Realities of high-throughput liquid chromatography/mass spectrometry purification of large combinatorial libraries: a report on overall sample throughput using parallel purification.

We report on the development of a validated, streamlined high-throughput process for the purification of parallel-synthesis-derived combinatorial libraries. The steps involved in this library purification process include dissolution of dry films of crude synthetic material, dual-column LC/MS purification, dual-column postpurification analysis, quantitation, reformatting, and submission of pure compounds for registration. Although the purification and postpurification analysis times decreased essentially linearly as a function of the number of HPLC columns employed, it was not possible to decrease the total purification process time linearly as a function of the number of columns employed in the system. This was due primarily to the fact that numerous steps in the total purification process are independent of sample analysis and purification (e.g., evaporation, reconstitution, and reformatting, etc.). Additionally, experiments were also performed to assess whether separate gradient pumps were necessary for each channel of this two-channel LC/MS or if acceptable results could be reliably obtained by splitting the flow from one set of gradient pumps between two HPLC columns. On the basis of the parallel, two-column LC/MS system employed in this work, throughput estimates were extrapolated to more massively parallel systems (e.g., four-channel LC/MS).

Mass-directed fractionation and isolation of pharmaceutical compounds by packed-column supercritical fluid chromatography/mass spectrometry.

An automated packed-column semi-preparative supercritical fluid chromatography/mass spectrometry (SFC/MS) system incorporating mass-directed fraction collection has been designed and implemented as an alternative to preparative HPLC and preparative HPLC/MS (PrepLC/MS) for the purification of pharmaceutical compounds. The system incorporates a single quadrupole mass spectrometer and a supercritical fluid chromatograph. Separations were achieved using a binary solvent system consisting of carbon dioxide and methanol. Purification of SFC-separated compounds was achieved incorporating mass-directed fraction collection, enabling selective isolation of the target molecular weight compound and eliminating the collection of undesired compounds (e.g., by-products, excess starting materials, etc.). Cross contamination between fractions and recoveries of the system were investigated. Mass spectrometer ionization with basic mobile additives is discussed, and examples of preparative SFC/MS chiral separations are presented. Early experiences suggest SFC will be a powerful and complementary technique to HPLC for the purification of pharmaceutical compounds.

Differences in phosphorylation of human and chicken stathmin by MAP kinase.

Stathmin/Op18 is a highly conserved 19 kDa cytosolic phosphoprotein. Human and chicken stathmin share 93% identity with only 11 amino acid substitutions. One of the substituted amino acids is serine 25, which is a glycine in chicken stathmin. In human stathmin, serine 25 is the main phosphorylation site for MAP kinase. In this study, we have compared the phosphorylation of human and chicken stathmin. The proteins were expressed in Sf9 cells using the baculovirus expression system and purified for in vitro phosphorylation assays. Phosphorylation with MAP kinase showed that chicken stathmin was phosphorylated 10 times less than human stathmin. To identify the phosphorylation sites we used liquid chromatography/mass spectrometry (LC/MS/MS). The only amino acid found phosphorylated was serine 38, which corresponds to the minor phosphorylation site in human stathmin. Phosphorylation with p34(cdc2)- and cGMP-dependent protein kinases gave almost identical phosphorylation levels in the two stathmins.

A multiple electrospray interface for parallel mass spectrometric analyses of compound libraries.

A parallel spray interface for mass spectrometry is described. This new electrospray interface enables effluent flow streams from an array of HPLC columns to be sampled independently and sequentially on a chromatographic time-scale. Unlike our previously reported parallel LC-MS interface, which incorporated a dual-sheath spray interface accommodating up to four flow streams that are sampled simultaneously, this new interface permits up to four columns to be sampled sequentially by means of a stepping motor and rotating plate assembly. Effluent flow streams from an array of four HPLC columns are connected to a parallel arrangement of electrospray needles co-axial to the mass spectrometer entrance aperture. Within the needle assembly, the individual spray tips are oriented in a circular array, where each needle is situated 90 degrees relative to one another for four-column operation. An eight-column system is described with needles spaced at 45 degree intervals. In between the needle assembly and the mass spectrometer entrance aperture is a Teflon disk with a through-hole that is mounted to a stepping motor assembly. By precisely controlling the stepping of the motor assembly, it is possible to sample each of the spray positions multiple times per second. By operating the quadrupole mass spectrometer in the single ion monitoring (SIM) mode, it was possible to acquire data at each of the spray positions during the course of the elution of compounds from the HPLC column array while maintaining both good sensitivity and peak shape. Preliminary results suggest this technique will be useful for high throughput combinatorial library analysis and profiling.

Identification of in vitro phosphorylation sites in the growth cone protein SCG10. Effect Of phosphorylation site mutants on microtubule-destabilizing activity.

SCG10 is a neuron-specific, membrane-associated protein that is highly concentrated in growth cones of developing neurons. Previous studies have suggested that it is a regulator of microtubule dynamics and that it may influence microtubule polymerization in growth cones. Here, we demonstrate that in vivo, SCG10 exists in both phosphorylated and unphosphorylated forms. By two-dimensional gel electrophoresis, two phosphoisoforms were detected in neonatal rat brain. Using in vitro phosphorylated recombinant protein, four phosphorylation sites were identified in the SCG10 sequence. Ser-50 and Ser-97 were the target sites for protein kinase A, Ser-62 and Ser-73 for mitogen-activated protein kinase and Ser-73 for cyclin-dependent kinase. We also show that overexpression of SCG10 induces a disruption of the microtubule network in COS-7 cells. By expressing different phosphorylation site mutants, we have dissected the roles of the individual phosphorylation sites in regulating its microtubule-destabilizing activity. We show that nonphosphorylatable mutants have increased activity, whereas mutants in which phosphorylation is mimicked by serine-to-aspartate substitutions have decreased activity. These data suggest that the microtubule-destabilizing activity of SCG10 is regulated by phosphorylation, and that SCG10 may link signal transduction of growth or guidance cues involving serine/threonine protein kinases to alterations of microtubule dynamics in the growth cone.

New developments in automated PrepLCMS extends the robustness and utility of the method for compound library analysis and purification.

New developments in the high throughput purification of combinatorial libraries by automated preparative LC-MS is presented. To facilitate high speed purifications at the multimilligram level, short columns operated at ultra high flow rates were incorporated. In order to match the linear velocity of the short analytical columns for high speed separations (operated at 4.0 mL/min), it was required to operate the preparative columns at flow rates in excess of 70 mL/min. For chromatographically well-behaved compounds, analytical LC-MS analyses and preparative LC-MS analyses could be achieved in as little as 5 min. For compounds exhibiting poor chromatographic peak shapes and/or for compound mixtures requiring higher resolution separations, slightly longer preparative LC-MS analysis times were required (8-10 min/sample).

Correlation of the phosphorylation states of pp60c-src with tyrosine kinase activity: the intramolecular pY530-SH2 complex retains significant activity if Y419 is phosphorylated.

Rapid digestion of pp60c-src tyrosine kinase (src TK) in combination with electrospray ionization mass spectrometry enabled the determination of the time course for autophosphorylation of three tyrosine sites (Y338, Y419, and Y530) and a correlation with src TK activity. A form of src TK was purified from baculovirus-infected cells which contains only Y338 partially phosphorylated. Incubation with MgATP increases the phosphorylation of all three sites. The autophosphorylation and dephosphorylation of Y419 are directly correlated with the level of src TK activity. The role of Y338 phosphorylation is unknown. Conditions resulting in complete autophosphorylation of Y530 were identified by electrospray ionization mass spectrometry. Surface plasmon resonance detection and size exclusion chromatography provide direct evidence for an intramolecular pY530-SH2 complex, supporting previous models [Matsuda, M., Mayer, B.J., Fukui, Y., & Hanafusa, H. (1990) Science 248, 1537-1539]. Contrary to these models, when the enzyme is fully phosphorylated on Y530, phosphorylated on Y419, and present only as the intramolecular pY530-SH2 complex, 20% of the kinase activity is retained. In addition, the k(m)'s for substrates are unaffected. Disruption of the pY530-SH2 interaction and activation of kinase activity by a high-affinity SH2 ligand yield a Kactivation which is 200-fold larger than the Kd for ligand binding to the uncomplexed src SH2 domain. These data suggest a Keq of 200 (unitless) for the intramolecular association of pY530 with the SH2 domain. We propose that the pY530-SH2 interaction modulates signal transduction by down-regulating src TK activity 5-fold, and perhaps more importantly by inhibiting protein-protein interactions with the SH2 domain. These results have significant implications relative to the development of SH2 ligands as therapeutics to control aberrant signal transduction. These ligands will be 200-fold less effective at inhibiting protein-protein interactions versus down-regulated src TK than versus activated src TK. This should minimize activation of src TK activity in normal cells and lead to an increased therapeutic index.

Phosphorylation of a cAMP-specific phosphodiesterase (HSPDE4B2B) by mitogen-activated protein kinase.

A cAMP-specific phosphodiesterase, HSPDE4B2B, was found to be phosphorylated when expressed in Sf9 cells or yeast. Deletion of amino acids 81-151 and 529-564 had no effect on the phosphorylation of HSPDE4B2B. Mass spectrometric analysis of purified HSPDE4B2B(1-564). HSPDE4B2B(81-564) and HSPDE4B2B(152-528) showed that phosphorylation occurred predominantly on Ser487 and Ser489. The stoicheiometry of phosphorylation was 1.2:1 (Ser487:Ser487, 489). There was no evidence by MS for a non-phosphorylated form of HSPDE4B2B(81-564) or HSPDE4B2B(152-528) when expressed in Sf9 cells. There was no detectable phosphorylation of purified HSPDE4B2B(152-528) expressed in Escherichia coli. Radiolabelling experiments with 32P revealed that phosphorylation of HSPDE4B2B(152-528) expressed in Sf9 cells was abolished when Ser487 and Ser489 were mutated to alanines. Analysis of the amino acid sequence revealed that Ser487 and Ser489 of HSPDE4B2B conform to the consensus motifs for phosphorylation by mitogen-activated protein kinase (MAP kinase) and casein kinase II respectively. Kinetic experiments in vitro showed that MAP kinase-phosphorylated E.coli expressed and purified HSPDE4B2B(151-528) with a K(m) of 63 microM and a Vmax of 3.0 mumol/min per mg. In comparison, MAP kinase phosphorylated myelin basic protein with a Km of 26.0 microM and a Vmax of 5.5 mumol/min per mg under the same conditions. Using MS and mutational analysis we found that MAP kinase-phosphorylated E. coli expressed HSPDE4B2B(152-528) exclusively at Ser487. These results suggest that phosphodiesterases could contribute to the cross-talk between the cAMP and MAP kinase signalling pathways.

In vitro phosphorylation of the epidermal growth factor receptor autophosphorylation domain by c-src: identification of phosphorylation sites and c-src SH2 domain binding sites.

During epidermal growth factor mediated signal transduction, intracellular receptor autophosphorylation on tyrosine residues results in the localization of several SH2 domain bearing proteins, including c-src, to the plasma membrane. This process is part of a complex pathway of specific protein associations that culminates in the regulation of cell growth and mitogenesis. The SH2 domain-mediated interaction of c-src with the EGF receptor has been demonstrated, yet the precise function of c-src in EGF receptor signaling remains unclear. The phosphorylation of EGFR by c-src was studied in order to evaluate the molecular basis for this interaction. The C-terminal autophosphorylation domain of EGFR was extensively phosphorylated by c-src and EGFR kinase activities in vitro as determined by electrospay ionization mass spectrometry. The sites of phosphorylation within the autophosphorylation domain (residues 976-1186) were identified by LC/MS, LC/MS/MS, and Edman sequencing. The majority of the sites identified corresponded to the known autophosphorylation sites of EGFR. Kinetic analyses of site-specific phosphorylation were made combining very fast enzyme digests (< = or 2 min) and high-speed, perfusion chromatography. These studies revealed that Y1086 was phosphorylated to a significantly higher extent by c-src than by EGFR. Additionally, Y1101 was identified as a unique c-src phosphorylation site. The function of these phosphorylation sites with respect SH2 domain interactions was investigated by affinity chromatography/mass spectrometry. A subset of peptides corresponding to the eight possible tyrosine phosphorylation sites within the EGFR autophosphorylation domain was demonstrated to bind to the SH2 domain of c-src. Those which bound to the SH2 domain included peptides derived from EGFR sequences flanking Y992, Y1086, Y1101, and Y1148. These data indicate that specific EGF receptor c-src phosphorylation sites are also ligands for the SH2 domain of c-src. Finally, extensive c-src phosphorylation of EGFR promoted its conversion to a form that exhibits high-affinity (KD = 380 nM) and cooperative (Hill coefficient; n = 2) binding to the SH2 domain of c-src as measured by surface plasmon resonance. The identification of c-src phosphorylation sequences on EGFR as c-src SH2 binding sites supports the notion that this interaction plays a significant role in the regulation of growth factor receptor function and signal transduction.

Exploration of the sequence specificity of pp60c-src tyrosine kinase. Minimal peptide sequence required for maximal activity.

The minimum length required for phosphorylation of a peptide by pp60c-src tyrosine kinase (srcTK) was delineated in this work. Budde (M.D. Anderson University of Texas, personal communication) suggested that the peptide (FGE)3Y(GEF)2GD (peptide I) was a "good" srcTK substrate. Peptide I yielded a 251-fold higher kcat/Km than RRLIEDAEYAARRG, a peptide substrate based upon the autophosphorylation site of srcTK. This was due to a 38-fold lower Km and a 6.6-fold increase in kcat.N-terminal truncation of up to 8 residues in a series of peptides yielded only a 3-fold decrease in activity. Removal of the final N-terminal residue resulted in a 10-fold loss in substrate activity, primarily as a result of an increase in the Km. C-terminal truncations ending in the amide yielded no significant loss in activity until the Y + 3 residue was removed, which resulted in a 73-fold decrease in kcat/Km relative to peptide I. The latter was due primarily to an increase in Km. The results from peptides truncated on both termini suggest that subsite recognition N- and C-terminal relative to the site of phosphorylation can be examined independently. In addition, the observation that only 5 residues are required for significant substrate activity suggests that small molecule inhibitors based upon interactions with the phosphoacceptor site may be developed.

Characterization of pp60c-src tyrosine kinase activities using a continuous assay: autoactivation of the enzyme is an intermolecular autophosphorylation process.

A continuous assay for pp60c-src tyrosine kinase (srcTK) was developed. A lag in phosphorylation of the peptide RRLIEDAEYAARG was observed that could be eliminated by preincubation with MgATP. The induction time for this lag was dependent upon MgATP and srcTK concentrations. When autophosphorylation was monitored by 32P incorporation from [gamma-32P]ATP, a lag in the time course was also observed. These results demonstrate that autoactivation is an intermolecular process. The electrospray ionization mass spectrum of the enzyme before and after activation demonstrated an increase in the phosphorylation state of the enzyme after incubation with MgATP. The delta 85-N-terminal mutant protein and a full-length G2A pp60c-src mutant, which removes the myristylation site, used in these studies were partially phosphorylated on Y338 and Y530 as isolated. This is the first report of phosphorylation on Y338, but the significance of this site of phosphorylation is unknown. These phosphorylations were insufficient to active the enzyme for transfer of the gamma-phosphoryl of MgATP to the peptides. The unphosphorylated enzyme initially present was converted to a monophosphorylated species upon treatment with MgATP. Y-419 phosphorylation was evident only after treatment with MgATP. These data are consistent with autophosphorylation on Y-419 as predicted. Intermolecular autophosphorylation is consistent with the ability of srcTK to dimerize, which is analogous to activation of receptor tyrosine kinases such as the EGF receptor kinase in response to growth factors. These results indicate that dimerization leading to activation does not require binding to the membrane or a hydrophobic N-terminus in the case of srcTK.

Examination of the dephosphorylation reactions catalyzed by pp60c-src tyrosine kinase explores the roles of autophosphorylation and SH2 ligand binding.

pp60c-src tyrosine kinase (srcTK) catalyzes the dephosphorylation of phosphotyrosine-containing peptides, including phosphopeptides that bind with high affinity to the src SH2 domain. The mechanism for these dephosphorylation reactions was investigated. Dephosphorylation was inhibited by a competitive inhibitor for the ATP binding site. In the presence of ADP, dephosphorylation of phosphopeptide substrates is primarily due to the reversal of the kinase reaction. Autoactivated and unactivated srcTK both catalyzed the reverse of the kinase reaction; however, autoactivated srcTK displayed an increase in kcat of approximately 4-11-fold relative to unactivated srcTK, depending on the reaction conditions. Autoactivation of srcTK does not affect the Km's for MgADP or phosphopeptide (FGE)3-pY-(GEF)2GD. Unphosphorylated srcTK becomes phosphorylated during the reverse of the kinase reaction upon accumulation of free MgATP. In the presence of MgATP, srcTK also dephosphorylates peptide substrates, by first hydrolyzing MgATP to MgADP. Binding of phosphotyrosine peptide ligands to the src SH2 domain stimulated the rate of MgATP hydrolysis approximately 2-fold, but had not effect on the Km for MgATP. These data suggest that autophosphorylation of tyrosine 419 is not required for nucleotide or peptide binding, or catalysis involving small peptide substrates. In addition, these results suggest that both the forward and the reverse src tyrosine kinase reactions may be important in regulating the intracellular levels of protein tyrosine phosphorylation.

Catalytic activity of the SH2 domain of human pp60c-src; evidence from NMR, mass spectrometry, site-directed mutagenesis and kinetic studies for an inherent phosphatase activity.

During solution structural studies it was apparent that the human recombinant pp60c-src SH2 domain (srcSH2, residues 144-249) possessed an inherent phosphatase (Pase) activity. Complexes of U[13C,15N]srcSH2 with unlabeled Ac-pYEEIE (I) were examined using 31P and 1H-detected isotope filtered NMR methods. The presence of a high-affinity complex in equimolar solutions of I and U[13C, 15N]-srcSH2 was demonstrated by chemical shift perturbations, line broadening, and the observation of intermolecular nuclear Overhauser effects from the pY and Ile side-chain protons of I to protons on amino acid residues present in the binding pocket of srcSH2. Solutions containing excess I relative to srcSH2 revealed a slow hydrolysis of I to produce Ac-YEEIE and inorganic phosphate. The hydrolysis rate determined from NMR and HPLC-electrospray ionization mass spectrometry data at 30 degrees C for solutions containing excess I was 0.002-0.003 h-1. srcSH2 also catalyzed the hydrolysis of p-nitrophenyl phosphate (pNPP). Isoelectric focusing gels of a number of mutant srcSH2s demonstrated that this activity comigrated with srcSH2. Km, kcat, and kcat/Km were 3.7 +/- 0.4 mM, 3.1 +/- 0.2 x 10(-2) min-1, and 8.4 +/- 0.4 M-1 min-1, respectively, toward pNPP. The C188A mutant of the srcSH2 domain displayed 15% of the activity displayed by wild-type srcSH2, demonstrating that this residue is not absolutely required for activity. Two additional mutations in the known pY binding site, R178K and R158K, also resulted in decreased pNPPase activity, suggesting that the activity resides in or near this site. The inhibitor profile and pH dependence suggest that this is a novel protein Pase activity.(ABSTRACT TRUNCATED AT 250 WORDS)

HIV-1 protease specificity derived from a complex mixture of synthetic substrates.

A rapid and semiquantitative method is described for determining the relative kcat/Km for individual peptides in defined substrate mixtures. The method utilizes electrospray ionization/mass spectrometry alone to semiquantitatively determine relative peptide substrate turnover rates. Unlike previous studies, in which chromatographic separation of individual peptide species was required, this mass spectrometric-based method relies strictly on the ability to ionize and detect simultaneously all peptide species in a defined mixture. Differences in the ion intensities of the individual components before and after incubation with protease are used to semiquantitatively determine preferred substrates. This method was used to the identify preferred peptide substrates for HIV-1 protease. Optimal substrates were identified from a defined synthetic peptide substrate mixture based on Ser-Gln-Asn-Tyr-Pro-Ile-Val, where the P1' proline was substituted with 20 naturally occurring amino acids. The hydrophobic residues Leu, Ile, Val, Phe, and Tyr were preferred in addition to Pro at the P1' site. The results were corroborated by performing the more laborious HPLC/Frit-fast atom bombardment/MS analyses.

Identification of phosphorylated peptides from complex mixtures using negative-ion orifice-potential stepping and capillary liquid chromatography/electrospray ionization mass spectrometry.

A rapid method for identifying and characterizing sites of phosphorylation of peptides and proteins is described. High-performance capillary liquid chromatography (HPLC) coupled with electrospray ionization mass spectrometry (ESI-MS) is used to distinguish non-phosphorylated and phosphorylated peptides originating from mixtures as complex as enzyme digests. The method relies on the ability to produce a fragment ion characteristic and unique to phosphopeptides (m/z 79, PO3) by stepping the orifice potential of the mass spectrometer as a function of mass. At low m/z values, a high orifice potential is applied to induce extensive fragmentation of the peptide, leading to the formation of the m/z 79 phosphate-derived ion. This method is analogous to that described by Carr et al. for the identification of glycopeptides from enzymatic digestion of glycoproteins (S.A. Carr, M.J. Huddleston, M.F. Bean, Protein Science 2, 183 (1993)). The method was first evaluated and validated for a mixture of non-, mono- and di-phosphorylated synthetic peptides. Both mono- and di-phosphorylated peptides were found to generate fragment ions characteristic of PO3 whereas the non-phosphorylated peptide did not. Application of the method was extended to identifying phosphopeptides generated from an endoprotease Lys-C digestion of beta-casein. Both the expected mono- and tetra-phosphorylated Lys-C peptides were observed and identified rapidly in the LC/SEI-MS analysis. The procedure was used additionally to identify the site(s) of phosphorylation of the cytosolic non-receptor tyrosine kinase, pp60(c-src).

Mercury in a commercial preparation of rat amylin.

The biological activity of amylin is reported to vary widely depending on the source and purity of the material. Three commercial samples of rat amylin were compared for structural differences. The samples were nearly identical using most of the available analytical measures--amino acid analysis, HPLC retention, even Edman sequencing data. When the samples were compared by ion spray ionization mass spectrometry, the molecular mass of one sample was 200 daltons higher than anticipated. Careful analysis of the sample, including atomic emission spectrometry, revealed that a mercury atom was associated with the polypeptide. The mercury presumably resulted from a deprotection step in the synthesis, involving the removal of an acetamidomethyl group from cysteine.

Electrospray ionization mass spectra of hemoglobin and transferrin by a magnetic sector mass spectrometer. Comparison with theoretical isotopic distributions.

Electrospray ionization mass spectra of human hemoglobin chains and of transferrin were acquired on a magnetic sector mass spectrometer. The observed molecular ion for each hemoglobin chain was in good agreement with the theoretical isotopic distribution at a reasonable resolution of 2000. The clear separation of a variant beta-chain in admixture with the normal counterpart at mass 15,867 that differed from it by 14 Da (0.09%) ensured that a smaller mass difference could be detected. The molecular ions for human transferrin were too broad compared with the theoretical shape to determine the molecular mass accurately, probably due to the heterogeneity of the carbohydrate moiety. A decrease in mass by neuraminidase digestion, however, determined the average number of sialic acids in the molecule.

Optimization of the fragmentation in a frit-fast atom bombardment ion source for the sequencing of peptides at the picomole level.

Peptides derived from enzymatic digestions (cathepsin D and trypsin) were characterized and amino acid sequences determined by using their LC/MS spectra. A Frit-FAB interface that produces extensive peptide fragmentation and permits amino acid sequencing at the low picomole level is described for a model antigen, Staphylococcus aureus nuclease (Nase), and an enzyme of unknown structure, yeast aminopeptidase B. The amino acid sequences of peptides derived from digestion of Nase with cathepsin D (a relatively nonspecific endoprotease) were readily deduced and have provided insights into the nature of antigen processing. Frit-FAB LC/MS spectra of the Nase peptides contained a sufficient number of fragment ions to conclusively identify peptides with a mass below 2000 Da. Capillary LC/MS provided a means for the separation and identification of these enzymatically derived peptides in a fraction of the time that would have been required by gas-phase Edman sequence analysis. The optimized Frit-FAB experiment was consequently evaluated for the partial characterization of aminopeptidase B recently purified to homogeneity from Saccharomyces cerevisiae. Sequence-specific ions observed in the Frit-FAB mass spectra of these tryptic peptides were identical with those commonly observed in high-energy collision-induced dissociation (CID) spectra and included side-chain fragment ions that differentiated leucine from isoleucine. These fragment ions were used to deduce entire amino acid sequences for several of the tryptic peptides.

Primary structure determination of peptides and enzymatically digested proteins using capillary liquid chromatography/mass spectrometry and rapid linked-scan techniques.

Primary protein sequences were determined for both peptides and enzymatically digested proteins by rapid linked-scan (B/E) liquid chromatography/mass spectrometry/mass spectrometry (LC/MS/MS) at the low-picomole level (10-50 pmol). During the course of a single LC/MS/MS analysis, we demonstrated that it is possible to generate interpretable collision-induced dissociation spectra of the eluting proteolytic peptides. Molecular weights of tryptic peptides were established by using 1/10 of the protein digest by operating in the capillary LC/frit-FABMS mode. Peptides exhibiting the strongest MH+ ions were then selected for subsequent LC/MS/MS analysis (typically 1/5 of the remaining protein digest). Elution times for each chromatographic peak were generally greater than 30 s. It was therefore possible to obtain a minimum of six B/E fast linked-scan spectra during the course of elution of each peptide component. Typically, B/E linked scans of the greatest ion abundance (obtained at the chromatographic peak maximum) were averaged to enhance the signal/noise ratio at these low-picomole levels. Unit resolution was observed for product ions below m/z 1000. Rapid linked scanning by LC/frit-FABMS/MS provided mass assignments for product ions within 0.2-0.3 amu of theoretical values. Side-chain fragment ions (wn and dn) were also observed, which allowed for the differentiation of isobaric amino acids (e.g., leucine and isoleucine). Examples of the application of this fast linked-scan technique to LC/MS/MS are presented for complex mixtures of unknown peptides and the tryptic digestion of phosphorylated beta-casein.