Adaptation of a thermospray liquid ... - ACS Publications

DOI: 10.1016/0165-9936(91)87009-U. A.L.L. Duchateau, B.H.M. Munsters, G.T.C. Kwakkenbos, R.G.J. Van Leuken. Selection of buffers and of an ion-pairing...
0 downloads 0 Views 561KB Size
248

Anal. Chem. 1990, 62, 248-252

(3) Lee, Y. W.; Cook. C. E.; Ito, Y. J . Liq. Cbromatogr. 1988, 7 7 , 37-53. (4) Ito. Y. J . LIO. Cbromtwr. 1985. 8 . 2131. i5j Lee, Y . to, Y.; Fa& Q. c.; Cook, C. E. J , Liq. chromatogr. 1988, 77(1), 75-89. (6) Vestal, M. L. Science 1984, 226, 275-281. (7) Covey, T. R.; Lee, E. D.; Bruins, A. D.; Henion, J. D. Anal. Cbem. 1986, 58, 1451A-1461A. (8) Games, D. E. Adv. Mass Spectrom. 1986, 704, 323-342. (9) Voyksner, R. D. High Performance Liquid Chromatography/Mass Spectrometry. In Analytical Aspects of Drug Tesring;Deutsch, D. G., Ed.: J. Wiley and Sons: New York, 1989; Chapter 7, pp 173-202. (10) Blakely, C. R.; Carmody, J. J.; Vestal, M. L. J . Am. Chem. Soc. 1980, 702, 5931-5933. (11) Blakely, C. R.; McAdams, M. J.; Vestal, M. L. J . Cbromatogr. 1978, 758,261-276. (12) Blakely, C. R.; Carmody, J. J.; Vestal, M. Anal. Cbem. 1980, 52, 1636-1641. (13) Blakely, C. R.; Vestal, M. L. Anal. Chem. 1983, 5 5 , 750-754. (14) Voyksner. R. D.: Haney, C. A. Anal. Cbem. 1985, 5 7 , 991-996.

(15) Conchillo, A.; Casas. J.; Messequer, A.; Abian, J. Biomed. Environm. Mass Spectrum 1988, 76,339-344. (16) Bellar, T. A.; Budde, W. L. Anal. Chem. 1988, 6 0 , 2076-2083. (17) Watson, D.; Taylor, G. W.; Murray, S. Biomed. Environ. Mass Spectrom. 1986, 73,65-69. (18) Barcelo, D. Biomed. Environ. Mass Spechom. 1988, 77, 363-369. (19) Lee, Y. W.; Voyksner. R. D.; Fang, 0. C.; Cook, C. E.; Ito, Y. J . L i q . Cbromatogr. 1988, 77(1), 153-171. (20) Voyksner, R. D.; Bursey, J. T.; Pellizzari, E. I. Anal. Chem. 1984, 56, 1507-1 5 14. (21) Martin, D. G.; Mizsak, S. A.; Nielsen, J. W. J . Antibiot. 1986, 39,603. (22) Martin, D. G.; Peltonen, R. E.; Nielsen, J. W. J . Antbiot. 1986, 39, 721. (23) Wang, H. J.; Chen, Y. Y. Acta Pharm. Sin. 1985, 2 0 , 832-835. (24) Ito, Y. Laboratory of Technical Development, NIH. Bethesda. MD, private communication.

RECEIVED for review July 10,1989. Accepted October 25,1989.

Adaptation of a Thermospray Liquid Chromatography/Mass Spectrometry Interface for Use with Alkaline Anion Exchange Liquid Chromatography of Carbohydrates Richard C. Simpson* and Catherine C. Fenselau Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, Maryland 21228

Mark R. Hardy, R. Reid Townsend, and Yuan C. Lee Department of Biology, Johns Hopkins University, Baltimore, Maryland 21218

Robert J. Cotter Department of Pharmacology, Johns Hopkins Ilniuersity, Baltimore, Maryland 21205

An interface is described that allows the direct coupling of high-performance alkaline anion exchange liquid chromatography with thermospray mass spectrometry. A membrane suppressor is used to remove nonvolatile alkaline salts from the mobile phase afler the chromatographic process is completed and prlor to introduction into the mass spectrometer. Examples are gtven of both isocratk and gradient separations of a three-component test mixture of N-acetylated mono- and disaccharides, followed by on-line mass spectral data acquisition. Sensitivity studies show minimum detection limits for the test compounds to be in the microgram range.

efficient resolution of underivatized saccharide positional isomers which is not possible to achieve with previous normalor reverse-phase approaches. Our laboratories are actively engaged in developing direct interfacing between alkaline anion exchange HPLC and mass spectrometry (MS), with the end goal being application of the interface to analyses of carbohydrates by HPLC/MS. The combination of excellent chromatographic resolution and mass spectral structure information could substantially contribute to the role of HPLC/MS in carbohydrate chemistry. This paper describes our initial efforts in the successful development of an interface to allow on-line coupling of alkaline anion exchange HPLC of underivatized N-acetylated saccharides with thermospray mass spectrometry.

INTRODUCTION

EXPERIMENTAL SECTION N-Acetylglucosamine (GlcNAc), N-acetylmannosamine (ManNAc),N-acetyllactosamine (LacNAc),glucose, sucrose, lactose, and stachyose were obtained from Sigma Chemical Co. (St. Louis, MO) and used without further purification. Doubly distilled deionized water was produced in-house and used for all solutions and reagents All other chemicals were of reagent grade purity. Apparatus. The HPLC was a Model 114M binary gradient system (Beckman,Berkeley, CA). The system utilized a Rheodyne 7125 injection valve (Cotati, CA) fitted with either a 20- or 250-pL4 sample loop. The alkaline mobile phase was either 10 or 100 mM NaOH. The variable-wavelength UV detector was a Model 757 fitted with a high-pressure flow cell (Applied Biosystems, Ramsey, Nd). The bonster HPLC pump was a Spectroflow 400 dual piston pump, also purchased from Applied Riosystems An AS-6 ana-

Various high-performance liquid chromatography (HPLC) techniques have been developed to facilitate carbohydrate analyses in an effort to better understand their role in biochemical processes. Some of these methods require precolumn derivatization of the sugars to enable or enhance detection ( I , 2). Many of the techniques utilize reverse-phase (Le. alkyl bonded phases) (3, 4 ) or normal-phase (i.e. amino bonded phases) ( 5 , 6) separation mechanisms. Additionally, cation exchange HPLC has also been used for the resolution of various saccharides ( 7 . 8). Most recently anion exchange HPLC, employing an alkaline mobile phase and electrochemical detection, has been successfully used for t,he resolution of sacchrides (9, 10). Work by Hardy and Townsend (11) indicates that alkaline anion exchange HPLC permits

Materials.

0003-2700/90/0362-0248$02.50/0 1990 American Chemical Society

ANALYTICAL CHEMISTRY, VOL. 62, NO. 3, FEBRUARY 1, 1990

249

-

Table I. Efficiency of Na+ Removal HPLC

PUMP

RNION

EXCHANGE

method of analysis: low-level sodium ion selective electrode suppressor: Dionex AMMS regenerant: 70 mN H2S04at 10 mL/min NaOH flow rate: 1 mL/min [NaOH], mM

residual [Na+l,rM

[NaOH], mM

residual [Na+I,rM

COLUnN

blank 10 40

i

A Figure 1. Schematic diagram of integrated anion exchange HPLC/MS system: TS, thermospray interface: MS. mass spectrometer.

lytical anion exchange column (250 X 4.6 mm, 10-pm polymeric packing), an AMMS anion micromembrane suppressor, and regenerant delivery system were obtained through Dionex Corp. (Sunnyvale, CA). A Corning low-level sodium ion selective combination electrode was purchased from VWR scientific (Bridgeport, NJ). A 10vL mixing chamber was supplied by The Lee Co. (Westbrook, CT). The mass spectrometer was a MS80RF magnetic instrument equipped with a thermospray interface (Karatos Analytical, Ramsey, NJ). Data collection and reduction were performed with DS-90 software on a Data General S/120 computer (Kratos). The mass spectrometer was calibrated over the appropriate mass ranges with a poly(ethy1ene glycol)/O.1 M ammonium acetate solution. The scan rate was held a t 3 s/decade and a resolution of 1500 was used throughout the entire study. To minimize base-line noise due to low molecular weight mobile phase components, the lowmass cut-off was set at 125 amu. The thermospray parameters were optimized for the mobile phase composition and flow rate. Typical temperatures were 190 "C for the vaporizer, 125 "C for the probe, 200 "C for the source block, and 250 "C for the vapor jet. Integrated System. A schematic of the integrated system is shown in Figure 1. The HPLC pump delivers the NaOH mobile phase through the analytical anion exchange column a t a flow rate of 1mL/min. The column effluent is directly passed through the suppressor for the removal of Na+ from the mobile phase. The suppressor is continuously regenerated with 70 mN H2S04,which is supplied by the regenerant delivery system pressurized with He to 25 psi. This pressure resulted in a regenerant flow rate of 10 mL/min. The mobile phase, which is now free of Na+ and mildly acidic, passes through the variable-wavelength UV detector for conventional UV monitoring. After exiting the UV detector, the mobile phase is directed into one arm of a "T" fitting. The second arm of the "T" fitting is connected to a reservoir of 0.6 M ammonium acetate. which serves as the thermospray ionization reagent. The base of the "T" fitting is connected to the inlet check valve of the booster pump. The booster pump, which is of the dual piston design, is set at a flow rate of 1.2 mL/min. Thus all of the mobile phase (at

10 30 1R

70 100

19 13

1 mL/min) is delivered through the inlet check valve and the remaining difference (0.2 mL/min) is supplied by the 0.6 M ammonium acetate solution. The resulting combined mobile phase/ammonium acetate solution is then 0.1 M in ammonium acetate, which is in the proper concentration range to assist in the thermospray ionization process. To reduce dispersion of the resolved chromatographic bands, the pulse damper in the booster pump is bypassed and the combined mobile phase/ammonium acetate solution is pumped at a flow rate of 1.2 mL/min from the booster pump outlet check valve through a 10-pL mixing chamber into the thermospray interface and subsequently intc the mass Spectrometer.

RESULTS AND DISCUSSION The direct interfacing of alkaline anion exchange HPLC to thermospray mass spectrometry requires removal of nonvolatile NaOH from the mobile phase prior to introduction into the mass spectrometer. Incorporation of an anion micromembrane suppressor into the system between the HPLC column outlet and mass spectrometer accomplishes this removal. r h e micromembrane suppressor achieves Na+ removal through a dialysis or ion replacement process, with Na+ passing out of the mobile phase, through the membrane, and into the regenerant solution where it is swept to waste. In turn, H+ is transported from the regenerant solution, through the membrane, and into the mobile phase. Thus the alkaline mobile phase is rendered slightly acidic and the OH- mobile phase component is converted to mater. To maximize the efficiency of Na+ removal, the regenerant solution was used a t the maximum concentration (70 mN H,SO,) and delivered a t the maximum pressure (25 psi, corresponding to 10 mL/ min) recommended by the manufacturer. Table I presents quantitative results concerning the efficiency of Ka+ removal by the micromembrane suppressor. The measurements, obtained by using a low-level sodium ion selective electrode, demonstrate the ability of the suppressor to efficiently remove Na+ over a wide concentration range in the mobile phase. The implication of these results is that the mobile phase concentration of NaOH may be substantially varied to obtain the desired chromatographic resolution of analytes yet still be removed prior to entry into the mass spectrometer. A major obstacle to interfacing with the thermospray mass spectrometer is the limit on the maximum operating pressure of the micromembrane suppressor. The manufacturer sets a maximum operating pressure of 100 psi, above which the membrane will rupture or the suppressor assembly will leak. Unfortunately the back pressure generated by the operating thermospray interface is approximately 600- 700 psi. Therefore the suppressor cannot be directly connected to the thermospray interface. Placement of a booster pump (12) between the suppressor and the thermospray interface provides the required pressure drop between the two components. Since the booster pump requires more liquid than is being supplied by the mobile phase, the mobile phase may exit the suppressor and enter the booster pump a t essentially atmospheric pressure. The volumetric difference between the mobile-phase flow rate and the set booster pump flow rate (0.2 mL/min) is provided by the 0.6 M ammonium acetate

250

}

t

VI

ANALYTICAL CHEMISTRY, VOL. 62, NO. 3, FEBRUARY 1, 1990

1

TOTAL