In-Sample Calibration Curve (ISCC) Using Multiple Isotopologue

Subscriber access provided by UNIV OF DURHAM

Article

In-Sample Calibration Curve (ISCC) Using Multiple Isotopologue Reaction Monitoring (MIRM) of a Stable Isotopically Labeled Analyte for Instant LC-MS/MS Bioanalysis and Quantitative Proteomics Huidong Gu, Yue Zhao, Marissa DeMichele, Naiyu Zheng, Yan J. Zhang, Renuka Pillutla, and Jianing Zeng Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b05656 • Publication Date (Web): 07 Jan 2019 Downloaded from http://pubs.acs.org on January 10, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

In-Sample Calibration Curve (ISCC) Using Multiple Isotopologue Reaction Monitoring (MIRM) of a Stable Isotopically Labeled Analyte for Instant LCMS/MS Bioanalysis and Quantitative Proteomics

Huidong Gu*, Yue Zhao, Marissa DeMichele, Naiyu Zheng, Yan Zhang, Renuka Pillutla, Jianing Zeng*

Bioanalytical Sciences, Research & Development, Bristol-Myers Squibb Route 206 & Province Line Road, Princeton, NJ 08543

Corresponding authors: Huidong Gu, Tel.: 609-252-7636, e-mail address: [email protected] Jianing Zeng, Tel: 609-252-5669, e-mail address: [email protected] *

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT A novel methodology of In-Sample Calibration Curves (ISCC) using Multiple Isotopologue Reaction Monitoring (MIRM) of multiple naturally occurring isotopologue transitions of a stable isotopically labeled (SIL) analyte for instant liquid chromatography-tandem mass spectrometry (LC-MS/MS) bioanalysis of biomarker, biotherapeutics and small molecule compounds is proposed and demonstrated for the first time. The theoretical isotopic abundances of the SIL analyte in its MIRM channels can be accurately calculated based on the isotopic distributions of its daughter ion and neutral loss. The isotopic abundances in these MIRM channels can also be accurately measured with a triple quadrupole mass spectrometer. By spiking a known amount of a SIL analyte into each study sample, an ISCC can be established based on the relationship between the calculated theoretical isotopic abundances (analyte concentration equivalents) in the selected MIRM channels of the SIL analyte and the measured MS/MS peak areas in the corresponding MIRM channels in each individual study sample. The analyte concentration of each study sample can then be calculated individually with the ISCC instantly without using an external calibration curve. The MIRM-ISCC-LC-MS/MS methodology was evaluated and demonstrated in this work with the examples of quantitation of a protein biomarker in human and monkey serum processed with immuno-capture and trypsin digestion; three surrogate peptides in trypsin digested human colon tissue homogenates; and a small molecule drug in human and rat plasma extracted with liquid-liquid extraction. The potential applications of the MIRM-ISCCLC-MS/MS methodology in quantitative proteomics, clinical laboratories and other areas are also discussed in this paper. Without the need of using external calibration curves, this novel MIRM-ISCC-LC-MS/MS methodology can provide accurate and reliable bioanalysis in many


Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


potential applications, especially for cases where authentic matrices for external calibration curves are not available.

Key Words: Multiple Isotopologue Reaction Monitoring (MIRM) In-Sample Calibration Curve (ISCC) Stable isotopically labeled (SIL) Isotopic abundance in MIRM channel LC-MS/MS Quantitative proteomics Biomarker



INTRODUCTION

With the increasing needs in translational medicine research, quantitative determination of biomarkers as well as quantitative proteomics in pre-clinical and clinical studies have been playing more and more important roles in drug discovery and development, such as identifying and validating potential biomarkers, facilitating patient stratification and dose selection, serving as surrogate endpoints and establishing PK/PD relationship at the site of action.1-4 The use of external calibration curves prepared in the same biological matrices as the incurred study samples, and the use of stable isotopically labeled (SIL) internal standards (IS) in both of the external calibration curves and incurred study samples are the keys in developing accurate and robust liquid chromatography-tandem mass spectrometry (LC-MS/MS) assays for the quantitative determination of small molecule drugs and biotherapeutics in biological matrices.5-8 However, for biomarker measurement in pre-clinical and clinical studies, due to biomarkers’ endogenous nature, the use of external calibration curves with authentic reference standards in authentic matrices is not possible in many cases. Several alternatives for the external calibration curve preparation were reported.9 Preparing an external calibration curve with a SIL surrogate analyte in authentic matrix is one of the alternatives. This option could provide the same assay performances as those assays using authentic analytes in authentic matrices since the SIL surrogate analyte and the authentic analyte have the “identical” physicochemical properties regarding sample extraction, chromatographic separation, MS ionization, fragmentation and detection. However, this approach is not easily available as another version of the SIL analyte is needed for internal standardization, and the access to two versions of the SIL analyte is very costly and time consuming, especially for the analysis of proteins.5, 10


Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Significant progresses have been made over the last decade in the area of stable isotope quantitative proteomics by using different stable isotopic labeling procedures,1 such as isotopecode affinity tag (ICAT),11 isotope-coded protein labeling (ICPL),12 isotope-differentiated binding energy shift tag (IDBEST),13-14 tandem mass tag (TMT),15 isobaric peptide termini labeling (IPTL)16 and stable isotope labeling with amino acids in cell culture (SILAC).17-19 The overall workflows are very similar for quantitative proteomics using different stable isotope labeling approaches. By tagging proteins or peptides using mass tag reagents with different numbers of stable isotope labels (different masses) on the tags for different samples, the labeled samples are then mixed and analyzed by mass spectrometry, and the relative abundances of the protein or peptide in the original samples could be obtained by comparing the peak areas corresponding to the different mass tags. The major drawbacks for this methodology are that it can only provide relative quantitation within the same analysis batch and its throughput is limited by the number of different tags available. Absolute quantitation in LC-MS proteomics with isotope dilution principle20 was achieved by spiking a known amount of a SIL peptide (AQUA approach21) or SIL protein (PSAQ approach22) into each study sample. The peptide (or protein) concentration of a study sample could be calculated using the ratio between the peak area of the non-labeled peptide (or protein) and the peak area of the labeled peptide (or protein). However, this quantitation approach is based on only one calibration point with an assumption that a linear relationship passing through the point of origin exists between the MS responses (or response ratios) and the corresponding concentrations.21

Recently, to address this issue, Chiva et al. reported23 that an accurate and robust assay was developed by spiking a calibration curve into each study sample for the absolute quantitation of a



targeted peptide in Formalin Fixed Paraffin Embedded (FFPE) samples. This calibration curve was pre-prepared using five different SIL peptide analytes (with the total labeling positions of 10, 16, 23, 33 and 39 for each SIL peptide analyte, respectively) at different concentration levels from 2 to 200 fmol. However, as multiple differently labeled peptide analytes with as many as 30 to 40 labeling positions are needed for the measurement of one targeted peptide, this approach is very costly and time consuming especially for quantitative proteomics of multiple targeted peptides.

Previously, we reported and demonstrated a methodology to accurately calculate the isotopic interferences from an analyte to its adjacent selective reaction monitoring (SRM) channels in LC-MS/MS assays.24 In that work, this methodology was used successfully to design stable isotopic labeling schema (labeling positions and number of labels etc.) for both microdosed IV drugs and assay internal standards in order to mitigate the “unwanted” isotopic interferences for several microdosing absolute bioavailability studies. In this paper, however, we will report and demonstrate a completely new methodology that uses these “unwanted” isotopic interferences in the adjacent SRM channels, which is termed as isotopic abundances in Multiple Isotopologue Reaction Monitoring (MIRM) channels in this work, to construct an In-Sample Calibration Curve (ISCC) in every study sample for quantitative LC-MS/MS bioanalysis. In this paper, the workflow for the MIRM-ISCC-LC-MS/MS methodology is presented. It is then evaluated and demonstrated with examples of quantitative analysis of a protein biomarker, three surrogate peptides and a small molecule drug in different biological matrices. The potential applications of the MIRM-ISCC-LC-MS/MS methodology in quantitative proteomics, clinical laboratories and other areas are also discussed.


Page 6 of 30

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL

Materials and Reagents Formic Acid (SupraPur grade), methyl-t-butyl ether (MTBE) were purchased from EMD Chemicals (Gibbstown, NJ, USA). HPLC grade methanol and acetonitrile were purchased from J.T. Baker (Phillipsburg, NJ, USA). LC grade ammonium bicarbonate and phosphate buffered saline with 0.05% tween (PBST) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dynabeads® M-280 Streptavidin was purchased from Invitrogen (Carlsbad, CA, USA). Sequencing grade modified trypsin was purchased from Promega Corporation (Madison, WI, USA). All non-labeled and labeled surrogate peptides for cluster of differentiation 73 (CD73): VIYPAVEGR and V[Ile(13C6,

15N)]YPAVEGR;

LAAFPEDR and LAAFPED[Arg(13C6,

15N

4)];

programmed cell death protein 1 (PD-1): and programmed death-ligand 1 (PD-L1):

LQDAGVYR and LQDAG[Val(13C5, 15N)]YR were purchased from Genscript (Piscataway, NJ, USA). Deionized water was generated in house using a NANOpure Diamond ultrapure water system from Barnstead International (Dubuque, IA, USA). Recombinant human CD73 (61,084 Da), anti-human CD73 monoclonal antibody (mAb), small molecule drug daclatasvir and SIL drug, 13C215N4-daclatasvir were generated internally at Bristol-Myers Squibb.

LC-MS/MS Instrumentation LC-MS/MS system used was a triple quadrupole 6500 mass spectrometer (AB Sciex, Foster City, CA) coupled with a Nexera UPLC system (Shimadzu, Columbia, MD). The UPLC system consists of two LC-30AD pumps, one SIL-30ACMP autosampler and one CTO-30AS column



heater. The separation was achieved on a Acquity HSS T3 analytical column (2.1 mm x 50 mm, particle size 1.8 m) (Waters, Milford, MA) with gradient elution using mobile phases of 0.01% formic acid in water (A) and 0.01% formic acid in acetonitrile (B). The LC-MS/MS data were acquired by Analyst® Software (1.6.2).

RESULTS AND DISCUSSIONS

Theory and MIRM-ISCC-LC-MS/MS Methodology Normally, the most abundant MIRM channel of an analyte (or SIL analyte) is monitored in a LCMS/MS assay for quantitative analysis. Due to elements’ naturally occurring isotopologues, in addition to the MS/MS response observed in its most abundant MIRM channel, isotopic abundances in MIRM channels adjacent to the most abundant MIRM channel can be accurately calculated and measured by LC-MS/MS. These MIRM channels are from the naturally occurring isotopologues of the analyte (or SIL analyte). Briefly, for a parent ion P (monoisotopic mass of p+1.00728*Zp, p is the monoisotopic mass of the parent molecule, Zp is the number of charge for the parent ion and proton mass is 1.00728 Da) with a daughter ion D (monoisotopic mass of d+1.00728*Zd, d is the monoisotopic mass of the daughter fragment, Zd is the number of charge for the daughter ion) and neutral loss N (monoisotopic mass of n), the most abundant (100%) MIRM channel (m/z) of the analyte (or SIL analyte) is shown below: (p+1.00728* Zp)/Zp  (d+1.00728* Zd)/Zd For a unit resolution triple quadrupole mass spectrometer using most commonly used charge states (singly-, doubly- and triply- charged ions), this MIRM channel (m/z) can be simplified as: (p+Zp)/Zp  (d+Zd)/Zd


Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The isotopic abundance in a MIRM channel (m/z) of (p+Zp+α)/Zp  (d+Zd+β)/Zd can be calculated as:24 Isotopic abundance in a MIRM channel of (p+Zp+α)/Zp  (d+Zd+β)/Zd = [relative isotopic distribution of the daughter ion at mass of (d+Zd+β)] * [relative isotopic distribution of the neutral loss at mass of n+(α-β)] where:

(1) p = d+n (2) α and β are integers, they are the number of additional neutrons on the parent ion and daughter ion, respectively, α≥0, β≥0 and α≥β (3) Zp and Zd are positive integers, they are the number of charges for the parent ion and daughter ion, respectively. (4) Isotopic distribution of a molecule can be found using an online calculator25 (5) Relative isotopic distributions of the daughter ion and neutral loss at mass of d+Zd (α=0) and n (α-β=0), respectively, are 100%

By using different combinations of α and β, the isotopic abundances in different MIRM channels (m/z) of (p+Zp+α)/Zp  (d+Zd+β)/Zd can be calculated and measured accurately. The isotopic abundances from the most abundant MIRM channel (α=0 and β=0) to the lowest abundant MIRM channel could reach as high as 5 to 6 orders of magnitude. In theory, as these isotopic abundances are coming from the combinations of different isotopologues of the daughter ion and neutral loss, and they have “identical” physicochemical properties, a linear relationship should exist between the MS/MS responses (peak areas) and the calculated theoretical isotopic abundances in all of the MIRM channels over the entire range. However, this linear relationship is limited by the mass spectrometer’s detection limit as well as its linear range, which is 3 to 4 orders of magnitude. It should be pointed out that to maintain this linear relationship, it is very critical to set the same mass spectrometer parameters (such as dwell time, collision energy and declustering potential etc. for AB Sciex triple quadrupole mass spectrometers) for all MIRM channels used for the SIL analyte and the SRM channel used for the analyte.



In our previous paper,24 the isotopic abundances in MIRM channels described here are actually “unwanted” assay interferences to be avoided. However, in this work, we propose a novel methodology which takes advantage of the isotopic abundances in these MIRM channels to construct an In-Sample Calibration Curve (ISCC) between the calculated theoretical isotopic abundances (analyte concentration equivalents) and the corresponding measured LC-MS/MS responses (peak areas) for instant measurement of biomarkers, biotherapeutics and small molecules, as well as the quantitative analysis in proteomics.

Figure 1 shows a general workflow for the MIRM-ISCC-LC-MS/MS methodology using quantitative determination of programmed death-ligand 1 (PD-L1) peptide LQDAGVYR as an example (details to be discussed later in the second example). By spiking a known amount of a SIL analyte LQDAG[Val(13C5, 15N)]YR (20 µL of 500 ng/mL = 10 ng of the SIL analyte) into each study sample (100 L), the calculated theoretical isotopic abundances in the MIRM channels of this labeled analyte can be converted to the SIL analyte isotope concentrations in the corresponding MIRM channels, which can be further converted to analyte concentration equivalents. Therefore, an ISCC between the calculated analyte concentration equivalents (x axis) and the measured MS/MS responses (y axis) can be established in each study sample, and the analyte concentration for this sample can be calculated instantly based on the established calibration curve and the analyte’s peak area.

Applications in Quantitative Bioanalysis of Protein and Peptide Compounds with MIRMISCC-LC-MS/MS Methodology


Page 10 of 30

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Example 1: MIRM-ISCC-LC-MS/MS Quantitation of Endogenous CD73 in Human and Monkey Serum CD73 is a 70 kDa protein highly expressed on many tumors.26 Quantitative determination of CD73 in human and monkey serum is needed to assist in dose selection and provide pharmacodynamic information for anti-CD73 pre-clinical and clinical drug development. Originally, an immuno-capture LC-MS/MS assay using traditional external calibration curves by serial dilution of a recombinant CD73 reference standard (61,084 Da) in a surrogate matrix was developed and validated.27 An anti-CD73 mAb was used for immuno-capture of CD73, followed by denaturation, trypsin digestion, and LC-MS/MS analysis. The surrogate peptide (unique to both human and monkey CD73) monitored in the LC-MS/MS assay was VIYPAVEGR with SRM transition (m/z) from a doubly charged parent ion to a singly charged daughter ion (y6 ion) with a transition of (m/z) 502.3++ 628.3+. Following trypsin digestion, a volume of 10 L of the SIL surrogate peptide, V[Ile(13C6, 15N)]YPAVEGR at a concentration of 100 ng/mL was added into each sample. As the original serum sample volume for this assay was 100 L, this is equivalent to 10 ng/mL ([10L•100ng/mL]/100 L) of the SIL peptide in the original serum samples, or 604.792 ng/mL of recombinant CD73 (10ng/mL•[61,084Da/1010Da]). In this experiment, this labeled peptide was used not only as the assay internal standard together with the external calibration curves, it was also used as an ISCC for each individual study sample as well. By doing this, the measured CD73 concentrations with the external calibration curves and ISCC can be compared side by side.

Before conducting the quantitation of CD73 in human and monkey serum using external calibration curve and ISCC approaches, the isotopic abundances in MIRM channels for the



labeled surrogate peptide, V[Ile(13C6, 15N)]YPAVEGR, need to be calculated and measured to confirm the selected MIRM channels are reliable and accurate to establish an ISCC without unexpected interferences.

The fragmentation of a peptide in triple quadrupole mass spectrometers can be easily resolved by using an online tool, such as Skyline (MacCoss Lab, Department of Genome Sciences, UW). In this case, as an y6 ion is monitored as the daughter ion, the daughter ion and neutral loss are determined to be C26H46N9O9+ and 13C615NC14H29N2O4, respectively. The isotopic distributions of the daughter ion (C26H46N9O9+) and neutral loss (13C615NC14H29N2O4) were calculated using an online calculator25 and listed in Table 1. The isotopic abundances in the most abundant MIRM channel (100% abundance) and adjacent MIRM channels for V[Ile(13C6, 15N)]YPAVEGR were calculated using the method we reported before24 and also described in the previous section. For example, for the MIRM channel of 506.3  629.3 (m/z) in Table 2 (channel 2), the mass for the neutral loss and daughter ion is 382.2 Da and 629.3 Da, respectively. The isotopic distributions for the neutral loss and daughter ion are 100% and 32.4695%, respectively, as shown in Table 1. Therefore, the isotopic abundance in this MIRM channel is 32.4695% (100% * 32.4695%). Similarly, the isotopic abundance in the MIRM channel of 507.8  631.3 (m/z) (Table 2, channel 12) can be calculated as 0.1822% (16.4891% * 1.1054%) based on the mass of its neutral loss (383.2 Da) and daughter ion (631.3 Da). The calculated theoretical isotopic abundances in these MIRM channels and the measured results (peak areas) with a Sciex API 6500 mass spectrometer in the corresponding MIRM channels for V[ILE(13C6, 15N)]YPAVEGR are shown in Table 2. The abundances cover over 6,000-fold from the most to the least abundant MIRM channel. Lower isotopic abundances could also be calculated and used if necessary.


Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


However, in many cases, it will be beyond the linear range for a triple quadrupole mass spectrometer. As shown in Table 2, the percentage differences for the measured results from the calculated theoretical isotopic abundances are within 13.5%, indicating the measured results are accurate and reliable without any interferences (e.g., interferences from isotope impurities and endogenous matrix), and therefore, these MIRM channels could be selected for the MIRMISCC-LC-MS/MS absolute quantitative analysis.

With the calculated theoretical isotopic abundances in the MIRM channels of the SIL peptide, the spiked SIL peptide isotope concentrations (and CD73 protein concentration equivalents) in the selected ten MIRM channels to be used for ISCC were calculated and listed in the right two columns in Table 2. An ISCC is constructed in each study sample using CD73 protein concentration equivalents (x axis) and the measured MS/MS peak areas (y axis) in the corresponding MIRM channels. Calibration curve regressions and concentration calculations were performed using an in-house developed software. A weighted (1/x2) least squares linear regression was used for all ISCCs. The ISCC performances for the first three injections for sample No. 1 in human plasma (all samples were extracted and analyzed in three replicates) are shown in the Supporting Information A, Table SI-A1. As shown in Table SI-A1, excellent ISCC performances (%DEV ≤ 8.7%) were observed for these three injections. Representative chromatograms for all ten MIRM channels used for ISCC and one SRM channel used for the analyte from the second injection are shown in the Supporting Information A, Figure SI-A1.

The results for the quantitative analysis of endogenous CD73 in human and monkey serum using external calibration curve and ISCC approaches are listed in Table 3. Overall, CD73



concentrations measured with ISCC approach are about 11% to 17 % lower than the concentrations measured using external calibration curve. This was caused by the 86.0% recovery for the immunocapture and digestion, as the immunocapture and digestion losses for the study samples were tracked and compensated by the external calibration curve. However, this was not the case for the ISCC approach as the SIL peptide was spiked after the digestion. Therefore, the concentrations measured with ISCC approach should be adjusted with the 86.0% recovery, and the adjusted concentrations matched with the concentrations measured using the external calibration curve very well, as shown in the Table 3.

It is worth mentioning that it is not necessary to include all promising MIRM channels in the final assay. The selected MIRM channels for the quantitation of CD73 in human and monkey serum are noted in Table 2. There are several considerations in selecting MIRM channels to be used in sample analysis runs, as listed below: 1. The ISCC curve range is defined by the isotopic abundance range of the selected MIRM channels. Therefore, appropriate MIRM channels should be selected to cover the expected concentration range. In this work, the selected MIRM channels had about 1,600-fold curve range to cover the expected increase of CD73 after dose. 2. Any MIRM channel with large %Dev (>15%) between the calculated and measured isotopic abundances should not be used in the final assay, as the large % Dev normally means that there is potential interference in the MIRM channel, including the interferences from isotope impurities and matrix endogenous. Therefore, multiple matrix lots should be tested during MIRM channel selection.


Page 14 of 30

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3. Only one MIRM channel should be selected among multiple MIRM channels with similar isotopic abundances. In this example, as shown in Table 2, the MIRM channels 4 and 8 were not selected because the MIRM channels 4 and 5, 7 and 8 have very close isotopic abundances, respectively. 4. A total of ten MIRM channels were used in this example for demonstration purpose only. Using fewer MIRM channels (four to five MIRM channels for 1,000-fold curve) does not impact data quality (data not shown).

Example 2: MIRM-ISCC-LC-MS/MS Quantitation of Surrogate Peptides for Protein Biomarkers PD-1, PD-L1 and CD73 in Digested Human Colon Homogenates The MIRM-ISCC-LC-MS/MS methodology described above can be easily applied into quantitative proteomics by spiking known amounts of multiple SIL surrogate peptides into the digested samples for the absolute quantitative proteomics for multiple peptide targets. Similar to AQUA approach, the MIRM-ISCC quantitation is also based on the concentrations of the SIL surrogate peptides spiked into the samples. However, as only one calibration point is used in AQUA approach for each target peptide, the accuracy of the quantitation could be greatly compromised, especially when the concentration for a target peptide is much higher or much lower than the concentration of the spiked SIL surrogate peptide. The MIRM-ISCC-LC-MS/MS approach, on the other hand, can offer a full calibration curve range with 3 to 4 orders of magnitude for each target peptide, and the accuracy of the quantitation can be assured within the entire curve range.



In this example, three surrogate peptides, LAAFPEDR for PD-1, LQDAGVYR for PD-L1 and VIYPAVEGR for CD73 were mixed and spiked into fully trypsin digested human colon tissue homogenates at concentrations of 1.00, 10.0 and 50.0 ng/mL, respectively. A volume of 100 L of the prepared sample was used in the assay. A mixture of 10 ng (20 L of 500 ng/mL) for each SIL peptide LAAFPED[Arg(13C6, 15N4)], LQDAG[Val(13C5, 15N)]YR and V[Ile(13C6, 15N)]YPAVEGR

in 10% methanol 90% water was added into the prepared samples for the

MIRM-ISCC-LC-MS/MS analysis. The equivalent concentration for each of the SIL peptide in the samples is 100 ng/mL (10ng/100L). Table SI-B1 in the Supporting Information B shows the MIRM channels and their analyte concentration equivalents used for the MIRM-ISCC-LCMS/MS quantitative analysis of these three peptides. A weighted (1/x2) least squares linear regression was used for all ISCC curves. Excellent ISCC curve performances were demonstrated by all calibration points are within 10.0% (data not shown) of the nominal concentrations, which is the analyte concentration equivalents. The measured concentrations for these three peptides are listed in the Supporting Information B, Table SI-B2, and the accuracy of the MIRM-ISCCLC-MS/MS measurement was confirmed by the results of all samples tested for all three peptides.

One potential issue for the MIRM-ISCC-LC-MS/MS approach in targeted quantitative proteomics for multiple peptides is that the total number of MIRM channels needs to be monitored in a LC-MS/MS run could be too many to be handled by a triple quadrupole mass spectrometer. This issue could be relieved by using scheduled MIRMs based on the different retention time windows for each target peptide, and using fewer MIRM channels in each ISCC. Our test results indicated that accurate and reliable quantitation still could be achieved by using 3


Page 16 of 30

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to 4 MIRM channels for a concentration range of 2 to 3 orders of magnitude in a multiplexed fashion.

Critical Considerations for MIRM-ISCC-LC-MS/MS Assays There are several critical considerations for the successful MIRM-ISCC-LC-MS/MS assay development and sample analysis. 1. Selection of a proper SIL analyte is one of the key factors to develop a reliable and robust quantitative MIRM-ISCC-LC-MS/MS assay. As the isotopic abundances in the selected MIRM channels of this labeled analyte will be used as a calibration curve for quantitative analysis of the analyte, the labeled analyte should be designed to avoid the isotopic interference from the analyte24 as this interference could compromise the assay accuracy. The rule of thumb is that four to six labels are needed to avoid the interference for most small molecule compounds and peptide analytes with 6 to 12 amino acids. 2. The impurity (amount of non-labeled analyte) in the SIL analyte should be low enough to avoid the interference from the SIL analyte to the analyte because, for ISCC approach, a large amount of the SIL analyte is needed in each study sample to define the assay upper limit of quantitation (ULOQ). In addition, the labeling impurity (amount of labeled analyte with fewer or more labeled positions than that of the SIL analyte) should also be low enough to avoid the interferences to the isotopic abundances in the MIRM channels of the SIL analyte. The impact of labeling impurity on the MIRM-ISCC-LC-MS/MS assay accuracy is discussed in the Supporting Information E. 3. Although deuterium labeling is very cost effective and easily available, deuterium labeled analytes should be avoided in ISCC approach due to the easy separation of the deuterium



labeled analytes from the non-labeled analytes. More importantly, the hydrogendeuterium exchange reaction, which can easily occur on exchangeable protons and deuterons of a deuterium labeled analyte, makes the accurate calculation of the isotopic abundances in MIRM channels very difficult. 4. Use a properly calibrated triple quadrupole mass spectrometer is another key factor for the success of the MIRM-ISCC-LC-MS/MS approach. For singly- and doubly- charged parent ions, unit resolutions (full width at half height-FWHH = 0.7 mass unit) for both Q1 and Q3 are good enough to generate accurate MS/MS responses close to the calculated theoretical isotopic abundances in the corresponding MIRM channels. If necessary, higher resolution (FWHH = 0.5 mass unit) in Q1 can improve the measurement accuracy, with the cost of losing some instrument sensitivity. Our test results showed that using higher resolution (FWHH = 0.5 mass unit) in Q3 is not helpful in improving the measurement accuracy. 5. As the isotope spacing (1 Da for Zp=1, 0.5 Da for Zp=2, 0.33 Da for Zp=3, and so on) gradually decreases with the increase of the number of charge (Zp), it is anticipated that accurate measurement of isotopic abundances in MIRM channels with triply (and higher) charged parent ions might be challenging even using resolutions with FWHH ≤ 0.5 mass unit. In this work, isotopic abundances in MIRM channels with triply (or higher) charged parent ions were not tested because triply (or higher) charged parent ions with high MS/MS responses were not available. 6. High resolution MS (HRMS) can also be used to construct ISCCs by monitoring multiple isotopologue parent ions of a SIL analyte. As a result, only the isotopic distribution of the SIL analyte’s parent ion is needed, avoiding the calculation of the isotopic abundances in


Page 18 of 30

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


each MIRM channel. Another advantage of using HRMS is that the triply (or higher) charged parent ions can be easily resolved by HRMS; however, both assay sensitivity and selectivity may be compromised and the selection of calibration curve points may also be limited by monitoring the parent ions only. 7. The performances for an established ISCC should be very consistent from sample to sample during the assay qualification and sample analysis as the ISCC is constructed using naturally occurring isotopologues in selected MIRM channels, with no human and instrument operations involved. Any unexpected significant bias in one MIRM channel for a few samples normally indicates endogenous interferences from those matrix lots, and excluding this point has no significant impact on the data accuracy. On the other hand, any unexpected significant biases in several MIRM channels for many samples indicate failure of MS instrument calibration. Inconsistent extraction recovery and matrix effect at different concentration levels or in different matrix lots can cause the calibration curve’s slope change from sample to sample. However, the inconsistencies have no impact on the assay performance and data quality if the SIL analyte is spiked at the beginning of sample preparation as the recovery and matrix effect are tracked and compensated by ISCC. In the cases where the SIL analytes are not spiked at the beginning of sample preparation, consistent recovery from sample to sample is very critical to ensure the data quality. 8. There is no need to add an additional assay internal standard in the MIRM-ISCC-LCMS/MS approach because an ISCC is in each study sample, and therefore all variations after the spiking of a SIL analyte into the study samples, including variations from extraction, injection, ionization, fragmentation and detection, are tracked and



compensated by the ISCC itself. Because of this, the assay performance could be further improved by spiking the SIL analyte as early as possible during the sample preparation, such as the analysis of small molecule drug daclatasvir in the Supporting Information C where the SIL daclatasvir was spiked into the samples at the beginning of the sample preparation. However, for protein analysis with immuno-capture, the labeled peptides can only be spiked after immuno-capture, and any variations during immuno-capture and trypsin digestion are not tracked and compensated, such as the analysis of CD73 protein in the example 1. This issue can be resolved by spiking a SIL protein (with labeled positions on the surrogate peptide) at the beginning of the sample preparation. 9. Similar to the traditional quality control samples used with external calibration curves, quality control samples could be placed in each MIRM-ISCC-LC-MS/MS batch using samples spiked with analyte at known concentrations to evaluate the assay precision and accuracy. As each sample has its own calibration curve, and the accuracy of the curve solely depends on the accuracy of the single spike of the SIL analyte, the spiking accuracy should also be monitored for each sample by comparing the SIL analyte’s peak areas from sample to sample in each batch. Other options of quality control for MIRMISCC-LC-MS/MS approach should also be explored and evaluated. 10. As an ISCC is in each individual sample and currently there is no commercial software which is capable to build ISCCs using peak areas from multiple MIRM channels, an inhouse developed software was used in this work for generating ISCCs with weighted least squares regression algorithm, and calculating sample concentrations in batch. Wide application of the MIRM-ISCC-LC-MS/MS methodology is relied on the commercial software development by major mass spectrometer companies.


Page 20 of 30

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


11. While equal molar MS/MS responses between an analyte and its SIL-analyte is assumed, it still needs to be tested and verified during assay development.9 12. The same MIRM-ISCC-LC-MS/MS workflow can be applied for the quantitative analysis of proteins (example 1), peptides (example 2) and small molecules (Supporting Information C). The step-by-step workflow for MIRM-ISCC-LC-MS/MS methodology is listed in Supporting Information D.

Other Potential Applications of MIRM-ISCC-LC-MS/MS Approach Unlike the traditional external calibration curve approach, the MIRM-ISCC-LC-MS/MS methodology allows the instant and accurate measurement of each individual sample without using external calibration curves. Therefore, this approach eliminates the need of using authentic biological matrix, simplifies the quantitative LC-MS/MS bioanalysis process and greatly reduces instrument time. While the MIRM-ISCC-LC-MS/MS methodology can be applied in regular pharmacokinetic (PK) sample analysis in drug discovery and development, this methodology is particularly useful for cases where authentic matrices are hardly available, such as biomarker measurement and quantitative proteomics; where the low throughput and long turnaround time are the main issues preventing the use of LC-MS/MS technique, such as the clinical diagnosis in clinical diagnostic laboratories;28 and where external calibration curve preparation is cumbersome, such as the analysis of fresh frozen and FFPE tissue samples as well as dried blood spot (DBS) samples. Additionally, an MIRM-ISCC could also be used as an external calibration curve by spiking a known amount of non-labeled analyte in blank matrix. As a result, an external calibration curve can be constructed in just one sample, eliminating the need for preparation of multiple samples for an external calibration curve. Similar to the traditional external calibration



curve, a SIL internal standard is needed for this single sample external calibration curve approach to ensure the assay performance. Additional work is needed to test the assay performance, evaluate the feasibility and validate this approach.

CONCLUSIONS The MIRM-ISCC-LC-MS/MS methodology for the quantitative bioanalysis was proposed and demonstrated for the first time. By spiking a known amount of a SIL analyte into each study sample, an ISCC can be easily established based on the relationship between the calculated theoretical isotopic abundances and the measured MS/MS peak areas in the corresponding MIRM channels to quantify each individual study sample. The MIRM-ISCC-LC-MS/MS methodology was tested and demonstrated in this work with three examples: (1) a protein biomarker target CD73 using protein level immunocapture, (2) three peptide targets for PD-1, PD-L1 and CD73 in digested colon tissue homogenates and (3) a small molecule drug daclatasvir in human and rat plasma using liquid-liquid extraction. With this novel MIRM-ISCC-LCMS/MS methodology, instant, accurate and reliable LC-MS/MS bioanalysis of biomarkers, biotherapeutics and small molecule drugs in support of drug discovery and development can be achieved without using external calibration curves. Moreover, the MIRM-ISCC-LC-MS/MS methodology can bring unique values of eliminated external calibration curves, simplified the workflows and improved throughput in the areas where absolute quantitation and overall sample turnaround still remain great challenges. The potential applications in quantitative proteomics, clinical laboratories and other areas were also discussed.

ACKNOWLEDGMENTS


Page 22 of 30

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The authors would like to thank Mr. Craig Titsch for providing both SIL and non-labeled daclatasvir compounds. REFERENCES 1. Chahrour, O.; Cobice, D.; Malone, J., Stable isotope labelling methods in mass spectrometry-based quantitative proteomics. Journal of Pharmaceutical and Biomedical Analysis 2015, 113, 2-20. 2. Rifai, N.; Gillette, M. A.; Carr, S. A., Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nature Biotechnology 2006, 24, 971. 3. Schuck, E.; Bohnert, T.; Chakravarty, A.; Damian-Iordache, V.; Gibson, C.; Hsu, C.-P.; Heimbach, T.; Krishnatry, A. S.; Liederer, B. M.; Lin, J.; Maurer, T.; Mettetal, J. T.; Mudra, D. R.; Nijsen, M. J. M. A.; Raybon, J.; Schroeder, P.; Schuck, V.; Suryawanshi, S.; Su, Y.; Trapa, P.; Tsai, A.; Vakilynejad, M.; Wang, S.; Wong, H., Preclinical Pharmacokinetic/Pharmacodynamic Modeling and Simulation in the Pharmaceutical Industry: An IQ Consortium Survey Examining the Current Landscape. The AAPS Journal 2015, 17 (2), 462-473. 4. Trusheim, M. R.; Berndt, E. R.; Douglas, F. L., Stratified medicine: strategic and economic implications of combining drugs and clinical biomarkers. Nature Reviews Drug Discovery 2007, 6, 287. 5. Ciccimaro, E.; Blair, I. A., Stable-isotope dilution LC–MS for quantitative biomarker analysis. Bioanalysis 2010, 2 (2), 311-341. 6. Gu, H.; Liu, G.; Wang, J.; Aubry, A.-F.; Arnold, M. E., Selecting the Correct Weighting Factors for Linear and Quadratic Calibration Curves with Least-Squares Regression Algorithm in Bioanalytical LC-MS/MS Assays and Impacts of Using Incorrect Weighting Factors on Curve Stability, Data Quality, and Assay Performance. Analytical Chemistry 2014, 86 (18), 8959-8966. 7. Liu, R. H.; Lin, D. L.; Chang, W.-T.; Liu, C.; Tsay, W.-I.; Li, J.-H.; Kuo, T.-L., Peer Reviewed: Isotopically Labeled Analogues for Drug Quantitation. Analytical Chemistry 2002, 74 (23), 618 A-626 A. 8. Miller, J. N., Basic statistical methods for Analytical Chemistry. Part 2. Calibration and regression methods. A review. Analyst 1991, 116 (1), 3-14. 9. Jian, W.; Edom, R. W.; Weng, N., Important considerations for quantitation of smallmolecule biomarkers using LC–MS. Bioanalysis 2012, 4 (20), 2431-2434. 10. LeBlanc, A.; Michaud, S. A.; Percy, A. J.; Hardie, D. B.; Yang, J.; Sinclair, N. J.; Proudfoot, J. I.; Pistawka, A.; Smith, D. S.; Borchers, C. H., Multiplexed MRM-Based Protein Quantitation Using Two Different Stable Isotope-Labeled Peptide Isotopologues for Calibration. Journal of Proteome Research 2017, 16 (7), 2527-2536. 11. Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R., Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nature Biotechnology 1999, 17, 994. 12. Schmidt, A.; Kellermann, J.; Lottspeich, F., A novel strategy for quantitative proteomics using isotope-coded protein labels. PROTEOMICS 2005, 5 (1), 4-15.



13. Hall, M. P.; Ashrafi, S.; Obegi, I.; Petesch, R.; Peterson, J. N.; Schneider, L. V., ‘Mass defect’ tags for biomolecular mass spectrometry. Journal of Mass Spectrometry 2003, 38 (8), 809-816. 14. Hall, M. P.; Schneider, L. V., Isotope-differentiated binding energy shift tags (IDBEST™) for improved targeted biomarker discovery and validation. Expert Review of Proteomics 2004, 1 (4), 421-431. 15. Craft, G. E.; Chen, A.; Nairn, A. C., Recent advances in quantitative neuroproteomics. Methods 2013, 61 (3), 186-218. 16. Koehler, C. J.; Strozynski, M.; Kozielski, F.; Treumann, A.; Thiede, B., Isobaric Peptide Termini Labeling for MS/MS-Based Quantitative Proteomics. Journal of Proteome Research 2009, 8 (9), 4333-4341. 17. Conrads, T. P.; Alving, K.; Veenstra, T. D.; Belov, M. E.; Anderson, G. A.; Anderson, D. J.; Lipton, M. S.; Paša-Tolić, L.; Udseth, H. R.; Chrisler, W. B.; Thrall, B. D.; Smith, R. D., Quantitative Analysis of Bacterial and Mammalian Proteomes Using a Combination of Cysteine Affinity Tags and 15N-Metabolic Labeling. Analytical Chemistry 2001, 73 (9), 2132-2139. 18. Langen, H.; Takács, B.; Evers, S.; Berndt, P.; Lahm, H.-W.; Wipf, B.; Gray, C.; Fountoulakis, M., Two-dimensional map of the proteome of Haemophilus influenzae. ELECTROPHORESIS 2000, 21 (2), 411-429. 19. Oda, Y.; Huang, K.; Cross, F. R.; Cowburn, D.; Chait, B. T., Accurate quantitation of protein expression and site-specific phosphorylation. Proceedings of the National Academy of Sciences 1999, 96 (12), 6591-6596. 20. Brun, V.; Masselon, C.; Garin, J.; Dupuis, A., Isotope dilution strategies for absolute quantitative proteomics. Journal of Proteomics 2009, 72 (5), 740-749. 21. Gerber, S. A.; Rush, J.; Stemman, O.; Kirschner, M. W.; Gygi, S. P., Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proceedings of the National Academy of Sciences 2003, 100 (12), 6940-6945. 22. Anderson, N. L.; Anderson, N. G.; Pearson, T. W.; Borchers, C. H.; Paulovich, A. G.; Patterson, S. D.; Gillette, M.; Aebersold, R.; Carr, S. A., A Human Proteome Detection and Quantitation Project. Molecular & Cellular Proteomics 2009, 8 (5), 883-886. 23. Chistina Chiva, O. P., Eduard Sabido, Internal calibration curves for accurate quantitation in clinical proteomics. In 66th ASMS Conference on Mass Spectrometry and Allied Topics, San Diego, CA, 2018; p ThP 481. 24. Gu, H.; Wang, J.; Aubry, A.-F.; Jiang, H.; Zeng, J.; Easter, J.; Wang, J.-s.; Dockens, R.; Bifano, M.; Burrell, R.; Arnold, M. E., Calculation and Mitigation of Isotopic Interferences in Liquid Chromatography–Mass Spectrometry/Mass Spectrometry Assays and Its Application in Supporting Microdose Absolute Bioavailability Studies. Analytical Chemistry 2012, 84 (11), 4844-4850. 25. Isotope distribution calculator. https://www.sisweb.com/mstools/isotope.htm (accessed October 18). 26. Zhang, B., CD73: A Novel Target for Cancer Immunotherapy. Cancer Research 2010, 70 (16), 6407-6411. 27. Zhao, Y.; Gu, H.; Postelnek, J.; Klippel, A.; Zhang, Y.; Zeng, J. In Quantitative analysis of total soluble CD73 in human serum as a pharmacodynamic biomarker by immuno-captureLC-MS/MS, 66th ASMS Conference on Mass Spectrometry and Allied Topics, San Diego, CA, San Diego, CA, 2018.


Page 24 of 30

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


28. Grebe, S. K.; Singh, R. J., LC-MS/MS in the Clinical Laboratory - Where to From Here? The Clinical biochemist. Reviews 2011, 32 (1), 5-31.



Page 26 of 30

Table 1, Isotopic distributions for neutral loss (13C615NC14H29N2O4) and daughter ion (C26H46N9O9+) for stable isotopically labeled peptide V[Ile(13C6, 15N)]YPAVEGR Mass shift for neutral loss: (α-β)

Lost in collision cell (neutral loss) 13C 15NC H N O 6 14 29 2 4 Mass Abundance (m/z) (%) 382.2 100 383.2 16.4891 384.2 2.0781 385.2 0.1935 386.2 0.0145 387.2 0.0008

Mass shift for daughter ion: β

Daughter ion (y6 ion)

[C26H46N9O9]+ Abundance Mass (%) 0 0 628.3 100 1 1 629.3 32.4695 2 2 630.3 6.9176 3 3 631.3 1.1054 4 4 632.3 0.1447 5 5 633.3 0.0159 6 634.3 0.0013 Note: Parent ion: [13C615NC40H76N11O13]++, doubly charged (Zp=2)


Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47


Table 2, Calculated and measured relative isotopic abundances in MIRM channels of SIL peptide V[ILE(13C6, 15N)]YPAVEGR

No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

MIRM channels (m/z)

Calculated theoretical relative isotopic abundance (%)

Neutral loss mass

Daughter ion mass

Measured responses (peak area)

505.8628.3 506.3629.3 506.3628.3 506.8630.3 506.8629.3 506.8628.3 507.3631.3 507.3630.3 507.3629.3 507.3628.3 507.8632.3 507.8631.3 507.8630.3 507.8629.3 507.8628.3 •••

100.0000 32.4695 16.4891 6.9176 5.3539 2.0781 1.1054 1.1406 0.6747 0.1935 0.1447 0.1822 0.1438 0.0628 0.0145 •••

382.2 382.2 383.2 382.2 383.2 384.2 382.2 383.2 384.2 385.2 382.2 383.2 384.2 385.2 386.2 •••

628.3 629.3 628.3 630.3 629.3 628.3 631.3 630.3 629.3 628.3 632.3 631.3 630.3 629.3 628.3 •••

51529500 18080800 8961640 3798610 2846230 1077450 618461 657095 374215 110815 84638.5 101662 77694.9 35250.9 8305.59 •••

Measured relative isotopic abundance (%) based on peak areas 100.0000 35.0883 17.3913 7.3717 5.5235 2.0909 1.2002 1.2752 0.7262 0.2151 0.1643 0.1973 0.1508 0.0684 0.0161 •••

% Dev from calculated relative isotopic abundance

Selected MIRM channels for ISCC

0.0 8.1 5.5 6.6 3.2 0.6 8.6 11.8 7.6 11.2 13.5 8.3 4.9 8.9 11.0 •••

Yes Yes Yes No Yes Yes Yes No Yes Yes Yes No No Yes No

Spiked ISCC SIL peptide isotope concentration (ng/mL)a 10.0000 3.2470 1.6489

Spiked ISCC CD73 protein concentration equivalent (ng/mL)b 604.792 196.376 99.724

0.5354 0.2078 0.1105

32.381 12.568 6.683

0.0675 0.0194 0.0145

4.082 1.173 0.877

0.0063

0.381

10 µL of 100 ng/mL (1 ng) of SIL peptide was spiked into the digested sample. As the original sample volume used for the assay was 100 µL, this is equivalent to that, in the original sample, there is 10 ng/mL of SIL peptide in its most abundant MIRM channel. Other SIL peptide isotope concentrations in adjacent MIRM channels were calculated based on the calculated theoretical relative isotopic abundances. b CD73 protein concentration equivalent = SIL peptide isotope concentration * (recombinant CD73 molecular weight of 61,084 /SIL peptide molecular weight of 1010). a

27



Table 3, Quantitative determination of endogenous CD73 in human and monkey serum using external calibration curve and MIRM-ISCC-LC-MS/MS approaches

Species/ matrix

Sample No.

Replicates

Measured CD73 protein concentration (ng/mL) using external calibration curve

Measured CD73 protein concentration (ng/mL) using ISCC

%Deva

Recovery adjusted CD73 protein concentration (ng/mL) using ISCCb

%Devc

7.838 -16.4 9.114 -2.7 1 9.370 9.960 -14.1 11.581 -0.1 2 11.595 8.940 -14.5 10.395 -0.6 3 10.457 3.740 -15.2 4.349 -1.4 1 4.412 3.453 -14.7 4.015 -0.8 2 2 4.047 4.216 -15.8 4.902 -2.1 3 5.005 3.920 -12.0 4.558 2.3 1 4.454 3.578 -14.1 4.160 -0.2 3 2 4.167 3.751 -12.9 4.362 1.2 3 4.309 1.696 -13.2 1.972 1.0 1 1.953 2.027 -13.1 2.357 1.1 4 2 2.332 1.825 -11.5 2.122 2.9 3 2.063 2.333 -15.9 2.713 -2.2 1 2.773 2.429 -13.6 2.824 0.5 2 2.811 5 2.471 -11.3 2.873 3.1 3 2.786 2.793 -16.6 3.248 -3.0 1 3.350 2.580 -13.6 3.000 0.5 6 2 2.986 2.644 -16.4 3.074 -2.8 3 3.162 a Percentage difference for the CD73 protein concentrations using ISCC from the CD73 protein concentrations using external calibration curve. b CD73 concentrations were adjusted with the recovery (86.0%) of immunocapture and digestion as the spiked SIL peptide did not go through the immunocapture and digestion steps. c Percentage difference for the recovery adjusted CD73 protein concentrations using ISCC from the CD73 protein concentrations using external calibration curve.

Human serum

1

Monkey Serum

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Page 28 of 30

28


Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. 2. 3. 4.

SIL analyte MIRM channel 1 (m/z: 464.2686.4), isotopic abundance 100%: 100 ng/mL of SIL analyte isotope concentration (99.4 ng/mL of analyte concentration equivalent) SIL analyte MIRM channel 2 (m/z: 464.7687.4), isotopic abundance 30.0%: 30.0 ng/mL of SIL analyte isotope concentration (29.8 ng/mL of analyte concentration equivalent) SIL analyte MIRM channel 3 (m/z: 465.2688.4), isotopic abundance 6.63%: 6.63 ng/mL of SIL analyte isotope concentration (6.59 ng/mL of analyte concentration equivalent) Analyte SRM channel (m/z: 461.2 680.4), measured concentration: 56.8 ng/mL

Figure 1, Scheme diagram of MIRM-ISCC-LC-MS/MS methodology workflow using quantitative determination of a PD-L1 surrogate peptide as an example

29



173x97mm (120 x 120 DPI)


Page 30 of 30

In-Sample Calibration Curve (ISCC) Using Multiple Isotopologue

Recommend Documents