Anal. Chem. 2000, 72, 4193-4198
Combination of Biomolecular Interaction Analysis and Mass Spectrometric Amino Acid Sequencing Tohru Natsume,*,† Hiroshi Nakayama,‡ O 2 sten Jansson,§ Toshiaki Isobe,| Koji Takio,‡ and †,⊥ Katsuhiko Mikoshiba
Calciosignal Net Project, Exploratory Research for Advanced Technology (ERATO), JST., c/o RIKEN Komagome Branch 2-28-8, Honkomagome, Bunkyo-ku, Tokyo 113-0021, Japan, RIKEN (The Institute of Physical and Chemical Research), 2-1, Hirosawa, Wako, Saitama 351-0198, Japan, Biacore AB, Rapsgartan 7, S-754 50 Uppsala, Sweden, Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University, 1-1, Minami-osawa, Hachioji, Tokyo 192-0397, Japan, and Department of Molecular Neurobiology, Institute of Medical Science, The University of Tokyo, 4-6-1, Shirokanedai, Minato-ku, Tokyo 108-8639, Japan.
We describe an approach for the combination of biomolecular interaction analysis (BIA) and electrospray tandem mass spectrometry (ESI/MS/MS) to obtain sequence information on the affinity-bound proteins on the sensor chip of BIA. The procedure is illustrated with stable and unstable interactions of recombinant proteins, i.e., histidine-tagged protein-Ni2+/NTA and 1,4,5-inositol trisphosphate receptor-ligand interactions. The E. coli lysates expressing the recombinant proteins were passed through the sensor chips, and biomolecular interactions were monitored in real time. The molecules detected on the sensor chip were digested by delivering proteolytic enzyme to the sensing flow cells. The resulting on-chip digested peptide mixture at the mid- to low-femtomole level was recovered on a microcapillary reversed-phase precolumn by an on-line system and analyzed using HPLC-MS/MS. In both cases, unambiguous sequence information on the recombinant proteins isolated on the sensor chip was obtained from only a single run of analysis. The combined BIA-MS/MS may prove to be a general and versatile system to discover novel biomolecular interactions and to analyze protein complexes. Biomolecular interaction analysis (BIA), a biosensor technology based on the principle of surface plasmon resonance, has become one of the important approaches used to understand the function of molecules in a wide biological field 1. BIA has been used to study protein-protein, protein-nucleic acid and nucleic acid-nucleic acid interactions as well as the low molecular weight substances binding to proteins or nucleic acids.1,2 This analytical * Corresponding author: Tohru Natsume, Ph.D., Integrated proteomics system project, Pioneer Research on Genome the Frontier, Science and Technology Agency, c/o Department of Chemistry Graduate School of Science, Tokyo Metropolitan University, Minami-ohsawa 1-1, Hachiohji city, Tokyo, 1920397 Japan. Tel.: +81-426-77-2452. Fax: +81-426-77-2525. E-mail: natsume@ comp.metro-u.ac.jp † Calciosignal Net Project. ‡ RIKEN (The Institute of Physical and Chemical Research). § Biacore AB. | Tokyo Metropolitan University. ⊥ The University of Tokyo. (1) Malmqvist, M.; Karlsson, R. Curr. Opin. Chem. Biol. 1997, 1, 378-83. 10.1021/ac000167a CCC: $19.00 Published on Web 08/01/2000
© 2000 American Chemical Society
technique can provide quantitative information, such as kinetic parameters and equilibrium constants for complex formation.3-6 Additional uses of the technology are the discovery of novel molecular interactions 7-10 and screening of small molecules, as potential therapeutics.11,12 In such experiments, molecules of interest, or orphan receptors, are immobilized on sensing flow cells formed on the sensor chip surface; then biological mixtures or fractions of liquid chromatography are passed over the sensor chip through the microfluidics system and specific binding activity is monitored in real time.7,9,10 The microfluidics is operated automatically by micro pumps and an autosampler;13 therefore, high-throughput screening can be performed, with high reproducibility. Once the fraction containing binding activity is determined, the conventional next step to identification of the binding molecule is a time- and resource-consuming process to scale up starting material, followed by large-volume affinity chromatography. This would often offset the efficiency of high-throughput analysis by (2) Szabo, A.; Stolz, L.; Granzow, R. Curr. Opin. Struct. Biol. 1995, 5, 699705. (3) Natsume, T.; Koide, T.; Yokota, S.; Hirayoshi, K.; Nagata, K. J. Biol. Chem. 1994, 269, 31224-8. (4) Natsume, T.; Tomita, S.; Iemura, S.; Kinto, N.; Yamaguchi, A.; Ueno, N. J. Biol. Chem. 1997, 272, 11535-40. (5) Karlsson, R.; Roos, H.; Fa¨gerstam, L.; Persson, B. Methods; Companion Methods Enzymol. 1994, 6, 99-110. (6) Johne, B.; Gadnell, M.; Hansen, K. J. Immunol. Methods 1993, 160, 1918. (7) Lackmann, M.; Bucci, T.; Mann, R. J.; Kravets, L. A.; Viney, E.; Smith, F.; Moritz, R. L.; Carter, W.; Simpson, R. J.; Nicola, N. A.; Mackwell, K.; Nice, E. C.; Wilks, A. F.; Boyd, A. W. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 2523-7. (8) Sakano, S.; Serizawa, R.; Inada, T.; Iwama, A.; Itoh, A.; Kato, C.; Shimizu, Y.; Shinkai, F.; Shimizu, R.; Kondo, S.; Ohno, M.; Suda, T. Oncogene 1996, 13, 813-22. (9) Seok, Y. J.; Sondej, M.; Badawi, P.; Lewis, M. S.; Briggs, M. C.; Jaffe, H.; Peterkofsky, A. J. Biol. Chem. 1997, 272, 26511-21. (10) Stitt, T. N.; Conn, G.; Gore, M.; Lai, C.; Bruno, J.; Radziejewski, C.; Mattsson, K.; Fisher, J.; Gies, D. R.; Jones, P. F.; Masiakowaski, P.; Ryan, T. E.; Tobkes, M. J.; Chen, D. H.; DiStefano, P. S.; Long, G. L.; Basilico, C.; Goldofarb, M. P.; Lemke, G.; Glass, D. J.; Yancopoulos, G. D. Cell 1995, 80, 661-70. (11) Gram, H.; Schmitz, R.; Zuber, J. F.; Baumann, G. Eur. J. Biochem. 1997, 246, 633-7. (12) Markgren, P. O.; Hamalainen, M.; Danielson, U. H. Anal. Biochem. 1998, 265, 340-50. (13) Josson, U. BioTechniques 1991, 11, 620-627.
Analytical Chemistry, Vol. 72, No. 17, September 1, 2000 4193
BIA. In some cases, it is impossible to obtain enough starting material and to purify target molecules in sufficient quantity for amino acid sequencing, especially when the binding activity is discovered in biological fluids which are available in limited quantities, such as Xenopus leaevis oocytes. To avoid the purification scale-up and accelerate screening processes, other investigators combined BIA with matrixassisted laser desorption/ionization time-of-flight mass spectrometry (MALDI/TOF) to determine the molecular mass of proteins retained on the sensor chip surface.14-16 However, determination of the intact molecular mass by MALDI/TOF analysis cannot be considered as a general method for identification of proteins, for the following reasons: First, the mass accuracy of MALDI/TOF is in the range of 10-50 ppm for small polypeptides,17 enabling direct identification, but for proteins over 30 kDa, the mass accuracy frequently drops to >0.1%, resulting in obscured identification. Second, proteins are often modified posttranslationally (e.g., enzymatic cleavage, glycosylation, phosphorylation, acetylation, methylation) and consequently all such modifications lead to change in the molecular mass. Therefore, identification of proteins by MALDI/TOF is generally made by peptide mass fingerprinting, based on database search.18 To achieve this, bound proteins would be eluted from the sensor chip and separated by electrophoresis followed by in-gel proteolytic digestion for subsequent peptide mass fingerprinting. An alternative, sensitive method for protein identification is mass spectrometric sequencing, using tandem mass spectrometry (MS/MS).18,19 In the most sensitive MS/MS analysis, using an electrospray ion trap mass spectrometer, unambiguous protein identification can be obtained at the 100 femtomole level.20,21 This is the same order of magnitude as the binding capacity of the sensor chip in typical BIA analysis. However, elution and transfer of such small quantities of samples from the BIA instrument to the ion source of mass spectrometer might be not feasible. Furthermore, mass spectrometric techniques can develop complete sequence information only for small peptides; therefore, proteins bound to the sensor chip should be enzymatically digested after the elution. This process also leads to a critical sample loss. We have now developed a novel system to combine BIA with MS/MS to sequence proteins bound to the sensor chip surface. These proteins are enzymatically digested on the sensor chip by delivering proteolytic enzyme to the sensing flow cell, through the microfluidics system. After the ‘on-chip digestion’ to minimize the risk of sample loss, the resulting peptide mixture is trapped in the RP capillary precolumn by the ‘on-line recovery’ technique for subsequent MS/MS analysis (Figure 1). To study the feasibility of using our combined system, epitopetagged recombinant protein and 1,4,5-inositol trisphosphate (IP3) (14) Nelson, R. W.; Krone, J. R.; Jansson, O. Anal. Chem 1997, 69, 4369-74. (15) Nelson, R. W.; Krone, J. R.; Jansson, O. Anal. Chem 1997, 69, 4363-8. (16) Sonksen, C. P.; Nordhoff, E.; Jansson, O.; Malmqvist, M.; Roepstorff, P. Anal. Chem 1998, 70, 2731-6. (17) Jensen, O. N.; Podtelejnikov, A.; Mann, M. Rapid Commun. Mass Spectrom. 1996, 10, 1371-8. (18) Yates, J. R., 3rd J. Mass Spectrom. 1998, 33, 1-19. (19) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Anal. Chem 1996, 68, 8508. (20) Wilm, M.; Shevchenko, A.; Houthaeve, T.; Breit, S.; Schweigerer, L.; Fotsis, T.; Mann, M. Nature (London) 1996, 379, 466-9. (21) Wilm, M.; Mann, M. Anal. Chem. 1996, 68, 1-8.
4194 Analytical Chemistry, Vol. 72, No. 17, September 1, 2000
Figure 1. General scheme of the on-line BIA-ESI/MS/MS system. Target proteins present in the biological mixture are monitored and isolated on sensor chips of BIA, after which the captured molecules are digested on a sensor chip and recovered in an on-line RP capillary precolumn. After the recovery, the precolumn is removed from the BIA instrument and subjected to HPLC-ESI/MS/MS analysis. Resulting recovered peptides are further separated on the microESI column, before the peptide mixture is sprayed.
binding protein present in the total cell lysates were isolated on the sensor chip and sequenced by MS/MS analysis. Our results show that unambiguous sequence information of proteins captured on the sensor chip at the mid- to low-femtomole level was obtained even with only by a single run of the BIA-MS/MS analysis. EXPERIMENTAL SECTION Small-Scale Expression of Recombinant Proteins in E. coli. To express the Xenopus FK506-binding protein (xFKBP), full length cDNA of xFKBP (accession AB006678) was amplified from Xenopus oocyte mRNA by RT-PCR, using a sense oligonucleotide (5′-GGAATTCCACCATGGATTACAAGGTATGACGACGATAAGGGAGTGCAAGTAGAAACCATT-3′) and an antisenseoligonucleotide (5′-GGAATTCCTCACTCCAGCCTCAGTAGCTC3′) based on the published sequence.22 The PCR product was subcloned into pRSET vector (Invitrogen) to create an in-frame fusion with the hexa-histidine sequence tag at the amino terminus. The ligand binding domain of 1,4,5-inositol phosphate receptor type 1 (IP3R1), a truncated mutant of cDNA that codes N-terminal 604 amino acids of mouse IP3R1,23 was cloned to the pET3a vector (Novagen). Escherichia coli BL21(DE3) was transformed with the pRSETHis-xFKBP or pET3a-T604. A single colony of each transformed cell was placed in 1 mL of Luria-Bertani medium containing 100 µg/mL ampicillin and incubated at 37 °C to an A600 of ∼1.6. The cells were then harvested by centrifugation and stored at -80 °C for subsequent use. The harvested cell pellet was resuspended in 500 µL of Tris buffer (50 mM Tris-HCl, 1 mM 2-mercaptoethanol, 1 mM EDTA, and 0.4% CHAPS, pH 7.4) containing protease inhibitors (10 µM pepstatin A, 10 µM leupeptin, and 0.6 mM phenylmethylsulfonyl fluoride) and lysed using a sonicator (Strason XL2020).23 The suspension was centrifuged at 30 000g for 20 min at 4 °C. Note that both recombinant proteins were produced in the absence of an inducer, such as IPTG, to test the developed system (22) Nishinakamura, R.; Matsumoto, Y.; Uochi, T.; Asashima, M.; Yokota, T. Biochem. Biophys. Res. Commun. 1997, 239, 585-91. (23) Natsume, T.; Hirota, J.; Yoshikawa, F.; Furuichi, T.; Mikoshiba, K. Biochem. Biophys. Res. Commun. 1999, 260, 527-33.
Figure 2. (A) Principle of the on-chip digestion; 3 µL of proteolytic digestion solution is delivered to the sensing flow cells, following a wash with distilled water. Then the flow is halted to allow for digestion, without dispersion and dilution. This process is monitored in real time, as the digestion curve. (B) Typical on-chip digestion curve. After the completion of digestion, the flow is restarted to recover the resulting peptide mixture.
under more restrictive conditions. By SDS-PAGE analysis, bands of recombinant proteins were not specified as being prominent, in both cases. Western blotting detected the corresponding bands, suggesting that the low concentration of the recombinant proteins (50-200 ng/mL) was produced at the basal expression level (data not shown). Surface Plasmon Resonance Biomolecular Interaction Analysis. For all analyses we used a Biacore X instrument (Biacore AB., Uppsala, Sweden). HEPES buffer (50 mM HEPES, 150 mM NaCl, 50 µM EDTA, 0.005% n-octylglycopyranoside, pH 7.4), as running buffer, flowed serially through two flow cells. To prepare a metal chelating sensor surface, a nitrilotriacetic acid immobilized sensor chip (Sensor chip NTA, Biacore AB., Uppsala, Sweden) was exposed to nickel solution (0.1 mM NiCl2, 50 mM HEPES, 150 mM NaCl, 50 µM EDTA, pH 7.4) for 1-2 min at a flow rate of 20 µL/min, following the injection of 20 µL of regeneration buffer (50 mM HEPES, 150 mM NaCl, 350 mM EDTA, pH 7.4) at the same flow rate to remove any metal ions chelated to the surface.24 For IP3 binding protein, the IP3 immobilized sensor chip surface was prepared, as described.23 Briefly, 100 ng/mL of a biotinylated (1,4,5)IP3 analogue in HEPES buffer (50 mM HEPES, 150 mM NaCl, 3.4 mM EDTA, pH 7.4) was injected over the streptavidin covalently immobilized sensor chip surface (Sensor Chip SA, Biacore AB., Uppsala, Sweden), and 150 RU of immobilization was observed. On-Chip Digestion and On-line Recovery System. After BIA analysis, the RP capillary column (Inertsil ODS-3, 3 µm, 1 × 0.3 mm i.d., GL Science, Tokyo) was connected to a flow cell channel of the Biacore X instrument via an in-house fabricated “flow cell out” port. To digest the bound protein on the sensor chip, proteolytic enzyme was delivered to flow cells by an air (24) Gershon, P. D.; Khilko, S. J. Immunol. Methods 1995, 183, 65-76.
partition method that prevents the enzyme from being diluted during the incubation and minimizes dispersion of the digested peptides in the flow-cell channel. To achieve this method, we modified previous work.16 Namely, a 100-µL pipet tip was prepared containing 20 µL of distilled water, 5 µL of air, 3 µL of Achromobacter lyticus protease I 25 (40 pM) dissolved in Tris buffer (50 mM Tris, 2 M urea, 0.005% n-octylglycopyranoside, pH 9.0), and 10 µL of air. Contents of the pipet tip passed over the sensor chip via a sample loop of the instrument in the arranged order (Figure 2A). Immediately after the first air passed over the flow cell (indicated by sensorgrams), continuous flow was halted to allow for digestion at 37 °C. As digestion of the bound protein on the sensor chip progressed, the response unit of the sensorgram gradually lowered (Figure 2B). After the sensorgram stabilized, the flow was restarted and the digested peptides were collected in the RP precolumn. Mass Spectrometric Analysis. The molecular masses and amino acid sequences of the on-chip-digested peptides were determined, using an HPLC-ESI/MS/MS system. The system illustrated in Figure 1 consisted of a high-performance liquid chromatograph (model 140D syringe pump, PE Biosystems, CA) with a flow split system and a Finnigan LCQ ion trap mass spectrometer (ThermoQuest, San Jose, CA) equipped with a microelectrospray interface (ESI) column. The micro ESI column was constructed as in an earlier study,26 but with modifications. Briefly, a fused-silica capillary (75 µm i.d. × 375 µm o.d.) was pulled by heating. A single POROS R2 bead (particle size, 50 µm, PE Biosystems) was placed at the tip of the pulled capillary as “particle frit”, and reversed-phase material (Mightysil C18, particle (25) Masaki, T.; Tanabe, M.; Nakamura, K.; Sejima, M. Biochem. Biophys. Acta 1981, 660, 44-50. (26) Gatlin, C. L.; Kleemann, G. R.; Hays, L. G.; Link, A. J.; Yates, J. R., 3rd Anal. Biochem 1998, 263, 93-101.
Analytical Chemistry, Vol. 72, No. 17, September 1, 2000
4195
Figure 3. Detection and capturing of histidine-tagged protein present in total cell lysate; 5 µL of E. coli lysate expressing xFKBP was injected over the Ni2+/NTA sensor chip surface at a flow rate of 5 µL/min for 60 s, following the injection of EDTA and Ni2+ solution. Change in the baseline after the injection corresponds to the amount of protein bound to the sensor surface of each flow cell (arrow).
Figure 4. ESI/MS/MS spectrum of on-chip digested histidine-tagged protein. The doubly charged ion at m/z ) 844.3 eluted from the RP capillary column (inset) was fragmented in the ion trap cell. The fragmentation produced ion series containing the C terminus and N terminus (designated y and b series, respectively). The set of ion series yielded the amino acid sequence of GQTVVVHYVGSLENGK, which corresponds to a peptide of xFKBP.
size; 1 µm, Kanto Chemical) was packed to a length of 2 cm. The A solvent was 2% acetonitrile/0.1% formic acid, and the B solvent was 80% acetonitrile/0.1% formic acid. The peptide mixtures trapped in the microcapillary precolumn were washed with 1% B solution and then connected to a micro ESI column. The two tandemly connected columns were then eluted with a 30-min linear gradient to 60% B solution at a flow rate of 500 nL/min, and the eluate was directly introduced to the mass spectrometer. During an HPLC run, if an ion was present in a scan, above a specified threshold, the mass spectrometer continued to automatically alternate between MS and MS/MS mode so that the molecular masses and the amino acid sequences of the peptides were obtained in a single HPLC-ESI/MS/MS run.
from the RP capillary column. The peptide ion at m/z ) 844.3 was fragmented in the ion trap cell and sequenced by MS/MS analysis. As shown in Figure 4, nine y and seven b series of fragment ions showed the peptide sequence G76QTVVVHYVGSLENGK91 of xFKBP. Two other doubly charged ions (m/z ) 918.0 and 1344.6) were also fragmented and identified as peptides of G58VQVETITEGDGRTFPK74 and D32RWIRPRDLQLVPWNSTMDYK52 of the same protein, respectively. Each fragmented ion spectrum of the peptide qualified solely for identification of the original protein. The sequence coverage of the observed peptides in the original protein was 33% (54/164 amino acids). In addition to these peptide ions, one relatively weak ion at molecular weight 8606, which did not match any of the complete and incomplete digested peptide of xFKBP, was observed (data not shown). This ion signal is presumably due to the digested fragment of endogenous E. coli protein. No other peptide was eluted from the capillary column. These results clearly demonstrate that xFKBP present in cell lysate was purified on the Ni2+/NTA sensor surface via the histidine tag and that the purified protein was effectively digested on the sensor chip. The resulting on-chip digested peptides were recovered from the sensor chip, using the on-line recovery system, and unambiguous amino acid sequences were obtained. The response unit (1 kRU) of the sensorgram just prior to digestion of the bound protein corresponds to a total of 2 ng of protein (1 ng × 2 flow cells), based on the assumption that the response unit derived only from the histidine-tagged protein.13 The amount of the protein digested on the sensor chip was calculated to be ∼160 fmol, as based on the molecular weight of xFKBP (12 kDa). The elution protocol of the proteins from the sensor chip has been reported.16 Using this off-line elution method instead of the on-line system described here, we could not recover an adequate amount of peptides for MS/MS analysis from the sensor chip on which the same amount of xFKBP was captured (data not shown). The system buffer used in this study (running buffer, NiCl2 solution, regeneration solution, and digestion buffer) contained detergent, n-octylglycopyranoside, instead of surfactant P20, which
RESULTS AND DISCUSSION Detection of Histidine-Tagged Protein Present in Total Cell Lysates. We cultured E. coli cells transformed with pRSETxFKBP in a small volume (1 mL), without a general induction procedure. To detect recombinant protein produced at the basal expression level, 5 µL of E. coli cell lysate was injected, for 1 min, over the Ni2+ chelated sensor chip NTA. The difference in response unit after the injection was 1 kRU in each flow cell (Figure 3). To digest the molecule bound to the Ni2+/NTA surface, Achromobacter protease I was delivered to the two flow cells, as described in the Experimental Section, and the flow of running buffer was halted to allow for a prolonged digestion reaction. Immediately after halting the flow, the response unit gradually began to lower (Figure 2B), indicating the progress of enzymatic cleavage of the bound molecule on the sensor surface. After the response stabilized (digestion for 130 min), thereby suggesting completion of the enzymatic reaction, the flow was restarted and the digested peptides were recovered in the reversed-phase (RP) capillary precolumn, by the on-line method. The column was then removed from the biosensor instrument and washed twice with 5% B solution for use in the following HPLC-ESI/MS/MS analysis. The HPLC-ESI/MS/MS study showed three intense ions at m/z ) 844.3, 918.0, and 1344.6, corresponding to peptides eluted 4196 Analytical Chemistry, Vol. 72, No. 17, September 1, 2000
Figure 5. Detection and capturing of IP3 binding protein present in the total cell lysate; 5 µL of E. coli lysate expressing the ligand binding domain of IP3R (T604) was injected over the IP3 immobilized sensor chip at a flow rate of 5 µL/min for 60 s. After the injection, a biphasic dissociation curve was evidentsthere was a rapid dissociation phase for 10 s followed by a slower dissociation one. The residual species in the second dissociation phase was subjected to on-chip digestion and subsequent ESI/MS/MS analysis.
is a standard component of well-established conditions for BIA analysis. When the surfactant P20 was present in the buffer system, the ions derived from digested peptides were hardly detectable because multiple intense ions derived from the surfactant were present. However, the recovery rate decreased drastically in the absence of detergent (data not shown). Hence, a detergent that is compatible for use in mass spectrometry, such as n-octylglycopyranoside, seems necessary to inhibit peptide adsorption to microfluidics channels and to increase recovery rates. The optimal concentration of the detergent is currently under investigation; the higher concentration of n-octylglycopyranoside (>0.05%) has adverse effects on ionization of the peptide. Detection of IP3 Binding Protein Present in Total Cell Lysates. Next, we applied our system to an unstable binding reactionsinteraction between the ligand binding domain of IP3R1 and its ligand. We earlier found that N-terminal 604 amino acid residues of IP3R1 (T604) expressed in E. coli are capable of binding to its ligand.23 Kinetic analysis showed that the dissociation rate constant of T604 was 1 s-1, indicating that the half-time of dissociation from the ligand is 700 ms and all of the once bound protein is dissociated from the surface within 5 s. However, on the high-capacity ligand surface, the apparent dissociation rate of T604 was slower than the actual chemical rate because of “rebinding” or the “parking effect”, which is attributed to mass transport limitation27 (The IP3 sensor chip used in our experiment has a “high-capacity” surface). Figure 5 shows the interaction between T604 present in the E. coli lysate and IP3 immobilized on the sensor chip surface. After injection of the lysate, dissociation of T604 from the surface was rapid, and about half of the amount of bound T604 dissociated rapidly, but residual species showed a slower dissociation rate, as reported.23 The molecules (27) Schuck, P.; Minton, A. P. Anal. Biochem. 1996, 240, 262-72.
Figure 6. ESI/MS/MS spectra of the on-chip digested IP3 binding protein. The doubly charged peptide ion at m/z ) 701.1 eluted from the RP column (inset) was fragmented and sequenced. An unambiguous amino acid sequence (PGANSTTDAVLLNK) was determined by the eleven y and eight b series of fragmented ions.
retained on the sensor surface were subjected to on-chip digestion for MS/MS analysis, similar to analysis of histidine tagged xFKBP, as described above. In the HPLC-ESI/MS/MS analysis, five intense peptide ions were detected and fragmented for sequence analysis. A doubly charged peptide ion at m/z 701.1 was identified to be P78GANSTTDAVLLNK91 of T604 (Figure 6). In this MS/MS analysis, all expected y series were confirmed, except for y1 and y2 that seemed to be too small to be retained in the ion trap, and eight b series out of eleven were also observed to be major peaks. In addition, clear sequence information was obtained from four other peptide ions which were identified to be digested fragments of T604. The consequent sequence coverage was 96/604 amino acids (data not shown). The background ion at m/z ) 8606 was again observed in this experiment. The response unit of a single flow cell just prior to digestion was 1800 RU (Figure 5), indicating that the total of ∼58 fmol of T604 was retained on the two flow cells, based on the assumption that the response was produced only by T604. The amount of resulting digested peptides was estimated to be
