Anal. Chem. 1996, 68, 2141-2146
Oligonucleotide Sequence and Composition Determined by Matrix-Assisted Laser Desorption/ Ionization Catherine M. Bentzley and Murray V. Johnston
Department of Chemistry and Biochemistry, University of Delaware, Newark, Delaware 19716 Barbara S. Larsen* and Steven Gutteridge
The DuPont Company, Science and Engineering Labs, P.O. Box 80228, Wilmington, Delaware 19800-0228
Molecular weight measurements of several oligonucleotides ranging in size from 12 to 60 bases were performed by matrix-assisted laser desorption/ionization with a timeof-flight mass spectrometer (MALDI-TOF). In each case, the mass accuracy was better than 0.1%. Sequences for two 12-base oligonucleotides and a 24-base oligonucleotide were determined using calf spleen phosphodiesterase to sequentially cleave from the 5′ end. A MALDITOF spectrum of the digest mixture shortly after the addition of the enzyme produced a characteristic oligonucleotide ladder. Molecular ions in the mass spectrum corresponded to the products of enzymatic cleavage, and the mass differences between these peaks identified the individual nucleotides. The resolution and mass accuracy of MALDI-TOF were sufficient to unambiguously identify the individual nucleotides in the 12- and 24-base strands. The ability to synthesize with accuracy specific sequences of oligonucleotides has in part driven the rapid and enormous expansion of the biotechnological revolution.1 Short oligonucleotides of 15-30 bases are used as primers for DNA sequencing,2 PCR amplification,3 and site-specific mutagenesis of proteins.4 Longer fragments of 30-100 bases or more are used in gene construction such as ribozymes, antigenes, or specific enzyme inhibitors (amptamers).5-7 Unfortunately, because of the difficulties in analyzing short stretches of DNA or RNA, most have been used with no verification of the correct molecular weight, sequence, or purity. The synthetic procedure requires that the bases are protected as the nucleotide chain is extended, and it is necessary to remove these groups postsynthesis for the oligonucleotide to be usable. More recent developments in the synthetic area include the ability to derivatize these molecules with fluorescent molecules or affinity labels. It is therefore essential that a rapid, sensitive, and effective means be devised for analyzing small amounts of product prior to their use to ensure quality and extent of deprotection or derivatization. Over the past two years, matrix-assisted laser desorption/ionization (MALDI) has developed into an effective tool for oligonucleotide analysis.8-12 (1) Itakura, K.; Rossi, J. J.; Wallace, R. B. Annu. Rev. Biochem. 1984, 53, 351. (2) Wu, R.; Tu, C.; Padmanabhun, R. Biochem. Biophys. Res. Commun. 1973, 55, 1092. (3) Gyllensten, U. BioTechniques 1989, 7, 700. (4) Zoller, M. J.; Smith, M. Methods Enzymol. 1983, 100, 468. (5) Stull, R. A.; Szoka, F. C., Jr. Pharm. Res. 1995, 12, 465. (6) Polushin, N. N.; Cohen, J. S. Nucleic Acids Res. 1994, 22, 5492. (7) Stein, C. A.; Cheng, Y.-C. Science 1993, 261, 1004. S0003-2700(95)01213-3 CCC: $12.00
© 1996 American Chemical Society
Oligonucleotide strands up to 100 base units can be determined with 0.05% mass accuracy and picomolar sensitivity. These capabilities should permit verification of the molecular weight and purity of synthetic analogs. We report here a method that provides sequence information as well. Two methods have found wide application for obtaining the sequence of bases in DNA. The Sanger method13 involves enzymatic sequencing through use of a deoxyoligonucleotide primer, a larger DNA template, and all four deoxynucleotide triphosphates and their dideoxy derivatives. Polyacrylamide gels with single base resolution are used to separate the mixture of extended strands and reveal the sequence. This technology can routinely generate up to 700 bases of sequence information from one primer. However, since the basis of the method is polymerase extension of a partial duplex of DNA formed between a primer and template generating a double-stranded product, it is unsuitable for short single-stranded oligonucleotides. The second technique, devised by Maxam and Gilbert,14 involves chemically degrading DNA selectively at individual bases. This technique was used routinely in the past to verify single-stranded sequences but has the disadvantage that the DNA not only requires derivatization to visualize on gels (usually a radioactive label) but also requires large amounts of material and time. Oligonucleotide analysis by mass spectrometry has the advantage of being applicable to small quantities of single-stranded DNA. Crain15 has reviewed the use of conventional mass spectrometric methods such as fast atom bombardment for oligonucleotide analysis. More recently, Little et al.16 used an infrared laser to photodissociate molecular ions of 8-25-base oligonucleotides produced by electrospray ionization. McLuckey et al.17 have shown that modified 6-base oligonucleotides can be (8) Lecchi, P.; Pannell, L. K. J. Am. Soc. Mass Spectrom. 1995, 6, 972. (9) Nordhoff, E.; Karas, M.; Cramer, R.; Hahner, S.; Hillenkamp, F.; Kirpekar, F.; Lezius, A.; Muth, J.; Meier, C.; Engels, J. W. J. Mass Spectrom. 1995, 30, 99. (10) Wang, B. H.; Biemann, K. Anal. Chem. 1994, 66, 1918. (11) Fitzgerald, M. C.; Parr, G. R.; Smith, L. M. Anal. Chem. 1993, 65, 3204. Tang, K.; Allman, S. L.; Chen, C. H. Rapid Commun. Mass Spectrom. 1993, 7, 943. (12) Wu, K. J.; Steding, A.; Becker, C. H. Rapid Commun. Mass Spectrom. 1993, 7, 142. (13) Sanger, F.; Nicklen, S.; Coulson, A. R. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 5463. (14) Maxam, A. M.; Gilbert, W. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 560. (15) Crain, P. F. Mass Spectrom. Rev. 1990, 9, 505. (16) Little, D. P.; Chorush, R. A.; Speir, J. P.; Senko, M. W.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1994, 116, 4893.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996 2141
Table 1. Calculated and Measured Molecular Weights of Three Oligonucleotides m/z sample
peak
calcd
obsda
accuracyb (%)
(M - H)(M - H)2(M - H)3-
10247 5123 3413
10252 5124 3412
0.05 0.02 0.03
(M - H)(M - H)2(M - H)3-
14268 7132 4753
14260 7128 4751
0.06 0.06 0.04
(M - H)(M - H)2(M - H)3-
18383 9192 6125
18373 9186 6120
0.05 0.07 0.06
33 Base Units 5′-CGG GAT CCG TCG ACG ACA GAC GTG ACG GCG TTG-3′
46 Base Units 5′-ATC ACG GTG CTA ACG ATA AGG GCA ATA GCA AGG AGG ATA TTC ATA G-3′
60 Base Units 5′-TCG ACC AAA CGC CAG GAT TGC AAC AGT TCC TCG GTG AAG GTG CTC ACC CAC-3′
a
Average of six spectra. b [|(m/z)calcd - (m/z)measd|/(m/z)calc] × 100%.
distinguished by tandem mass spectrometry. Nordhoff et al.9 have detected prompt fragment ions from oligonucleotides by MALDI with an infrared laser. In each case, the fragmentation products are sequence dependent, and spectral interpretation can be difficult. Pomerantz et al.18 have determined the compostion of 4-8-base oligonucleotides through molecular weight analysis and subsequent application of computational algorithm. This method requires prior knowledge of the composition of one base, and it provides only composition, not sequence. In principle, oligonucleotides can be sequenced by producing an oligonucleotide ladder by enzymic digestion, and the resulting mixture can be analyzed by MALDI. The enzyme calf spleen phosphodiesterase serves as an exonuclease cleaving from the 5′ end. The products of the digest differ in mass by successive loss of individual nucleotides consisting of a phosphate group, a deoxyribose molecule, and one of the four bases. MALDI of the digest quenched shortly after the addition of the diesterase produces a characteristic oligonucleotide ladder. Molecular ions in the mass spectrum, labeled as “exo” peaks, correspond to the products of enzymatic cleavage along the oligonucleotide strand. The mass differences between exo peaks, 313.2 Da for loss of adenine, 329.2 Da for loss of guanine, 289.2 Da for loss of cytosine, and 304.2 Da for loss of thymine, reveal the sequence. Oligonucleotide ladder sequencing by MALDI was first demonstrated by Pieles et al.19 Here we extend this approach to partial sequence determination of 12- and 24-base oligonucleotides, and we quantitatively assess mass accuracy. EXPERIMENTAL SECTION Oligonucleotide Synthesis and Enzymatic Digestion. Oligonucleotide samples were synthesized on an Applied Biosystems 394 DNA-RNA synthesizer (Perkin Elmer, Modesto, CA). The sequences used in this study are given in Table 1 and Figures 5 and 6. Applied Biosystems reagents were used according to the recommended procedure with a 0.2 µmol column. Purification was accomplished by washing the oligonucleotide, retaining the terminal dimethoxytrityl (DMT) protecting group onto an oligonucleotide purification (OP) column (Perkin Elmer) in 10% (17) McLuckey, S. A.; Van Berkel, G. J.; Glish, G. L. J. Am. Soc. Mass Spectrom. 1992, 3, 60. (18) Pomerantz, S. C.; Kowalak, J. A.; McCloskey, J. A. J. Am. Soc. Mass Spectrom. 1993, 4, 204. (19) Pieles, U.; Zurcher, W.; Schar, M.; Moser, H. Nucleic Acids Res. 1993, 21, 3191.
2142 Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
aqueous ammonium hydroxide. Subsequent washes with ammonium hydroxide removed impurities and truncated sequences. The DMT group of the purified compound was removed with 1% TFA solution, and the deprotected oligonucleotide eluted with 20% acetonitrile solution from the OP column. The purified material was dried and resuspended in doubly distilled water prior to use. Enzymatic digestion of the purified oligonucleotides was performed by mixing 5 µL of a 5′-exonuclease calf spleen phosphodiesterase solution (Boehringer Mannheim, Indianapolis, In.; 2 units/mg) with 10 µL of the oligonucleotide (50-100 pmol/µL) at 41 °C. The reaction mixtures were sampled at regular time intervalssevery 3 min for the samples with 12 bases and every 10 min for those with 24 bases. Removing the digest from the heating bath and adding it to the matrix solution (see below) served to quench the reaction. The method was repeated three times to ensure reproducibility. MALDI Sample Preparation and Spectra Acquisition. Samples were prepared for MALDI analysis by adding 1 µL of the reacted oligonucleotide solution to 5 µL of a saturated aqueous solution of 3-hydroxypicolinic acid (HPA) and 1 µL of 0.1 M ammonium citrate (Sigma Chemical Co., St. Louis, MO). A 1 µL aliquot of the matrix-analyte solution was applied to the probe tip and allowed to air dry. The probe was then inserted into the mass spectrometer. A visualization stage on the mass spectrometer allowed the morphology of the matrix-analyte mixture to be assessed. A “good” sample spot was composed of oblong, transparent crystals, with the sample being concentrated on the edges, as illustrated in Figure 1. The encircled area in Figure 1 shows where a strong oligonucleotide signal was obtained. The presence of the enzyme did not appear to alter the morphology of the dried spot or contribute to background peaks in the mass spectra. As reported by Pieles et al.,19 the addition of ammonium citrate to the sample solutions decreased the formation of adduct peaks and improved resolution. Each reacted sample was spotted on the probe tip twice as a check for reproducibility of the sample preparation method. Mass spectra were obtained on a Finnigan MAT Vision 2000 (Bremen, Germany) equipped with a 337 nm nitrogen laser (Laser Science, Inc., Newton, MA). Unless otherwise indicated, all spectra were acquired in the negative ion mode with the ion reflector on. In this mode, a 20 kV postacceleration region precedes the detector. Each spectrum is an average of 20-30 selected laser shots over the analyte surface.
Figure 3. MALDI spectrum of 5′-CTA GAC CGC GGT 3′.
Figure 1. Sample spot for 5′-CTA GAC CGC GGT-3′ in 3-HPA. A location giving an intense analyte signal is circled.
Figure 2. MALDI spectrum of 46-base oligonucleotide.
RESULTS AND DISCUSSION Determination of Molecular Weight and Purity. Figure 2 shows a MALDI spectrum acquired for the oligonucleotide containing 46 bases (see Table 1 for sequence). Three ions are observed: (M - H)-, (M - 2H)2-, and (M - 3H)3-. The intensities of the doubly and triply charged ions were found to increase relative to the molecular ion peak with increasing concentration. A similar concentration dependence, though in the positive ion mode, has been observed for cytochrome c by Zhong and Zhao.20 Figure 2 shows additional peaks and a broadened profile on the low m/z side of the molecular ion peaks. These features were observed in the ion reflection mode but not in the linear mode. This behavior suggest that these ions arise from metastable decomposition of the molecular ion in the flight tube between the ion reflector and detector. Oligonucleotide ions are known to exhibit metastable decomposition.9,21 In the reflector mode of the Vision 2000, the detector is preceded by a 20 kV (20) Zhong, F.; Zhao, S. Rapid Commun. Mass Spectrom. 1995, 9, 570.
postacceleration region. (The linear mode does not require postacceleration.) Postacceleration causes the products of metastable decomposition to arrive at the detector before the undissociated molecular ions. Currently, we cannot assign accurate m/z values for the metastable peaks because of software restrictions in the mass spectrometer data system. However, the oligonucleotide molecular weight can be accurately determined from the centroid of the sharp peak at the high m/z end of the undissociated molecular ions. The effect of postacceleration on a metastably broadened molecular ion peak shape has recently been discussed by King et al.22 Table 1 compares the measured and calculated molecular weights of oligonucleotides containing 33, 46, and 60 bases. The accuracies of the mass measurements range from 0.02 to 0.07%. Molecular weight measurement with this high degree of confidence verifies not only that the base composition is correct but that the sample does not suffer significant contamination from other species, i.e., truncated or incompletely deprotected strands. Although the ranges of accuracies are independent of molecular size, the absolute deviations increase with increasing molecular size, which, as discussed below, makes sequencing of longer oligonucleotides progressively more difficult. In part, this increase is attributed to decreasing resolution. The resolution (m/∆m, where ∆m ) full width at half-maximum) is typically 250 for the 33-base compound, 130 for the 46-base compound, and 35 for the 60-base compound. Sequencing 12-Base Oligonucleotides. The results in the previous section illustrate the ability of MALDI to confirm composition by determining the molecular weight of a synthetic oligonucleotide that it is not contaminated with truncated species. Figure 3 shows the MALDI spectrum of a purified 12 mer (1); the spectrum for 2 is similar and, therefore, is not shown.
5′-CTA GAC CGC GGT-3′ 1 5′-GAT CTC CGC GGA-′ 2 As in Figure 2, peaks to the low m/z side of the molecular ion are attributed to metastable decomposition in the flight tube (21) Schneider, K.; Chait, B. T. Org. Mass Spectrom. 1993, 28, 1353. (22) King, T. B.; Colby, S. M.; Reilly, J. P. Rapid Commun. Mass Spectrom. 1994, 8, 865.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
2143
Figure 4. (a) MALDI spectrum of 5′-CTA GAC CGC GGT-3′ after 9 min of digestion. (b) MALDI spectrum of 5′-GAT CTC CGC GGA3′ after 15 min of digestion. Peaks resulting from metastable decay are labeled with an asterisk.
between the ion reflector and the detector. The sequence cannot be confirmed from the molecular weight. In principle, the sequence can be obtained through enzymatic approaches combined with mass spectrometry. Calf spleen phosphodiesterase, which has a 5′-phosphodiesterase activity, was used to digest 1 and 2 into successively smaller fragments over short incubation times, producing a mixture of single-stranded fragments differing in size by individual nucleotides. Figure 4a gives the resulting MALDI spectrum of the oligonucleotide ladder after 9 min of digestion of 1. Each peak observed is a molecular ion labeled with exo to distinguish it from the molecular ion of the undigested strand. Peaks corresponding to the oligonucleotide after cleavage of C (exo1), then T (exo2), A (exo3), G (exo4), and A (exo5) are observed. Figure 4b shows the MALDI spectrum of 2 after 15 min of digestion. In this case, the peaks are observed at different mass-to-charge ratios than those in Figure 4a. In Figure 4b, the peaks correspond to the removal of G (exo1), then A (exo2), T (exo3), C (exo4), and T (exo5), clearly indicating the difference in the sequence between 1 and 2. There are differences in the relative intensities of the exo peaks for the two sequences, reflecting some base specificity of the enzyme or secondary 2144 Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
Figure 5. MALDI spectrum of a 24-base oligonucleotide after (a) 0 min, (b) 20 min, and (c) 50 min digestion.
structure influences on the digestion rate. Additional peaks on the low m/z side of the exo peaks in Figure 4a and b are again attributed to metastable decomposition. Table 2 shows the mean of six measurements obtained for exo peaks observed in the spectra of 1 acquired after successive 3 min intervals of digestion. The 95% confidence limits of the standard deviation of the exo peak masses range from (0.1 to 2.7 Da and are sufficiently low to permit unambiguous identification of each base in the sequence. Table 3 shows the similar
Table 2. MALDI Analysis of Exo Peak Produced by Enzymatic Digestion of 5′-CTA GAC CGC GGT-3′ time (min) 0 3 6 9 12 15
(M - H)-
c, exo1
T, exo2
A, exo3
G, exo4
3642.3a ((1.6)b 3641.7 ((1.2) 3642.0 ((1.5) 3643.0 ((1.7) 3643.0 ((1.6) 3642.1 ((2.3)
3353.7 ((1.8) 3355.6 ((1.5) 3355.5 ((1.7) 3354.6 ((1.9) 3353.1 ((2.6)
3049.2 ((1.9) 3049.8 ((2.7) 3048.5 ((0.9) 3050.6 ((1.8) 3049.2 ((2.0)
2734.9c
2402.9c
2734.5 ((1.8) 2736.8 ((0.1) 2737.0 ((1.5) 2737.7 ((2.3)
A, exo5
2409.2c 2408.6 ((1.8) 2408.0 ((1.7)
2094.0 ((1.8) 2093.0 ((1.7)
a Mean molecular mass calculated from typically six spectra. b 95% confidence limits. c Peak observed in only one spectrum.
Table 3. MALDI Analysis of Exo Peaks Produced by Enzymatic Digestion of 5′-GAT CTC CGC GGA-3′ time (min) (M - H)-
G, exo1
A, exo2
T, exo3
C, exo4
T, exo5
15
3313.8c
3001.6 ((2.3) 3001.5 ((1.4) 3003.8 ((1.5) 3003.2 ((1.4) 3001.7 ((1.0)
2697.9 ((2.3) 2698.0 ((1.4) 2699.2 ((0.56) 2699.0 ((1.4) 2698.9 ((1.0)
2408.3 ((0.72) 2408.0 ((0.64) 2409.0 ((0.48) 2410.7 ((1.6) 2410.0 ((1.3)
2105.7c
12
3311.2 ((5.1) 3313.0 ((2.6) 3314.6 ((0.96) 3316.8c
0 3
3642.2a ((1.0)b 3638.1 ((3.0)
6 9
2105.5c 2105.0 ((0.4) 2107.6 ((1.8) 2106.1 ((2.2)
a Mean molecular mass calculated from typically six spectra. b 95% confidence limits. c Peak observed in only one spectrum.
results for the digestion of 2. Again, the accuracy of the measurement of the exo peak masses is sufficient to permit unambiguous identification of each base in the sequence. The smaller metastable peaks on the low m/z side of the exo peaks occur fortuitously at positions in the spectrum where the mass differences are much less than a single nucleotide loss and therefore cannot be misinterpreted as products of the diesterase activity.
Sequencing a 24-Base Oligonucleotide. As previously discussed, the molecular weight measurement of an oligionucleotide with 24 bases, Figure 5a, confirms the composition and purity of the oligionucleotide. The enzymatic digest with the calf spleen phosphodiesterase is shown in Figure 5b and c after 20 and 50 min, respectively. As the digestion proceeds, the molecular ion intensity decreases rapidly. Figure 5b shows that, after 20 min, the molecular ion completely disappears, and peaks ranging from exo1 to exo7 are observed. Although the metastable decomposition peaks are nearly as large as the exo peaks, they are easily disregarded since the apparent mass loss is ∼168 Da, while loss of a base unit must be 289 Da or greater. Figure 5c shows that, after 50 min of digestion, peaks ranging from exo1 to exo16 are observed. The signal intensities of exo1 through exo4 are quite small, suggesting that the sequence itself affects the digestion rate of this enzyme. Further work is in progress to analyze this aspect of phosphodiesterase specificity. Table 4 show the mean of 4-6 measurements for the exo peaks observed after successive 10 min intervals of digestion. Samples analyzed beyond 60 min of digestion resulted in exo peaks of low intensity, which make unambiguous determination of the sequence more difficult. The variability in digestion rate causes the intensity of the early exo peaks to vary, which requires initial spectra to be acquired at shorter time intervals, particularly to determine the first base from a single spectrum. The initial errors for the mass differences between the (M - H)- and exo1 peaks in the 10 min spectra spans A, G, and T are higher than usual. This is a direct effect of the rate of digestion. Other bases in the sequence are less problematic. For example, the 40 min spectra permit unambiguous determination of 10 of the 12 bases spanned by the exo peaks observed. The one exceptions are the mass difference between exo4 and exo5 (both A and T are within the 95% confidence limits) and between exo7 and exo8 (both C and T are within the 95% confidence limits). However, the 30 min spectra are sufficient to distinguish A from T between exo4 and exo5, and the 50 min spectra are sufficient to distinguish C from T between exo7 and exo8. The 50 min spectra unambiguously identify G between exo13 and 14, but all bases are excluded by the 95% confidence limits for the mass difference between exo14 and exo15. Clearly, experimental procedures that yield higher resolution and mass accuracy are needed to permit unambiguous sequencing of large oligonucleotides.
Table 4. MALDI Analysis of Exo Peaks Produced by Enzymatic Digestion of 5′-GCC AAT GCG CTG CGC TGT CAA TGC-3′ time (min) (M-H)1 0 10 20 30 40 50
G, exo1
7327a ((0.67) 7311b 6988a ((13.7) ((11.5) 6986a ((9.1) 6993b ((1.9) 6986a ((0.5) 6987b ((2.4)
C, exo2
6693a ((8.0) 6693a ((8.6) 6697b ((10.3) 6702a ((3.9) 6684b
C, exo3
A, exo4
A, exo5
T, exo6
6402a ((9.8) 6412b ((3.8) 6409b ((2.9)
6084d
5774d
5484d
6100b ((4.5) 6093a ((4.8) 6091b ((9.4)
5788b ((2.8) 5784a ((1.3) 5778b ((7.6)
5485b ((1.8) 5481a ((2.5) 5472a ((7.0)
G, exo7
C, exo8
5159b ((0.7) 5156a ((3.2) 5154a ((3.7)
4854d 4861a ((6.6) 4866a ((3.8)
G, exo9
C, exo10
T, exo11
G, exo12
C, exo13
G, exo14
C, exo15
4535a ((0.9) 4536a ((3.8)
4248b ((2.0) 4242a ((8.2)
3944b ((0.2) 3943a ((2.7)
3615b ((0.5) 3613a ((7.1)
3326b ((0.5) 3326a 2998a 2739c ((2.0) ((1.3) ((1.0)
a Mean molecular mass and 95% confidence limits (in parentheses) calculated from six spectra. b Calculations based upon four spectra. c Calculations based upon 2 spectra. d Peak observed in only one spectrum.
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
2145
CONCLUSIONS MALDI-TOF can be applied to measure the molecular weights of oligonucleotide strands containing 12-60 bases, allowing the composition and purity to be verified. Determining the molecular weights of large oligonucleotides is convenient and routine. Sequence information can be obtained through enzymatic digestion of the oligonucleotides using the exonuclease calf spleen phosphodiesterase. Exposure of single-stranded DNA to the enzyme as a function of time creates an oligonucleotide ladder in the MALDI spectrum, with peaks at mass differences characteristic of the progressive loss of individual nucleotides from the 5′ end. Oligonucleotide ladders for two 12 mers with identical base composition but different sequences could be distinguished. The sequencing method was extended to a 24 mer with little difficulty. From the relative intensity of the resulting exo peaks, it is apparent that the rate of digestion of the enzyme is not constant but is influenced by strand sequence and secondary structure.
2146
Analytical Chemistry, Vol. 68, No. 13, July 1, 1996
Future studies will focus on the base specificity of the enzyme as well as extending the length of the strand analyzed. To improve our results, other matrixes, such as 6-azathiothymine, were suggested, as well as the use of time lag focusing to increase resolution. ACKNOWLEDGMENT This research was supported in part by funds from the Delaware Research Partnership. We thank Mary Bailey for synthesizing and purifying the oligonucleotides used in this study.
Received for review December 15, 1995. Accepted March 22, 1996.X AC951213N X
Abstract published in Advance ACS Abstracts, May 1, 1996.
