Anal.Chem. 2000, 72, 4386-4393
`
`High-Performance Liquid
`Chromatography-Electrospray Ionization Mass
`Spectrometry of Single- and Double-Stranded
`Nucleic Acids Using Monolithic Capillary Columns
`
`Andreas Premstaller, Herbert Oberacher, and Christian G. Huber*
`
`InstituteofAnalyticalChemistryandRadiochemistry,Leopold-Franzens-University,Innrain52a,A-6020Innsbruck,Austria
`
`Monolithic capillary columns were prepared by copolym-
`erization of styrene and divinylbenzene inside a 200-(cid:237)m
`i.d. fused silica capillary using a mixture of tetrahydro-
`furan and decanol as porogen. With gradients of acetoni-
`trile in 100 mM triethylammonium acetate, the synthe-
`sized columns allowed the rapid and highly efficient
`separation of single-stranded oligodeoxynucleotides and
`double-stranded DNA fragments by ion-pair reversed-
`phase high-performance liquid chromatography (IP-RP-
`HPLC). Compared with capillary columns packed with
`micropellicular, octadecylated poly-(styrene/divinylben-
`zene) particles, an improvement in column performance
`of approximately 40% was obtained, enabling the analysis
`of an 18-mer oligodeoxynucleotide with a column ef-
`ficiency of more than 190 000 plates per meter. The
`chromatographic separation system was on-line-coupled
`to electrospray ionization mass spectrometry (ESI-MS).
`To improve the mass spectrometric detectabilities, 25 mM
`triethylammonium bicarbonate was utilized as an ion-pair
`reagent at the cost of only little reduction in separation
`performance and acetonitrile was added postcolumn as
`the sheath liquid through the triaxial electrospray probe.
`High-quality mass spectra of femtomole amounts of 3-mer
`to 80-mer oligodeoxynucleotides were recorded showing
`very little cation adduction. Double-stranded DNA frag-
`ments ranging in size from 51 to 587 base pairs were
`separated and detected by IP-RP-HPLC-ESI-MS. Accurate
`mass determination by deconvolution of the mass spectra
`was feasible for DNA fragments up to the 267-mer with a
`molecular mass of 165 019, whereas the spectra of longer
`fragments were too complex for deconvolution because
`of incomplete separation due to overloading of the col-
`umn. Finally, on-line IP-RP-HPLC tandem MS was ap-
`plied to the sequencing of short oligodeoxynucleotides.
`
`For more than 30 years, columns packed with microparticulate
`sorbents have been successfully applied as separation media in
`high-performance liquid chromatography (HPLC).1 Despite many
`advantages, HPLC columns packed with microparticulate, porous
`
`* Corresponding author. Tel.: +43 512 507 5176. Fax: +43 512 507 2767.
`E-mail: Christian.Huber@uibk.ac.at
`(1) Majors, R. E. Supplement to LC GC 1998, 7-21.
`
`4386 AnalyticalChemistry,Vol.72,No.18,September15,2000
`
`stationary phases have some limitations, such as the relatively
`large void volume between the packed particles and the slow
`diffusional mass transfer of solutes into and out of the stagnant
`mobile phase present in the pores of the separation medium.2,3
`One approach to alleviate the problem of restricted mass transfer
`and intraparticular void volume is the concept of monolithic
`chromatographic beds, where the separation medium consists of
`a continuous rod of a rigid, porous polymer which has no
`interstitial volume but only internal porosity consisting of mi-
`cropores and macropores.4-8 Because of the absence of intrapar-
`ticular volume, all of the mobile phase is forced to flow through
`the pores of the separation medium.9 According to theory, mass
`transport is enhanced by such convection10,11 and has a positive
`effect on chromatographic efficiency.12 Monolithic chromato-
`graphic beds are usually prepared by polymerization of suitable
`monomers and porogens in a stainless steel or fused silica tube
`which acts as a mold.6,13,14 The porous structure is achieved as a
`result of the phase separation which occurs during the polymer-
`ization of a monomer or monomer mixture containing appropriate
`amounts of both a cross-linking monomer and a porogenic solvent
`or a mixture of porogenic solvents.15,16
`Miniaturized chromatographic separation systems applying
`capillary columns of 10-500-(cid:237)m inner diameter are frequently the
`method of choice for the separation and characterization of
`biopolymer mixtures, when the amount of available sample is
`limited. The concept of monolithic stationary phases is especially
`
`(2) Martin, A. J.; Synge, R. L. M. Biochem. J. 1941, 35, 1358.
`(3) Unger, K. K. In Packings and Stationary Phases in Chromatographic
`Techniques; Unger, K. K., Ed.; Marcel Dekker: New York, 1990; p 75.
`(4) Hansen, L. C.; Sievers, R. E. J. Chromatogr. 1974, 99, 123-133.
`(5) Hjerte´n, S.; Liao, J.-L.; Zhang, R. J. Chromatogr. 1989, 473, 273-275.
`(6) Svec, F.; Fre´chet, J. M. J. Anal. Chem. 1992, 64, 820-822.
`(7) Minakuchi, H.; Nakanishi, K.; Soga, N.; Ishizuka, N.; Tanaka, N. Anal. Chem.
`1996, 68, 3498-3501.
`(8) Gusev, I.; Huang, X.; Horva´th, C. J. Chromatogr., A 1999, 855, 273-290.
`(9) Petro, M.; Svec, F.; Fre´chet, J. M. J. J. Chromatogr., A 1996, 752, 59-66.
`(10) Liapis, A. I.; McCoy, M. A. J. Chromatogr., A 1994, 660, 85.
`(11) Rodrigues, A. E.; Lu, Z. P.; Loureiro, J. M.; Carta, G. J. Chromatogr. 1993,
`653, 189.
`(12) Afeyan, N. B.; Gordon, N. F.; Mazsaroff, I.; Varady, L.; Fulton, S. P.; Yang,
`Y. B.; Regnier, F. E. J. Chromatogr. 1990, 519, 1-29.
`(13) Hjerte´n, S.; Li, Y. M.; Liao, J. L.; Mohammad, J.; Nakazato, K.; Pettersson,
`G. Nature (London) 1992, 356, 810-811.
`(14) Viklund, C.; Svec, F.; Fre´chet, J. M. J. Chem. Mater. 1996, 8, 744-750.
`(15) Seidl, J.; Malinsky, J.; Dusek, K.; Heitz, W. Adv. Polym. Sci. 1967, 5, 113-
`213.
`(16) Svec, F.; Fre´chet, J. M. J. Macromolecules 1995, 28, 7580-7582.
`
`10.1021/ac000283d CCC: $19.00 © 2000 American Chemical Society
`Published on Web 08/18/2000
`
`1
`
`MTX1045
`
`

`

`favorable for the fabrication of capillary columns, because the
`immobilization of the monolith at the capillary wall eliminates the
`necessity of preparing a tiny retaining frit,17 which is one of the
`more tedious and difficult-to-control steps during the manufacture
`of packed-bed capillary columns.18 Moreover, the possibility of
`direct on-line conjugation of capillary HPLC to mass spectrometry
`makes available highly valuable information about the structure
`and identity of the separated compounds.19 Electrospray ionization
`mass spectrometry (ESI-MS), by virtue of the multiple charging
`of biopolymers and the very soft ionization process, has become
`one of the most important mass spectrometric techniques for the
`analysis of nucleic acids.20 Nevertheless, the success of ESI-MS
`for the characterization of nucleic acids largely depends on the
`purity of the sample that is introduced into the mass spectrom-
`eter.21 The major difficulties arise due to the tendency of nucleic
`acids to form quite stable adducts with cations, resulting in mass
`spectra of poor quality.22,23 The on-line sample preparation of
`nucleic acids by chromatographic separation prior to ESI-MS is
`very attractive because it not only removes cations from nucleic
`acid samples but also fractionates nucleic acids in mixtures that
`are too complex for direct-infusion ESI-MS.
`Although continuous-bed columns have proved to be efficient
`for the separation of biopolymers such as peptides24,25 and
`proteins,6,13,26 the concept of continuous beds has not successfully
`been transferred to the separation of nucleic acids by ion-pair
`reversed-phase HPLC (IP-RP-HPLC).27 We have utilized 200-(cid:237)m
`i.d. capillary columns packed with micropellicular polymer beads
`for the chromatographic separation and mass spectrometric
`characterization of nucleic acids up to 40 nucleotides by on-line
`IP-RP-HPLC-ESI-MS.28 Therefore, our goal in this study was to
`investigate the applicability of monolithic chromatographic beds
`to high-resolution separations of nucleic acids by IP-RP-HPLC and
`to extend the size range of the separated and detected nucleic
`acids. It is shown that both single- and double-stranded nucleic
`acids in a size range of 3 nucleotides (nt) to 600 base pairs (bp)
`can be separated with high efficiency and identified after separa-
`tion by ESI-MS.
`
`EXPERIMENTAL SECTION
`Chemicals and Oligodeoxynucleotide Samples. Acetonitrile
`(HPLC gradient grade), divinylbenzene (synthesis grade), metha-
`nol (HPLC gradient grade), styrene (synthesis grade), and
`tetrahydrofuran (analytical reagent grade) were obtained from
`
`(17) Ericson, C.; Liao, J.-L.; Nakazato, K.; Hjerte´n, S. J. Chromatogr., A 1997,
`767, 33-41.
`(18) Oberacher, H.; Krajete, A.; Parson, W.; Huber, C. G. J. Chromatogr., A
`2000, in press.
`(19) Tomer, K. B.; Moseley, M. A.; Deterding, L. J.; Parker, C. E. Mass Spectrom.
`Rev. 1994, 13, 431-457.
`(20) Nordhoff, E.; Kirpekar, F.; Roepstorff, P. Mass Spectrom. Rev. 1996, 15,
`76-138.
`(21) Portier, N.; Van Dorsselaer, A.; Cordier, Y.; Roch, O.; Bischoff, R. Nucleic
`Acids Res. 1994, 22(19), 3895-3903.
`(22) Stults, J. T.; Marsters, J. C. Rapid Commun. Mass Spectrom. 1991, 5, 359-
`363.
`(23) Huber, C. G.; Buchmeiser, M. R. Anal. Chem. 1998, 70, 5288-5295.
`(24) Wang, Q. C.; Svec, F.; Fre´chet, J. M. J. J. Chromatogr., A 1994, 669, 230-
`235.
`(25) Li, Y. M.; Brosted, P.; Hjerte´n, S.; Nyberg, F.; Silberring, J. J. Chromatogr.,
`B 1995, 664, 426-430.
`(26) Wang, Q. C.; Svec, F.; Fre´chet, J. M. J. Anal. Chem. 1993, 65, 2243-2248.
`(27) Huber, C. G. J. Chromatogr., A 1998, 806, 3-30.
`(28) Huber, C. G.; Krajete, A. Anal. Chem. 1999, 71, 3730-3739.
`
`Merck (Darmstadt, Germany). Styrene and divinylbenzene were
`distilled before use. Acetic acid (analytical reagent grade), azo-
`bisisobutyronitrile (synthesis grade), decanol (synthesis grade),
`and triethylamine (p.a.) were purchased from Fluka (Buchs,
`Switzerland). A 1.0 M stock solution of triethylammonium acetate
`(TEAA) was prepared by dissolving equimolar amounts of tri-
`ethylamine and acetic acid in water. A 0.50 M stock solution of
`triethylammonium bicarbonate (TEAB) was prepared by passing
`carbon dioxide gas (AGA, Vienna, Austria) through a 0.50 M
`aqueous solution of triethylamine at 5 (cid:176)C until pH 8.4 -8.9 was
`reached. For preparation of all aqueous solutions, high-purity water
`(Epure, Barnstead Co., Newton, MA) was used. The standards of
`phosphorylated and nonphosphorylated oligodeoxynucleotides
`(p(dT)12-18, p(dT)12-18, p(dT)19-24, p(dT)25-30) were purchased as
`sodium salts from Pharmacia (Uppsala, Sweden) or Sigma-Aldrich
`(St. Louis, MO). The synthetic oligodeoxynucleotides (dT)24 (Mr
`7 238.71), a 5¢-dimethoxytritylated 5-mer (DMTr-ATGCG, Mr
`1805.42), and an 80-mer (CCCCAGTGCT GCAATGATAC CGC-
`GAGACCC ACGCTCACCG GCTCCAGATT TATCAGCAAT AAAC-
`CAGCCA GCCGGAAGGG, Mr 24 527.17) were ordered from
`Microsynth (Balgach, Switzerland) and used without further
`purification. The size standard of double-stranded DNA restriction
`fragments (pBR322 DNA-Hae III digest) was purchased from
`Sigma Aldrich.
`Preparation of Monolithic and Packed-Bed Capillary
`Columns.29 Polyimide-coated fused silica capillary tubing of 350-
`(cid:237)m o.d. and 200-(cid:237)m i.d. was obtained from Polymicro Technolo-
`gies (Phoenix, AZ). A 1-m piece of fused silica capillary tubing
`was silanized with 3-(trimethoxysilyl)propyl methacrylate accord-
`ing to the procedure published in ref 3030 in order to ensure
`immobilization of the monolith at the capillary wall. Then, a 300-
`mm piece of the silanized capillary was filled, using a plastic
`syringe, with a mixture comprising 50 (cid:237)L of styrene, 50 (cid:237)L of
`divinylbenzene, 130 (cid:237)L of decanol, 20 (cid:237)L of tetrahydrofuran, and
`10 mg/mL of azobisisobutyronitrile. The mixture was polymerized
`at 70 (cid:176)C for 24 h. After polymerization,
`the capillary was
`extensively flushed with acetonitrile at a flow rate of 5.0 (cid:237)L/min
`and finally cut into 60-mm-long pieces. Octadecylated PS-DVB
`particles (PS-DVB-C18) were synthesized as published in the
`literature.31 The PS-DVB-C18 stationary phase has been com-
`mercialized as DNASep by Transgenomic Inc. (Santa Clara, CA).
`Packed-bed capillary columns were prepared according to the
`procedure described in ref 18.18
`High-Performance Liquid Chromatography. The HPLC
`system consisted of a low-pressure gradient micro pump (model
`Rheos 2000, Flux Instruments, Karlskoga, Sweden) controlled by
`a personal computer, a vacuum degasser (Knauer, Berlin, Ger-
`many), a column thermostat made from 3.3-mm o.d. copper tubing
`which was heated by means of a circulating water bath (model K
`20 KP, Lauda, Lauda-Ko¨nigshofen, Germany), a microinjector
`(model C4-1004, Valco Instruments Co. Inc., Houston, TX) with a
`200- or 500-nL internal sample loop, a variable wavelength detector
`
`(29) Huber, C. G.; Premstaller, A.; Oberacher, H. Method for preparation of
`monolithic capillary columns for high performance liquid chromatography
`of single and double stranded nucleic acids. U. S. primary patent application,
`60/178.553.
`(30) Huang, X.; Horva´th, C. J. Chromatogr., A 1997, 788, 155-164.
`(31) Huber, C. G.; Oefner, P. J.; Bonn, G. K. Anal. Biochem. 1993, 212, 351-
`358.
`
`AnalyticalChemistry,Vol.72,No.18,September15,2000 4387
`
`2
`
`

`

`Figure 1. High-resolution capillary IP-RP-HPLC separation of
`phosphorylated oligodeoxynucleotide ladders in a monolithic capillary
`column. Column, continuous PS-DVB, 60 (cid:2) 0.20 mm i.d.; mobile
`phase, (A) 100 mM TEAA, pH 6.97, (B) 100 mM TEAA, pH 6.97,
`20% acetonitrile; linear gradient, 15-45% B in 3.5 min, 45-55% B
`in 2.5 min, 55-65% B in 4.0 min; flow-rate, 2.5 (cid:237)L/min; temperature,
`50 (cid:176) C; detection, UV, 254 nm; sample, p(dA)12-18, p(dT)12-30,
`40-98 fmol of each oligodeoxynucleotide.
`
`Figure 2. High-resolution capillary IP-RP-HPLC separation of a
`mixture of double-stranded DNA fragments in a monolithic capillary
`column. Column, continuous PS-DVB, 60 (cid:2) 0.20 mm i.d.; mobile
`phase, (A) 100 mM TEAA, pH 7.00, (B) 100 mM TEAA, pH 7.00,
`20% acetonitrile; linear gradient, 35-75% B in 3.0 min, 75-95% B
`in 12.0 min; flow-rate, 2.2 (cid:237)L/min; temperature, 50 (cid:176) C; detection, UV,
`254 nm; sample, pBR322 DNA-Hae III digest, 1.81 fmol of each
`fragment.
`
`(model UltiMate UV detector, LC Packings, Amsterdam, Neth-
`erlands) with a Z-shaped capillary detector cell ULT-UZ-N-10, 3nL
`cell, LC Packings), and a PC-based data system (Chromeleon 4.30,
`Dionex-Softron, Germering, Germany).
`Electrospray Ionization Mass Spectrometry and Coupling
`with Capillary Liquid Chromatography. ESI-MS was performed
`on a Finnigan MAT LCQ quadrupole ion trap mass spectrometer
`(Finnigan MAT, San Jose, CA, used in Figures 4-7) or a Finnigan
`MAT TSQ 7000 triple quadrupole mass spectrometer (used in
`Figure 3) equipped with an electrospray ion source. The capillary
`column was directly connected to the spray capillary (fused silica,
`105-(cid:237)m o.d., 40-(cid:237)m i.d., Polymicro Technologies) by means of a
`microtight union (Upchurch Scientific, Oak Harbor, WA). A
`syringe pump equipped with a 250-(cid:237)L glass syringe (Unimetrics,
`
`4388 AnalyticalChemistry,Vol.72,No.18,September15,2000
`
`Figure 3. Scanning electron micrographs of (a) underivatized PS-
`DVB particles, (b) octadecylated PS-DVB particles, and (c) an
`underivatized PS-DVB monolith.
`
`Shorewood, IL) was used for continuous-infusion experiments and
`for pumping sheath liquid. For analysis with pneumatically assisted
`ESI, an electrospray voltage of 3.2-3.7 kV and a nitrogen sheath
`gas flow of 20-30 arbitrary units (LCQ) or 28-33 psi (TSQ) were
`employed. The temperature of the heated capillary was set to 200
`(cid:176)C. Total ion chromatograms and mass spectra were recorded
`on a personal computer with the LCQ Navigator software version
`1.2 or on a DEC-Alpha 3000 workstation with the ICIS software
`version 8.3.0 (Finnigan). Mass calibration and coarse tuning was
`performed in the positive-ion mode by direct infusion of a solution
`
`3
`
`

`

`Figure 4. Separation and mass analysis of a series of oligodeoxy-
`thymidylic acids. Column, continuous PS-DVB, 60 (cid:2) 0.20 mm i.d.;
`mobile phase, (A) 10 mM TEAA, pH 7.00, (B) 10 mM TEAA, pH 7.00,
`20% acetonitrile; linear gradient, 20-60% B in 10.0 min; flow-rate,
`temperature, 50 (cid:176)C; scan, 800-2000 amu in 2 s;
`3.2 (cid:237)L/min;
`electrospray voltage, 3.8 kV; sheath gas, 34 psi N2; sheath liquid,
`acetonitrile; flow rate, 3.0 (cid:237)L/min; sample, (dT)6-18, 50 ng.
`
`of caffeine (Sigma, St. Louis, MO), methionyl-arginyl-phenylalanyl-
`alanine (Finnigan), and Ultramark 1621 (Finnigan). Fine tuning
`for ESI-MS of oligodeoxynucleotides in the negative-ion mode was
`performed by infusion of 3.0 (cid:237)L/min of a 20 pmol/(cid:237)L solution of
`(dT)24 in 25 mM aqueous TEAB containing 20% acetonitrile
`(v/v). A sheath flow of 3.0 (cid:237)L/min acetonitrile was added through
`the triaxial electrospray probe. For all direct infusion experiments,
`cations present in the oligodeoxynucleotide samples were re-
`moved by on-line cation exchange using a 20 (cid:2) 0.50 mm i.d.
`cation-exchange microcolumn packed with 38-75-(cid:237)m Dowex 50
`WX8 particles (BioRad, Richmond, CA).23 For IP-RP-HPLC-ESI-
`MS analysis, oligodeoxynucleotides and DNA fragments were
`injected without prior cation removal.
`
`RESULTS AND DISCUSSION
`Performance of the Monolithic Capillary Columns for
`Oligodeoxynucleotide- and dsDNA Separations. Since slow
`mass transfer kinetics is often the limiting factor for speed and
`efficiency in biopolymer separations, the enhancement of intra-
`particular mass transfer is particularly important for the rapid
`chromatographic separation of large molecules having low diffu-
`sivities such as nucleic acids.32 One way to circumvent intrapar-
`ticular diffusion is the complete elimination of the support pores
`resulting in stationary phases of the micropellicular configura-
`tion.27,33,34 The main advantage of such sorbents rests with the
`rapid mass transfer, because the only remaining particle-based
`diffusion limitations are in a thin layer at the surface of nonporous
`particles. This allows the rapid separation of oligodeoxynucleotides
`and DNA fragments within a fraction of the time that is required
`with conventional porous packing materials.27,35 Consequently, to
`
`(32) Chen, H.; Horva´th, C. J. Chromatogr., A 1995, 705, 3-20.
`(33) Unger, K. K.; Jilge, G.; Kinkel, J. N.; Hearn, M. T. W. J. Chromatogr. 1986,
`359, 61-72.
`(34) Kalghatgi, K.; Horva´th, C. In Analytical Biotechnology-Capillary Electrophoresis
`and Chromatography; Horva´th, C., Nikelly, J. G., Eds.; American Chemical
`Society: Washington, DC, 1990; pp 163-180.
`(35) Huber, C. G.; Oefner, P. J.; Bonn, G. K. Chromatographia 1993, 37, 653-
`658.
`
`Figure 5. Quality control of a synthetic 80-mer oligodeoxynucleotide.
`Column, continuous PS-DVB, 60 (cid:2) 0.20 mm i.d.; mobile phase, (A)
`25 mM TEAB, pH 8.40, (B) 25 mM TEAB, pH 8.40, 20% acetonitrile;
`linear gradient, 20-100% B in 15 min;
`flow-rate, 3.0 (cid:237)L/min;
`temperature, 50 (cid:176) C; scan, 1000-3000 amu; electrospray voltage, 3.2
`kV; sheath gas, 30 units; sheath liquid, acetonitrile; flow rate, 3.0 (cid:237)L/
`min; sample, 5.0 pmol raw product.
`
`maintain the separation speed and performance upon the transition
`from a packed to a monolithic chromatographic bed, the synthesis
`of a monolith has to be tuned such that its morphology resembles
`that of a chromatographic bed formed by nonporous particles with
`relatively large channels for convective flow and without any
`micropores. Therefore, we used decanol and tetrahydrofuran as
`porogens with relatively poor solvency for poly(styrene/divinyl-
`benzene) (PS-DVB), resulting in the formation of large channels
`in the monolithic bed.
`Following polymerization, extensive washing with acetonitrile,
`and equilibration with 100 mM TEAA-5.0% acetonitrile solution,
`
`AnalyticalChemistry,Vol.72,No.18,September15,2000 4389
`
`4
`
`

`

`Table 1. Comparison of the Resolution Values for
`Oligodexynucleotides and Double Stranded DNA using
`Packed and Monolithic Capillary Columns
`
`compounds
`
`p(dT)12/p(dT)13
`p(dT)29/p(dT)30
`51/57 bp
`540/587 bp
`
`resolution
`with packed column
`with monolithic column
`
`3.05
`1.04
`3.88
`1.11
`
`5.38
`2.38
`5.15
`2.70
`
`ticles of the same dimensions. The permeabilities of the monolithic
`columns and the packed columns were similar, resulting in back
`pressures between 18 and 20 MPa at a flow rate of 2.6 (cid:237)L/min
`and a 50 (cid:176)C column temperature, which indicates that the dimen-
`sion of the channels for convective flow in both chromatographic
`beds is of approximately the same size. The relative standard
`deviations of the peak widths at half-height, both among various
`batches of packed capillary columns and monolithic capillary
`columns, were better than 10%, which demonstrates that column
`preparation was reproducible and allowed the comparison of the
`chromatographic performance of both column types. The chro-
`matographic performance was evaluated by gradient separation
`of a mixture of (dT)12-18 with a gradient of 5.0-12.0% acetonitrile
`in 100 mM TEAA in 10 min. Three injections of the standard onto
`each of the three columns gave average peak widths at half-height
`for (dT)18 of 2.28 ( 0.22 s (sample size N ) 9, standard deviation
`sd ) 0.29 s, level of significance P ) 95%) for the monolithic
`columns and 3.84 ( 0.16 s (N ) 9, sd ) 0.20 s, P ) 95%) for the
`packed-bed capillary columns. These values demonstrate that the
`chromatographic performance of monolithic columns for oligode-
`oxynucleotide separations is approximately 40% better than that
`of packed-bed columns. The chromatographic efficiency of the
`monolithic columns was determined by isocratic elution of (dT)18
`with an eluent containing 7.8% acetonitrile in 100 mM TEAA at a
`flow rate of 2.4 (cid:237)L/min. At a column temperature of 50 (cid:176)C, the
`number of theoretical plates exceeded 11 500 plates for a 60-mm
`column, corresponding to 191 000 theoretical plates per meter.
`Figure 1 illustrates the high-resolution separation of phospho-
`rylated oligodeoxyadenylic- and oligothymidylic acids ranging in
`size from 12 to 30 nt. Gradient elution with 3.0-9.0% acetonitrile
`in 3.5 min, followed by 9.0-11.0% acetonitrile in 2.5 min, and finally
`11.0-13.0% acetonitrile in 4.0 min in 100 mM TEAA resulted in
`peak widths at half-height of 1.3 s for p(dA)12 to 2.4 s for p(dT)30,
`which allowed the baseline resolution of the whole series up to
`the 30-mer within 8.2 min. The resolution of homologous oligode-
`oxynucleotides obtained with the monolithic column clearly
`surpasses that of a capillary column packed with PS-DVB-C18
`beads (Table 1, compare also Figure 1 in ref 2828).
`IP-RP-HPLC has been shown to be efficient not only for the
`rapid separation of single-stranded oligodeoxynucleotides but also
`for the fractionation of double-stranded DNA fragments up to chain
`lengths of 2000 bp.36 The applicability of the monolithic PS-DVB
`stationary phase to the IP-RP-HPLC separation of double-stranded
`DNA was tested by injection of a pBR322 DNA-Hae III digest,
`which was separated in 12.5 min using a gradient of 7.0-15.0%
`acetonitrile in 3 min, followed by 15.0-19.0% acetonitrile in 12
`
`(36) Huber, C. G.; Oefner, P. J.; Bonn, G. K. Anal. Chem. 1995, 67, 578-585.
`
`Figure 6. Separation and mass analysis of double-stranded DNA
`fragments from a restriction digest of the pBR322 plasmid. Column,
`continuous PS-DVB, 60 (cid:2) 0.20 mm i.d.; mobile phase, (A) 25 mM
`TEAB, pH 8.40, (B) 25 mM TEAB, pH 8.40, 20% acetonitrile; linear
`gradient, 15-30% B in 3.0 min, followed by 30-50% B in 12 min;
`flow rate, 2.8 (cid:237)L/min; temperature, 40 (cid:176)C; scan, 1000-3000 amu;
`electrospray voltage, 3.2 kV; sheath gas, 32 units; sheath liquid,
`acetonitrile; flow rate, 3 (cid:237)L/min; sample, pBR322 DNA-HaeIII digest,
`180 fmol of each fragment.
`
`Figure 7. Extracted mass spectra of the 80 bp (a), 123/124 bp (b),
`and 267 bp (c) fragments of
`the pBR322 DNA-Hae III digest.
`Conditions as in Figure 6.
`
`the performance of three different 60 (cid:2) 0.20 mm i.d. monolithic
`capillary columns was compared with that of three columns
`packed with octadecylated, 2.3-(cid:237)m micropellicular PS-DVB par-
`
`4390 AnalyticalChemistry,Vol.72,No.18,September15,2000
`
`5
`
`

`

`min in 100 mM TEAA at a flow rate of 2.2 (cid:237)L/min. Again, the
`chromatogram of the mixture depicted in Figure 2 with fragments
`ranging from 51 to 587 bp, as well as the resolution values given
`in Table 1, demonstrate that the separation performance of
`monolithic columns is superior to that of packed-bed columns with
`respect to their separation capability for nucleic acids (compare
`also Figure 1 in ref 3636). In this context, one distinctive difference
`between the octadecylated PS-DVB stationary phase and the PS-
`DVB monolith deserves discussion. While derivatization with
`octadecyl groups has been shown to be essential to obtain high
`chromatographic efficiency with PS-DVB particles,31,37 monolithic
`stationary phases exhibit superior efficiency already without
`derivatization. One possible explanation for this different behavior
`is the formation of the polymer in two different chemical environ-
`ments. The PS-DVB particles were polymerized in aqueous
`suspension, where poor solvation of the hydrophobic polymer by
`the hydrophilic solvent resulted in a relatively flat surface, as
`revealed by the scanning electron micrograph depicted in Figure
`3a. The particle surface became rugulose after derivatization with
`octadecyl groups offering a contact area greater than that of a
`smooth spherical particle (Figure 3b). The formation of the
`monolithic bed, on the other hand, took place in an entirely
`organic environment. During polymerization, small primary par-
`ticles of approximately 0.5 (cid:237)m coagulated to form the porous
`monolith, resulting in a surface structure which came close to
`that of the octadecylated PS-DVB particles (Figure 3c).
`On-line Separation and Mass Determination of Synthetic
`Oligodeoxynucleotides. For many of the analytical problems
`encountered with oligodeoxynucleotides, chromatographic separa-
`tion in combination with UV detection is not sufficient to get a
`conclusive answer. The on-line conjugation of chromatographic
`separation to mass spectrometry, however, offers a potent tool
`for the characterization and identification of oligodeoxynucleotides
`on the basis of accurate mass determinations and fragmentation
`patterns. For example, the HPLC-UV analysis of a (dT)12-18
`standard that was left overnight at room temperature showed a
`number of small peaks eluting before the seven major peaks
`(chromatogram not shown). We supposed that the small peaks
`were phosphorylated or nonphosphorylated hydrolysis products
`of (dT)12-18, but this assumption was not definitive until the
`separation system was on-line-coupled to ESI-MS, which revealed
`that they were nonphosphorylated hydrolyzates ranging from the
`6-mer to the 11-mer (Figure 4). Application of a gradient from 4
`to 12.0% acetonitrile in 10 mM TEAA enabled the separation of
`all oligothymidylic acids from the 6-mer to the 18-mer. As
`suggested in a recent report, acetonitrile was added postcolumn
`as sheath liquid to enhance the mass spectrometric detectability
`of the separated oligodeoxynucleotides.38 This example demon-
`strates that, by using on-line IP-RP-HPLC-ESI-MS, the unequivocal
`identification of low femtomole amounts of oligodeoxynucleotides
`is feasible on the basis of their molecular masses (Table 2). With
`a gradient of 4.0-12.0% acetonitrile in 50 mM TEAA in 10 min,
`oligothymidylic acids as small as the 3-mer were eluted as sharp
`and symmetric peaks (chromatogram not shown), whereas mono-
`nucleotides could not be chromatographed, even with a neat
`
`(37) Huber, C. G.; Oefner, P. J.; Preuss, E.; Bonn, G. K. Nucleic Acids Res. 1993,
`21, 1061-1066.
`(38) Huber, C. G.; Krajete, A. J. Chromatogr., A 2000, 870, 413-424.
`
`Table 2. Measured and Theoretical Masses of (dT)6-18
`
`oligodeoxy-
`nucleotide
`
`retentiontime
`(min)
`
`molecular mass
`measured
`theoretical
`
`rel deviation
`(%)
`
`(dT)6
`(dT)7
`(dT)8
`(dT)9
`(dT)10
`(dT)11
`(dT)12
`(dT)13
`(dT)14
`(dT)15
`(dT)16
`(dT)17
`(dT)18
`
`1.77
`2.63
`3.59
`4.35
`4.94
`5.44
`5.76
`6.13
`6.39
`6.66
`6.92
`7.12
`7.35
`
`1763.09
`2066.96
`2371.90
`2675.28
`2978.95
`3284.43
`3589.29
`3892.78
`4197.47
`4501.81
`4806.26
`5109.19
`5413.35
`
`1763.21
`2067.40
`2371.59
`2675.79
`2979.98
`3284.18
`3588.37
`3892.57
`4196.76
`4500.96
`4805.15
`5109.35
`5413.54
`
`0.006
`0.021
`-0.013
`0.019
`0.035
`-0.008
`-0.026
`-0.006
`-0.017
`-0.019
`-0.023
`0.003
`0.004
`
`aqueous eluent, due to the lack of retention. From the crystal
`structure of the trinucleotide (A)3 it can be inferred that a 3-mer
`oligodeoxynucleotide has an almost globular structure with a
`diameter of approximately 1.0 nm.39 Because penetration of
`analytes into micropores of commensurate size would cause
`considerable band broadening, the capability of the monolithic
`stationary phase to efficiently separate such small molecules is a
`good indicator for the absence of micropores.
`Refined chemistry has significantly improved the efficiency of
`automated solid-phase synthesis of long oligodeoxynucleotide
`sequences. However, assuming a coupling efficiency of 98-99%
`per synthesis cycle, the maximum yield of an 80-mer oligodeoxy-
`nucleotide will be only 20-45%, and contamination of the target
`sequence with a number of failure sequences or partially depro-
`tected sequences is generally observed.28,40 Figure 5a illustrates
`the analysis of 5.0 pmol of a crude 80-mer oligodeoxynucleotide.
`The high number of partly resolved peaks eluting between 2 and
`6 min made identification and quantitation of the target sequence
`from the reconstructed ion chromatogram impossible. However,
`extraction of a selected ion chromatogram at m/z 1167.0, 1225.5,
`and 1290.0 clearly identified the target sequence eluting at 3.8
`min (Figure 5b). Averaging and deconvolution of four mass
`spectra between 3.7 and 3.8 min yielded a molecular mass of
`24 525.0 (Figure 5c), which correlates well with a theoretical mass
`of 24 527.17 (0.009% relative deviation). Moreover, the deconvo-
`luted mass spectrum (inset in Figure 5c) did not show notable
`cation adduction, which verifies that IP-RP-HPLC is an efficient
`method for the desalting of oligodeoxynucleotides. Comparison
`of the mass spectrum extracted from the chromatogram (Figure
`5c) with that of an 80-mer obtained by direct-infusion ESI-MS
`(compare Figure 3 in ref 2323) clearly corroborates the high value
`of on-line coupling of chromatographic separation to mass
`spectrometry, because the chemical background in the mass
`spectrum is greatly reduced upon chromatographic separation,
`and exact mass measurement is possible using IP-RP-HPLC-ESI-
`MS with only one fiftieth of the amount of sample that is consumed
`during direct-infusion ESI-MS.
`On-line Separation and Mass Determination of dsDNA
`Fragments. The potential to obtain high-quality ESI mass spectra
`
`(39) Suck, D.; Manor, P. C.; Saenger, W. Acta Crystallogr., Sect. B 1976, 32,
`1727-1737.
`(40) Huber, C. G.; Stimpfl, E.; Oefner, P. J.; Bonn, G. K. LC-GC 1996, 14, 114-
`127.
`
`AnalyticalChemistry,Vol.72,No.18,September15,2000 4391
`
`6
`
`

`

`Table 3. Molecular Masses of Double-Stranded DNA
`Fragments from the pBR322 DNA-Hae III Digest
`
`fragment
`
`51
`57
`64
`80
`89
`104
`123
`124
`184
`192
`213
`234
`267
`434
`458
`502
`540
`587
`
`positiona
`942-992
`993-1049
`534-597
`3410-3489
`832-920
`298-401
`175-297
`402-525
`1263-1446
`4344-174
`1050-1262
`598-831
`3490-3756
`2518-2951
`2952-3409
`1447-1948
`1949-2488
`3757-4343
`
`molecular mass
`measuredb
`theoretical
`31 565 ( 24 (4)
`35 252 ( 54 (6)
`39 573 ( 84 (7)
`49 494 ( 43 (10)
`55 058 ( 41 (14)
`64 391 ( 56 (22)
`76 059 ( 49 (15)
`76 731 ( 44 (17)
`113 802 ( 140 (15)
`118 722 ( 123 (17)
`131 733 ( 148 (18)
`144 708 ( 127 (25)
`165 091 ( 230 (12)
`n. d.c
`n. d.
`n. d.
`n. d.
`n. d.
`
`31 559.57
`35 263.04
`39 592.83
`49 475.35
`55 038.97
`64 312.99
`76 045.76
`76 675.05
`113 747.36
`118 668.82
`131 674.02
`144 646.56
`165 019.11
`268 240.41
`283 002.81
`310 240.12
`333 738.33
`362 707.09
`
`rel deviation
`(%)
`
`0.018
`-0.032
`-0.026
`0.038
`0.034
`0.12
`0.017
`0.073
`0.048
`0.045
`0.045
`0.042
`0.044
`n. d.
`n. d.
`n. d.
`n. d.
`n. d.
`
`a Position relative to the Eco RI restriction site in pBR322. b Mo-
`lecular mass given as average ( standard deviation (number of charge
`states used to calculate the average molecular mass). c Not determined.
`
`of large, double-stranded DNA is essentially determined by the
`amount of salt as well as the number of different compounds
`present in the sample mixture.21,41 Recently, Muddiman et. al.
`published the mass spectrum of a 500-bp polymerase chain
`reaction product, which had been purified by ethanol precipitation
`followed by microdialysis.42 Although the amount of DNA that
`was analyzed in the ion cyclotron resonance mass spectrometer
`was in the low femtomole range, much more material was required
`for purification before mass measurement. Hence, there is an
`urgent need for rapid on-line separation and purification protocols
`requiring only minute sample amou