis.
(1967).
2. Rustum,
Y. M., and Schwartz,
H. S., Methods
for the separation
and identification
of ribonucleotides
on DEAE-cellulose.
Anal.
Biochem. 53, 411 (1973).
21. Khym, J. X., Partition
of transfer
ribonucleic
acid between
aqueous
salt solutions
and trichlorotrifluoroethane
containing
a
quaternary
amine. J. Biol. Chem. 241, 4529 (1966).
22. Henry, R. A., Schmit, J. A., and Williams, R. C., Application
of
a new chemically
bonded,
superficially
porous
anion-exchange
packing to nucleotide
analysis. J. Chromatogr.
11, 358 (1973).
3. Hunger, H. D., and Reinbothe,
H., S#{227}ulenchromatographie von
5’-Nukleotiden
am Anionenaustauscher
Sephadex
A-25 mit einem
Lineargradienten
aus Ammoniumbicarbonat
und Ammoniumcarbaxnat. J. Chromatogr. 97, 273 (1974).
4. Singhal, R., Anion-exchange
chromatography
on reversed-phase
columns: Isolation and assay of nucleosides,
nucleotides,
and oligonucleotides
from nucleic acids and from cytoplasm.
Biochim.
Biophys. Acta 319, 11 (1973).
5. Holton, R. A., Spatz, D. M., van Tamelen,
E. E., and Wierenga,
W., The rapid separation
and isolation of mono- and oligonucleotides by high speed liquid chromatography:
An ion exchange,
re-
1252 CLINICALCHEMISTRY,Vol. 21, No. 9, 1975
F. J., Nucleotide
metabolism
J. Biol. Chem.
242, 1933
23. De Leenheer, A. P., Rigo, S. S., Gelijens, C. F., and
Vaerenberg,
P. M., Column chromatography
of 5-fluorouradil
Sephadex
G-10. J. Chromatogr.
92, 339 (1974).
Van
on
24. Kirkland, J. J., Columns
for analytical
liquid chromatography.
Anal. Chem. 43, 37A (1971).
25. HalSsz, I., Liquid chromatography-an
overview.
In Modern
Practice of Liquid Chromatography,
J. J., Kirkland,
Ed., WileyInterscience,
New York, N. V., 1971, p 325.