ANALYTICAL
BIOCHEMISTRY
135, 120-127 (1983)
A Modified High-Performance Thin-Layer Plate for the Separation of Purines and Pyrimidines W. JOST AND H. E. HAUCK E. Merck, P.O. Box 4119, D-6100 Darmstadt, Federal Republic of Germany Received June 10, 1983 Several ways of using the recently developed high-performance thin-layer chromatography (HRTLC) precoated plate NH2 F 254s to separate purines and pyrimidines are described. This precoated plate is coated with silica gel 60 which has been chemically modified with alkylamino groups. In view of the chemical properties of the functional groups bonded to the silica gel matrix, the HPTLC precoated plate NH2 F 254s can be considered to be a weak basic ionexchange plate. In aqueous eluants the substances are separated principally according to charge differences. The HFTLC precoated plate NH2 F 254s can, however, also be used to separate uncharged, polar compounds with organic solvents. Examples of separations and chromatograms for its use in both aqueous and organic ehrants are given. KEY WORDS: high-performance thin-layer chromatography; in situ evaluation; separation of purines, pyrimidines, nucleosides, and nucleotides; ion-exchange thin-layer plate.
The analytical separation of nucleobases, nucleosides, nucleotides, and other purine and pyrimidine derivatives plays an important role in biochemistry. As in column chromatography, ion exchangers have proven themselves as stationary phases for the thin-layer chromatographic investigation of nucleic acid components. Most of the coatings used for the thin-layer chromatographic separation consist of polyethyleneimine cellulose (PEI cellulose) (l-7) or of organic ion-exchange resins (8-12). Buffered aqueous eluants, to which salts have been added and whose pHs have been adjusted to certain values, are usually used as mobile phases. With the aid of partition chromatography, it is furthermore possible to separate nucleotides by using cellulose coatings with alcohol-water mixtures ( 13). The substance spots can be quantitatively evaluated by preparing an autoradiogram using radioactively labeled test substances or by scratching out the spots, with subsequent elution of the test substances and determination of the molar absorbance. As a consequence 0003-2697183 $3.00 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form resewed.
of their nonhomogeneous surface structure, a determination of the separated substances by direct optical evaluation has only given unsatisfactory results on the coatings used until now. The following article describes a recently developed precoated plate for thin-layer chromatography-the HPTLC’ precoated plate NH2 F 254s ( 14), which is particularly suitable for separating purines, pyrimidines, and nucleotides and which furthermore makes it possible to evaluate quantitatively in situ the separated substances. The separation performance and selectivity of the HPTLC precoated plate NH2 F 254s are demonstrated with the aid of some chromatograms. EXPERIMENTAL
SECTION
Plates and solvents. The HPTLC precoated plate NH2 F 254s (Cat. 15647) from E. Merck, Darmstadt, was used. All of the solvents required (water, methanol, acetone, ethanol, ’ Abbreviation used: HPTLC, high-performance thinlayer chromatography. 120
SEPARATION
OF
PURINES
acetonitrile, and chloroform obtained from E. Merck) were of LiChrosolv or analytical grade. Samples and sample sizes. The following substances obtained from E. Merck, Darmstadt, were used: thymine, uric acid, xanthine, hypoxanthine, cytosine, uracil, 2-thiouracil, guanine, adenine, adenosine, nicotinamide adenine dinucleotide (NAD), nicotinamide adenine dinucleotide phosphate (NADP), uridine Sdiphosphoglucose (UDPG), guanosine 5’-triphosphate (GTP), adenosine 5’-monophosphate (AMP), adenosine 5’-diphosphate (ADP), adenosine 5’-triphosphate (ATP), and sodium chloride. The following were supplied by BoehringerMannheim: reduced nicotinamide adenine dinucleotide (NADH), reduced nicotinamide adenine dinucleotide phosphate (NADPH), cytidine, cytidine 5’-monophosphate (CMP), cytidine 5’-diphosphate (CDP), cytidine 5’-triphosphate (CTP), guanosine, guanosine 5’monophosphate (GMP), guanosine 5’-diphosphate (GDP), uridine, uridine 5’-monophosphate (UMP), uridine 5’-diphosphate (UDP), and uridine 5’-triphosphate (UTP). Caffeine, theobromine, and theophylline were supplied by Sigma, Munich. The sample substances were dissolved in water, methanol, or chloroform at concentrations between 1 and 3 mg/ml. Portions of 100 to 300 nl of the individual substance solutions or mixtures were applied to the coating using a Hamilton syringe. The plate was subsequently developed linearly in a normal chamber without chamber saturation. The chromatograms were evaluated in situ with a TLC/ HPTLC scanner with monochromator from Camag, Muttenz (Switzerland) at a uv wavelength of 254 nm. RESULTS
AND
Properties of the HPTLC NH2 F 254s
DISCUSSION
Precoated Plate
The HPTLC precoated plate NH2 F 254s is coated with a chemically modified, readyfor-use sorbent layer. Silica gel with an average
AND
PYRIMIDINES
121
pore diameter of 6 nm (60 A, silica gel 60) was used as the base material. y-Aminopropyl groups are chemically bonded to the surface of the silica gel. The chromatographic properties of the new NH1 precoated plate are primarily determined by the alkylamino groups. In view of the chemical properties of the functional groups, the HPTLC precoated plate NH2 F 254s can be considered to be a weak basic ion-exchange plate (15). The sorbent used has a very narrow particle size distribution and an average particle diameter that is suitable for HPTLC. These result in a high packing density and a very homogeneous surface, which means that the chromatograms can be directly evaluated by optical means without problems. The sorbent coating contains an acid-stable fluorescence indicator so that uv-active substances can be made visible. The fluorescence produced is pale blue and the wavelength of excitation is 254 nm. The amino plate can be wetted with pure water without the addition of salts or organic solvents. Furthermore, no problems arise when this plate is developed in organic solvents or organic solvent mixtures, which means that there are no limitations in the choice of eluants required for chromatography. Figure 1 is a graphic representation of the dependence of the development time, t, on the water content of various eluants (mixtures of water with methanol, ethanol, acetone, and acetonitrile). The migration distance, ZJ, was 5 cm in all cases and the plates were developed in a normal chamber without chamber saturation. In all four cases, curves which pass through a maximum are obtained. The longest relative development times were observed for the ethanol-water system. The shortest development times on the amino plate were measured for the acetonitrile-water solvent system. The maxima for the individual development time curves are in the region of 40 to 50% water for the ethanol-, methanol-, and acetone-water systems. The curve of the acetonitrile-water system runs through a maximum at about 70% water.
122
JOST AND HAUCK
60
: [min]
50
40
30
20
10
0
t 0
50
, 100%
water
FIG. 1. The development time, t, as a function of the water content of the eluant for aqueous mixtures of methanol (O), ethanol (V), acetone (W), and acetonitrile (A). Migration distance, z,: 5 cm.
The development times for pure acetonitrile and pure acetone are about the same as the development time of pure water. The other two pure organic solvents (ethanol and methanol) on the other hand have development times longer than that of pure water.
Chromatographic Investigations When the HPTLC precoated plate NH2 F 254s is used with aqueous eluants, the property of the alkylamino group to function as an ion exchanger is particularly apparent. Under such chromatographic conditions, it is to be expected that the amino plate exhibits a special
selectivity for polyanions. In order to confirm this, investigations into the retention behavior of uracil, uridine, and uridine phosphates with an acetone-water eluant system of various compositions were made. These experiments indeed showed the selectivity expected, but spot formation was very diffise. This phenomenon can be eliminated without considerable change in the selectivity by adding sodium chloride to the eluant. The dependence of the R/values of uracil, uridine, UMP, UDP, UDPG, and UTP on the water content of the acetone-water eluant system is shown in Fig. 2. Sodium chloride (0.2 mol/liter) was added to the eluant. Be-
SEPARATION
‘“E
123
OF PURINES AND PYRIMIDINES
50%
100
0% Acetone % water
FIG. 2. The dependence of the R, values of uracil and uridine and its mono- and polyphosphates on the water content of the eluant. Eluant: acetone/water from O/100 to 100/O (v/v) with the addition of 0.2 mol/ liter NaCl. Cold saturated solutions of NaCI were used for acetone/water = 80/20, 90/10 and for pure acetone. Migration distance, z,: 5 cm. Test substances: uracil (0), uridine (O), UMP (V), UDP (A), UDPG (W, UTP (0).
cause of the limited solubility of the salt in eluant with a large amount of the organic component, cold saturated solutions of sodium chloride were used for acetone-water = 80/20 and 90/ 10 (v/v) and for pure acetone. The diagram shows an increase in the RJ values with increasing water content in the eluant for the test substances which occur as uncharged compounds in the neutral pH range used (uracil and uridine), as well as for the anions (UMP, UDP, UDPG, and UTP). This is principally due to the fact that the solubility of these polar substances in the mobile phase increases with increasing water content, which leads to a reduction in the interaction with the stationary phase. This retention behavior, especially the high Rf level of uridine and uracil with eluants which contain more than 40% water, indicates that the
alkylamino groups bonded to the silica gel matrix effect only weak interactions with the uncharged sample substances. The R/sequence of the nucleotides is mainly determined by the number of negative charges. The number of protons in the phosphate residues of the nucleoside polyphosphates used in this experiment-which are able to dissociate-increases from two to four. UMP and UDPG have two, UDP has three, and UTP has four acidic protons. The interaction of the nucleotides with the stationary phase, which acts as an anion exchanger, becomes stronger with increasing number of negative charges. Hence, within the series of nucleoside polyphosphates, UTP is retarded the most, and UMP and UDPG the least. Table 1 gives the Rf values of some nucleobases, nucleosides, and the corresponding
124
JOST AND HAUCK TABLE 1 R, VALUE~OFSOMENUCLEOBASES,NUCLEOSIDES,ANDNUCLU)SIDEPOLYPHOSPHATES ONTHEHPTLC PRECOATEDPLATE NH2F 254s
Substance
I-4
Substance
4
Substance
4
Substance
Rf
Adenine Adenosine AMP ADP ATP
0.80 0.92 0.54 0.25 0.14
Guanine Guanosine GMP GDP GTP
0.71 0.75 0.28 0.14 0.07
Uracil Uridine UMP UDP UTP
0.80 0.80 0.41 0.18 0.10
Cytosine Cytidine CMP CDP CTP
0.87 0.86 0.50 0.24 0.14
Note. Eluant: ethanol/water = 30/70 (v/v) with the addition of 0.2 mol/liter NaCI. Migration distance, q: 5 cm.
nucleoside polyphosphates with ethanol/water = 30/70 (v/v) as the eluant. Similar retention sequences are obtained for the individual series of nucleobases, nucleosides, and nucleoside polyphosphates. Under the chromatographic 4
conditions given, it is not possible to separate the pyrimidine bases (uracil and cytosine) from their nucleosides (uridine and cytidine). In the case of the two purines (adenine and guanine), separation is possible, with the base being more strongly retarded than the corresponding nucleoside. As mentioned earlier, the nucleoside polyphosphates are separated according to their charge numbers. As a comparison, see Figs. 3,4, and 5, where the chromatographic conditions for each case were optimized. The chromatogram shown in Fig. 6 can also be explained in terms of the same separation 3
4
2
J I 0
I 5cm
3. The separation of adenine, adenosine, and its phosphates. Eluant: ethanol/water = 30170 (v/v) with the addition of 0.2 mol/liter NaCl. Migration distance, z,: 5 cm. Sample substances: ATP (1), ADP (2), AMP (3), adenine (4), adenosine (5). Sample sizes: 300 nl (ah 0.1%). FIG.
I 0
I 5cm
FIG. 4. The separation of guanine, guanosine, and its phosphates. Ehrant: methanol/water = 5/95 (v/v) with the addition of 0.2 mol/liter NaCl. Migration distance, z,: 5 cm. Sample substances: GTP (I), GDP (2), GMP (3), guanine (4), guanosine (5). Sample sizes: 300 nl (all 0.1%).
SEPARATION
;wn
ci
125
OF PURINES AND PYRIMIDINES
FIG. 5. The separation of cytosine, cytidine, and its phosphates. Eluant: acetone/water = 30170 (v/v) with the addition of 0.2 mol/liter NaCl. Migration distance, z,: 5 cm. Sample substances: CTP (I), CDP (2), CMP (3), cytosine, cytidine (4). Sample sires: 300 nl (all 0.1%).
quence, according to increasing Rf values, is NADPH, NADP, NADH, and NAD. The diagram shows a clear group separation. NADPH and NADP are retarded in the R, range between 0.12 and 0.3, and NADH and NAD between 0.6 and 0.75. In both cases, i.e., for both the nicotinamide adenine dinucleotides and for the nicotinamide adenine dinucleotide phosphates, the reduced compounds are more strongly retarded than the nonreduced dinucleotides, which occur as zwitterions under these chromatographic conditions. In the examples described so far, the substances being investigated were mainly separated on the basis of charge differences. It is possible, however, to use the HPTLC precoated plate NH2 F 254s to retard substances according to other criteria. Figure 7 shows the
3
principle. The separation of the four dinucleoside polyphosphates NAD, NADH, NADP, and NADPH is shown. The Rf se3 4
IL L
G
:,
: cm
FIG. 6. The separation of some dinucleoside polyphosphates. Eluant: ethanol/water = 30/70 (v/v) with the addition of 0.2 mol/liter NaCl. Migration distance, ZJ: 5 cm. Sample substances: NADPH (l), NADP (2), NADH (3), NAD (4). Sample sizes: 300 nl (all 0.1%).
0
1 7cm
FIG. 7. The separation of some purine derivatives. Eluant: ethanol/water = 80/20 (v/v) saturated with NaCl. Migration distance, z,: 7 cm. Sample substances: uric acid (I), xanthine (2), hypoxanthine (3), guanine (4) adenine (5). Sample sizes: 300 nl (all 0.1%).
126
JOST AND HAUCK
thin-layer chromatographic separation of some purine derivatives on the amino plate. The purines investigated differ in the number of their hydroxyl and amino groups. The Rf values in the sequence uric acid, xanthine, hypoxanthine increase with decreasing number of OH groups. Guanine has both an OH group and an NH1 group, but is retarded less than hypoxanthine. The NH2 group in the purine ring obviously reduces the interaction of the molecule with the NH2 plate. Adenine has no OH group, but has an NH2 group. It is hence retarded the least and therefore has the largest RI-value in the series of the purines investigated. The following two examples are intended to illustrate the use of the amino plate with organic eluants with very low water contents or with water-free eluants. The chromatogram in Fig. 8 shows the separation of some pyrimidine derivatives using an 80% water-saturated mixture of chloroform/methanol = SO/50 (v/v) as the eluant. The substances investigated differ only slightly from each other in their molecular weights, but they differ considerably with respect to their acidity and polarity.
FIG. 8. The separation of some pyrimidine derivatives. Ehtant: chloroform/methanol = SO/SO(v/v) 80% saturated with water. Migration distance, z,: 7 cm. Sample sub stances: 2-thiouracil (I), uracil (2), cytosine (3), thymine (4). Sample sixes: 300 nl (all 0.1%).
5cm
FIG. 9. The separation of xanthine and its N-methyl derivatives. Eluant: methanol. Migration distance, zf: 5 cm. Sample substances: xanthine (1) 0.03% theophylline (2) 0.03% theobromine (3) 0.06% caffeine (4) 0.02%. Sample sizes: 100 nl.
Thiouracil, which is the most acidic compound, is retarded with the smallest Rf value. Next, with increasing RJ values, follow uracil and cytosine, although they are not completely separated. Thymine is the least polar of the four pyrimidines investigated and is hence retarded the least on the hydrophilic, weakly basic stationary phase. The separation of xanthine and its N-methyl derivatives theophylline, theobromine, and caffeine shown in Fig. 9 can be explained using a similar retention model. Pure methanol was used as the eluant. The Rf values of the purines increase with increasing number of the methyl groups. In this case this means that xanthine has the smallest R/value and caffeine the largest. An interesting aspect of this separation is that the positional isomers theophylline and theobromine are separated with a high degree of selectivity. REFERENCES 1. Randerath, E., and Randerath, K. (1964) J. Chromatogr. 16, 126-129.
SEPARATION
OF
PURINES
2. Reyes, P. (1972) Anal. B&hem. 50, 35-39. 3. Payne, R. C., and Tram, T. W. (1982) Anal. Biochem 121,49-54.
4. Baudendistel, L. J., and Ruh, T. S. (1978) J. Chromatogr. 148, 500-503. 5. Reibel, D. K., and Rovetto, M. J. (1978) J. Chromatogr. 161, 406-409. 6. Barton, R. A., Schulman, A., Jacobson, E. J., and Jacobson, M. K. (1977) J. Chromatogr. 130, 145150. 7. Czebotar, V., and Dietz, A. A. (1976) J. Chromatogr. 124, 141-144. 8. Tomasz, J., and Simoncsits, A. (1975) Chromatographra 8, 348-349.
AND
PYRIMIDINES
127
9. Tomasz, J. (1979) J. Chromatogr. 169, 466-468. 10. Tomasz, J. (1976) J. Chromatogr. 128, 304-308. 11. Tomasz, J. (1973) J. Chromatogr. 84,208-213. 12. Bols, N. C., Bowen, B. W. M., Khor, K. G., and Bolsika, S. A. (1980) Anal. B&hem. 106, 230237. 13. Cohen, L., and Kaplan, R. (1975) Anal. B&hem. 69, 283-288. 14. Jost, W., and Hauck, H. E. (1983) J. Chromatogr. 261,235-244.
15. Helfferich, H. (1959) Ionenaustauscher, Vol. I, p. 14, Verlag Chemie, Weinheim.