Hillel K. Brandes and David S. Bell
Supelco, Liquid Separations, Bellefonte PA 16823
T403163
GIH
Abstract
Nucleotides are ubiquitous in cells, with major
functions including the control of cellular energetics,
cell signaling cascades, biosynthesis of nucleic acids,
and involvement in countless enzymatic reactions. It is
therefore of widespread interest to resolve and identify
components of various such mixtures.
Traditionally, reversed-phase (RP) separations of
nucleotides have relied on ion-pair reagents, typically
salts of tetrabutylammonium. We sought to determine
whether viable alternatives of RP nucleotide
separations existed which would not be dependent on
quaternary ammonium ion-pair (IP) reagents. However,
this proved unsatisfactory; herein we demonstrate
optimal RP nucleotide separations by use of alternative
ion-pairing reagents on a silica-based alkyl phase.
2
Introduction
• RP nucleotide separations typically are performed on alkyl phases
and rely on quaternary amine IP reagents to provide adequate
retention.
• Quaternary amine IP reagents present problems for MS detection.
• IP reagents can present problems of reproducibility, particularly
where gradient elution is performed
• We sought to explore alternate stationary and mobile phases that
might circumvent the need for quaternary amine IP reagents, or
preferably any IP reagent.
• Regarding alternatives, zirconia-based stationary phases were of
particular interest given the fact that phosphates behave as strong
Lewis bases for coordination with the Zr surface, thus providing a
means for the nucleotide analytes to be retained in the absence of
any mobile phase modifier (IP reagent).
3
Initial Strategies
In our effort to find alternate RP separations of nucleotides
which did not rely on traditional use of quaternary amine ionpair reagents the following courses were initially pursued:
1. Perform the separation at pH 2 in which the negative charge on the
phosphate moieties would be reduced to ≤ ~0.5. TFA was included to
ion-pair with any protonated amines of the purine/pyrimidine bases.
Samples to be run on various phases (alkyl, polar-embedded, and
fluorinated).
2. Perform the separation at pH 1 on a zirconia column, at which
phosphate moieties would have a net negative charge of only ~ 0.1.
3. Perform the separation at neutral pH (to enhance ion-pairing with
phosphate moieties) on the various phases (alkyl, polar-embedded,
and fluorinated) with triethylamine, a weak basic IP reagent. This
might still permit MS detection in a negative-ion mode.
Initial evaluations were with a sample mixture of AMP, ADP, and ATP.
4
Structures of Initial Analytes
NH2
NH2
N
N
N
N
O
O
P
HO
HO
HO
HO
O
HO
O
O
P
P
O
HO
O
O
HO
NH2
N
N
N
N
HO
HO
O
O
O
P
P
P
O
HO
O
HO
O
O
HO
ATP
OH
N
N
OH
AMP
5
N
N
ADP
OH
Results of Initial Strategies
Summaries from the initial strategies:
1. The TFA pH 2 mobile phase system provided inadequate retention
and/or resolution on alkyl, polar-embedded, or fluorinated phases
(data not shown).
a. Other perfluoro organic acids might still provide adequate
retention while permitting mass-spectral detection.
b. Inclusion of phases with polar functionalities was to observe
if polar interactions with analytes might contribute to unusual
selectivity. The fluorinated phase was not expected to perform
favorably considering its electronegative surface and the negative
charge on the phosphate moieties.
2. Sample components appeared well retained on the zirconia
support, but eluted in an unsatisfactory broad peak (see Figure A).
3. Triethylamine provided inadequate retention and/or resolution of
the nucleotide mixtures (including the guanidyl-, cytodyl- and
uridyl- sample mixtures of mono-, di- and triphosphates) on alkyl,
polar-embedded, or fluorinated phases; data not shown.
6
Figure A. Chromatogram of Nucleotide
Mixture on Zirconia-PDB Column
0
mAU 260nm
2
4
Column: Discovery Zr-PBD, 7.5cm x 2.1mm ID, 3µm
Mobile Phase: 80:20, (25mM H3PO4/HCl, pH 1.0) : MeOH
Flow: 0.21mL/min
Temp.: 50ºC
Det.: 260nm
Inj.: 2µL, 0.2µg each analyte
Sample: AMP, ADP, & ATP
0
7
10
Time (min)
20
Optimization With Ion-Pair Reagents
• None of the previous separation systems (no IP utilized)
yielded satisfactory results.
• Tetraethylammonium (TEA) was initially studied as the
IP reagent. With sample mixtures of adenyl-, guanidyl-,
cytodyl- and uradyl- nucleotide mono-, di- and triphosphates, even at concentrations of ~15mM TEA, the
volume fraction of methanol had to be as low as 1- 2%
for good retention and resolution. This is not desirable
for standard C18 phases as it could lead to problems
with reproducibility due to inadequate “wetting” of the
stationary phase.
8
• Tetrapropylammonium (TPA) was therefore studied as
the IP reagent to see if it would provide adequate
retention for the nucleotide analytes.
Structures of Additional Analytes
Purine / Pyrimidine Bases
O
O
NH2
NH2
N
N
N
HN
N
H2N
N
N
HN
N
N
O
N
O
cytidine (C)
uridine (U)
guanosine (G)
adenosine (A)
N
Nucleotides
O
O
P
HO
HO
1-3
B
B
O
HO
O
HO
OH
O
O
O
P
HO
O
P
O
OH
OH
O
2'-3'-cNMP
9
B
O
3'-5'-cNMP
Tetrapropylammonium (TPA)
Mobile Phase System Methods
Two mobile phase buffers (pH 7.0) were explored:
1. phosphate because is it so commonly used, and
2. Bis-Tris, because phosphate at neutral pH is aggressive toward
silica, and may present column lifetime issues.
Phosphate Mobile Phase System
Mobile Phase A: 10mM H3PO4, 5mM TPAOH / NH4OH, pH 7.0
Mobile Phase B: 75:25, (13.3mM H3PO4, 6.7mM TPAOH / NH4OH, pH 7.0) : MeOH
Bis-Tris Mobile Phase System
Mobile Phase A: 10mM Bis-Tris, 5mM TPAOH / HCO2H, pH 7.0
Mobile Phase B: 75:25, (13.3mM Bis-Tris, 6.7mM TPAOH / HCO2H, pH 7.0) : MeOH
10
For both mobile phase systems:
Various proportions of mobile phases A & B (isocratic)
Column: Discovery HS C18, 15cm x 2.1mm ID, 5µm
Flow: 0.21mL/min
Injection: 2µL
Temperature: 35°C
Sample: 0.1g/L each analyte
Detection: 260nm
Phosphate Mobile Phase System:
Results Summary
Adenosyl Nucleotides
Guanidyl Nucleotides
1.0
1.0
3'-5'-cAMP
3'-5'-cGMP
0.8
2'-3'-cAMP
0.8
ATP
GDP
AMP
0.4
log k'
log k'
GTP
0.6
ADP
0.6
2'-3'-cGMP
0.4
0.2
GMP
0.2
0.0
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
10
12
14
16
18
8
10
12
14
16
% MeOH
Cytidyl Nucleotides
Uradyl Nucleotides
0.8
0.8
0.6
UTP
0.6
CTP
UDP
CDP
UMP
0.4
CMP
0.2
log k'
log k'
0.4
0.0
0.2
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.6
8
10
12
14
% MeOH
11
18
% MeOH
16
18
8
10
12
14
% MeOH
16
18
Phosphate Mobile Phase System:
Sample Chromatograms
50%B (12.5% MeOH)
35%B (8.75% MeOH)
GMP
mAU260nm
40
2’-3’-cAMP
ADP
ATP
3’-5’-cAMP
GDP
3’-5’-cGMP
20
mAU 260nm
20
40
2’-3’-cGMP
60
AMP
0
0
GTP
0
10
Time (min)
0
20
4
12
14
16
UDP
mAU 260nm
20
mAU 260nm
20
8
10
Time (min)
UMP
CMP
CDP
UTP
0
0
CTP
0
12
6
40%B (10% MeOH)
40
40%B (10% MeOH)
2
2
4
6
Time (min)
8
10
12
0
2
4
6
8
Time (min)
10
12
Bis-Tris Mobile Phase System:
Results Summary
Adenyl Nucleotides
Guanidyl Nucleotides
1.4
1.2
3'-5'-cAMP
1.2
ATP
0.8
ADP
2'-3'-cGMP
AMP
0.8
GTP
0.6
GDP
GMP
0.6
0.4
log k'
log k'
3'-5'-cGMP
1.0
2'-3'-cAMP
1.0
0.4
0.2
0.2
0.0
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.8
-0.6
10
12
14
16
18
20
22
24
26
10
28
12
14
16
18
20
22
24
Cytidyl Nucleotides
Uradyl Nucleotides
1.0
0.8
CTP
1.0
CDP
0.8
UTP
UDP
CMP
0.6
0.6
0.4
0.4
log k'
log k'
28
1.2
1.2
0.2
UMP
0.2
0.0
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.6
-0.8
-0.8
12
14
16
18
20
% M eOH
13
26
% M eOH
% M eOH
22
24
26
28
12
14
16
18
20
% M eOH
22
24
26
28
Bis-Tris Mobile Phase System:
Sample Chromatograms
mAU 260nm
40
ADP
2’-3’-cGMP
3’-5’-cGMP
GDP
GTP
20
20
3’-5’-cAMP
0
0
ATP
10
Time (min)
20
0
40
CMP
mAU 260nm
20
20
60% B (15% MeOH)
60
60% B (15% MeOH)
10
Time (min)
CDP
UDP
UTP
0
0
CTP
UMP
mAU 260nm
20
40
0
0
14
GMP
mAU 260nm
40
60
2’-3’-cAMP
AMP
60
55% B (13.75% MeOH)
80
60% B (15% MeOH)
2
4
6
8
Time (min)
10
12
14
0
2
4
6
8
Time (min)
10
12
14
Observations
• Retention patterns of the NMP, NDP, and NTP series
with IP reagents, do correlate with the number of
phosphate moeties, which is consistent with an
operational IP mechanism.
• Selectivities of the adenyl and guanidyl nucleotide
sample sets vary not just between themselves but also
between buffer systems.
• The Bis-Tris buffer system permits greater retention of
the nucleotides.
• Of the various silica-bases phases investigated (alkyl,
polar-embedded, fluorinated) with any of the IP
reagents, the alkyl phase provided better retention and
resolution (data not shown).
15
Conclusions
• Quaternary amine-based IP reagents, as components of
mobile phases, provide the necessary retention and
selectivity for nucleotide separations.
• Reduced retention of the nucleotide analytes in the
phosphate mobile phase system could be due to the
scavaging of the IP reagent by the buffer, with a lower
level of IP available to ion-pair with the analytes.
16
Conclusions (cont’d.)
• Conceivably, Bis-Tris, a tertiary amine, could also ionpair with the analytes. However, its selection was
based on its pKa (6.5) and the observation that
triethylamine provided inadequate retention. While
triethylamine would be expected to carry a positive
charge of 1 at pH 7 (even in the presence of 10-20%
methanol, given its pKa of 10.8), that of Bis-Tris would
be ≤ 0.25. Thus, as an IP reagent, it would be expected
to exhibit a very minor role, if any.
• Silica-based alkyl phases appear to provide the
optimum support for RP nucleotide separations under
the conditions examined.
17
Conclusions (cont’d.)
• While zirconia-based phases may display good
retention for nucleotides without ion-pairing
reagents, further investigations have yet to be
conducted to determine conditions that could provide
good chromatography.
• Other perfluoro organic acids will be investigated as
IP reagents that might still permit MS detection.
18
Conclusions (cont’d.)
• Methods have been demonstrated, that while
necessitating the use of IP reagents, they are isocratic
and thus not plaqued by reproducibility issues that can
occur with gradient methods (especially when the
strong solvent does not contain any IP reagent as is
common practice).
• An ideal chromatographic stationary phase for
nucleotide separations would be hydrophobic, inert
(display no other secondary interactions), and stable to
pH 1 or 12. Thus chromatography at extremes of pH
could neutralize either the phosphates or amines, with
absence of secondary interactions permitting good
chromatographic behavior.
19