Achieve Sensitive Quantification of GLP-1 Analogs

Application Note Biopharma/Pharma Authors Xi Qiu, Steve Murphy, and David L. Wong Agilent Technologies, Inc. Abstract Sensitive bioanalytical assays to determine pharmacokinetics and toxicokinetics are crucial for advancing glucagon-like peptide-1 (GLP-1) receptor agonist medications. This application note presents an automated workflow using LC/MS to quantify the GLP-1 analog tirzepatide in human plasma. Our results indicate that this workflow achieved an LLOQ of 0.05 ng/mL for tirzepatide by using 100 μL biological sample, and was linear up to 1,000 ng/mL. This workflow can be used for quantitative analysis of GLP-1 therapeutics without the need for specific antibodies. It also provides excellent sensitivity, high specificity, and fast method development, which all play a crucial role in drug discovery and development. Sensitive Quantitation of Glucagon‑Like Peptide-1 (GLP-1) Analog Tirzepatide in Plasma Using an automated workflow with the Agilent 6495D triple quadrupole LC/MS 2 Introduction In recent years, the research and application of GLP-1 receptor agonists have significantly increased due to their effectiveness in treating obesity and type 2 diabetes. The global market for GLP-1 analog is projected to grow rapidly, reaching $140 billion by change to 2030. Developing sensitive bioanalytical assays to determine pharmacokinetics and toxicokinetics is crucial for advancing these medications. Liquid chromatography/mass spectrometry (LC/MS) has emerged as an alternative method for analyzing these large molecules due to its high specificity, sensitivity, wide dynamic range, and rapid method development. Additionally, LC/MS can avoid cross-reactivity, enhance productivity, and reduce costs and delays associated with reagent and antigen availability. Among the notable advancements is the peptide tirzepatide, which is sold under the brand names "Zepbound" for weight loss and "Mounjaro" for diabetes.1,2 Traditionally, the plasma concentration of tirzepatide is measured by a ligand-binding assay, which requires time to develop antibodies and can lack selectivity and specificity. In this application note, we present an automated workflow with LC/MS technologies to quantify GLP-1 analog drugs in human plasma using an Agilent 1290 Infinity II bio LC and an Agilent 6495D triple quadrupole LC/MS (LC/TQ) system (Figure 1). Figure 1. Agilent AssayMAP Bravo protein sample prep platform, 1290 Infinity II bio LC, and 6495D LC/TQ. 3 Experimental Materials and methods Tirzepatide, semaglutide, formic acid (FA), and ammonium formate were purchased from MilliporeSigma (St. Louis, MO, U.S.). Acetonitrile (ACN) and methanol (MeOH) were purchased from Honeywell (Charlotte, NC, U.S.). The 96-well LoBind plates and protein LoBind tubes were purchased from Eppendorf (Hauppauge, NY, U.S.). Also used were Agilent AssayMAP RP-S cartridges. Instrumentation The following instrumentation was used: – Agilent AssayMAP Bravo protein sample prep platform (G5571AA) – Agilent 1290 Infinity II bio LC system including: – Agilent 1290 Infinity II bio high-speed pump (G7132A) – Agilent 1290 Infinity II bio multisampler (G7137A) – Agilent 1290 Infinity II thermostat column compartment (G7116A) equipped with a standard flow Quick Connect bio heat exchanger (G7116-60071) – Agilent 6495D LC/TQ (G6495DA) Software The following software was used: – Agilent MassHunter Acquisition software (version 12.0) – Agilent MassHunter Quantitative Analysis software (version 12.0) Sample preparation Protein precipitation: A mixture of 150 μL ACN and 150 μL MeOH was added to a 100 μL aliquot of human plasma fortified with different concentrations of tirzepatide. Also, 5 μL of 1,000 ng/mL semaglutide was used as the internal standard. The mixture was vortexed for 2 minutes, then spun down at 17,000 g for 10 minutes. A 350 μL volume of each supernatant was transferred to a 96-well plate and dried down under nitrogen gas with heating. Then, 110 μL water was added to each well to reconstitute. The solution was further purified using AssayMAP RP-S cartridges. AssayMAP automated purification: AssayMAP RP-S cartridges were conditioned with 80% ACN and equilibrated with water. Then, 100 μL of the previously mentioned reconstituted samples were loaded onto AssayMAP RP-S cartridges at 3 μL/min, and cartridges were washed once with water and once with 10% ACN in 10 mM ammonium formate buffer. The final elution step was carried out by loading 30 μL of 5% FA in 80% ACN to the cartridge at 3 μL/min. Lastly, 30 μL water with 5% FA was added to the final elution and 5 μL was injected into LC/MS for analysis. LC/MS analysis Data acquisition was performed using an Agilent 1290 Infinity II bio UHPLC coupled to an Agilent 6495D LC/TQ with the Agilent Jet Stream electrospray ion source. Separation was obtained with an Agilent AdvanceBio Peptide Mapping column (2.1 × 150 mm, 120 Å, 2.7 μm). Tables 1 and 2 list the LC and MS parameters used for this workflow. Positive electrospray ionization of tirzepatide yielded [M + 4H]4+ signal at m/z 1,204.4 as the most intense ion. MRM transitions were optimized and 1,204.4 & 396.2 was chosen as the quantifier and 1,204.4 & 299.2 was chosen as the qualifier, and the semaglutide MRM transition (1,029.4 & 1,238.1) was chosen. Both peptides' MRM transitions are listed in Table 3 with optimal collision energies. Parameter Value Column Agilent AdvanceBio Peptide Mapping 120 Å, 2.1 × 150 mm, 2.7 μM Column Temperature 80 °C Injection Volume 5 μL Autosampler Temperature 4 °C Needle Wash 3 s in wash port (50:50 water:methanol) Mobile Phase A) water + 0.1% FA B) ACN + 0.1% FA Flow Rate 0.4 mL/min Gradient Program Time %B 0 30 3.0 45 5.0 60 5.1 95 8.0 95 8.2 30 10.0 30 Stop Time 10.0 min Table 1. LC parameters. 4 Parameter Value Ion Mode Positive Gas Temperature 290 °C Drying Gas Flow 15 L/min Nebulizer Gas 30 psi Sheath Gas Temperature 400 °C Sheath Gas Flow 11 L/min Capillary Voltage 6,000 V Nozzle voltage 2,000 V Table 2. MS acquisition parameters. Peptide Precursor Ion Product Ion Collision Energy Tirzepatide 1,204.4 396.2 26 Tirzepatide 1,204.4 299.2 65 Semaglutide 1,029.4 1,238.1 30 Table 3. Tirzepatide and semaglutide MRM transitions. Data processing All MS data were processed using Agilent MassHunter Quantitative Analysis software. Results and discussion Method optimization for GLP-1 analog peptide quantitative analysis To improve the sensitivity and reproducibility for GLP-1 peptide quantitative analysis, all the sample preparation steps were evaluated to achieve the best results. 1. Organic solvent composition to precipitate proteins was evaluated, including just ACN, just MeOH, and ACN:MeOH (1:1). The ACN:MeOH (1:1) was found to be the best for tirzepatide and semaglutide extraction (Figure 2). 2. Loading buffer was evaluated, including pure water, and water with 0.1% FA and 10%, 20%, or 40% ACN. Pure water was found to be the best loading buffer for GLP-1 peptides. 3. Washing buffer was evaluated, including pure water, water with 0.1% FA, water with 1% FA, and water with 10 mM NH4COOH pH 10 buffer, with or without 10% ACN. The combination of water and 10% ACN in 10 mM NH4COOH pH 10 buffer was found to be the best washing solution. 4. Elution buffers were evaluated, including 10 mM NH4COOH pH 10 buffer with 30% or 80% ACN, and 80% ACN with 0.1%, 1%, and 5% FA. The 80% ACN with 5% FA produced the best results. Figure 2. Comparison of organic solvents. Left side is tirzepatide MRM (1,204.4 & 396.2) response after organic solvent precipitation, and right side is semaglutide MRM (1,029.4 & 1,238.1) response after organic solvent precipitation. ×1 0301235,370.30×10300.51.01.52,706.69×10301235,362.57Acquisition time (min)1.52.02.53.03.54.04.5Counts×10202461,280.68×102012412.89×1020241,055.21Acquisition time (min)1.52.02.53.03.54.04.5CountsTirzepatideSemaglutideACN:MeOH (1:1)ACNMeOHACN:MeOH (1:1)ACNMeOH 5 LC and MS conditions were all optimized, and the peptides' MRM transitions and collision energies were also optimized to achieve the best MS sensitivity for peptide quantitative analysis. The optimized source parameters are listed in Table 2. Quantitative analysis of tirzepatide in human plasma MassHunter Quantitative Analysis software was used to perform quantitative analysis of calibration curve and QC samples. Blank human plasma and tirzepatide had a low limit of quantification (LLOQ) of 0.05 ng/mL, as shown in Figure 3A. The calibration curve was linear up to 1,000 ng/mL with linear fit and 1/x2 weight, as shown in Figure 3B. Table 4 shows that all calibration points' precision and accuracy are with ± 15% acceptance criteria, demonstrating excellent assay performance. Figure 3. (A) Overlayed tirzepatide MRM (1,204.4 & 396.2) chromatograms of blank human plasma and lowest calibration point. (B) Calibration curve of tirzepatide in human plasma from 0.05 to 1,000 ng/mL. The insert is a zoomed-in snapshot of the lowest five calibration points. 0 0.51.01.52.02.53.03.54.04.55.05.5Acquisition time (min)0.51.01.52.02.53.03.54.04.55.05.56.06.57.07.58.08.59.09.5×101CountsA050–501001502002503003504004505005506006507007508008509009501,0501,00000.5–0.51.01.52.02.53.03.54.04.55.56.06.55.0y = 0.065646x + 0.006852 R2 = 0.9960R = 0.9999Relative responseRelative concentration×10100.51.01.52.02.53.03.54.04.55.06.05.500.51.01.52.02.53.03.54.0Relative responseRelative concentration×101B0.05 ng/mL tirzepatideBlank human plasma QC Concentration (ng/mL) 0.05 (LLOQ) 0.1 (Low QC) 5.00 (Mid QC) 800 (High QC) Run 1 Mean 0.0507 0.0915 5.0159 794.3824 % Bias 1.3 –8.5 0.3 –0.7 % CV 9.2 5.3 2.2 1.0 Run 2 Mean 0.0506 0.0997 4.8651 821.4187 % Bias 1.3 –0.3 –2.7 2.7 % CV 5.1 5.2 3.2 7.2 Run 3 Mean 0.0518 0.0952 5.350 723.432 % Bias 3.7 –4.8 7.0 –9.6 % CV 12.0 8.8 6.5 0.4 Interday Mean 0.0511 0.0955 5.077 779.744 % Bias 2.1 –4.5 1.5 –2.5 % CV 8.5 7.0 5.8 6.8 Table 5. Tirzepatide QC samples' (n = 4) precision and accuracy over three runs. The intra and interday analytical precision and accuracy of QC samples were determined from three independent runs performed over three days. The precision and accuracy results for tirzepatide in human plasma are shown in Table 4. All levels of QC samples (n = 4) met acceptance criteria of 15%, as recommended by the regulatory agency. The results demonstrate excellent assay performance using automated sample preparation. Conclusion The developed GLP-1 peptide quantification workflow, combining automation, the Agilent 1290 Infinity II bio LC, and Agilent 6495D LC/TQ, is an ideal platform for GLP-1 peptide quantitative analysis in biological matrices. The automated LC/MS workflow combines the advantages of both systems and provides excellent assay sensitivity and reproducibility. This workflow achieved an LLOQ of 0.05 ng/mL for tirzepatide by using 100 μL biological sample, and was linear up to 1,000 ng/mL, covering a 4.3 order of magnitude of dynamic range. In three individual qualification runs, the QC samples in all levels met regulatory agency requirements with excellent accuracy and precision. Another advantage of this workflow is that it is universal and can be applied to many other GLP-1 analogs. It also requires minimal method development time to greatly help drug discovery and development. References 1. Mounjaro- tirzepatide injection, solution. DailyMed. 13 May 2022. 2. Zepbound- tirzepatide injection, solution. DailyMed. 9 Nov. 2023 Calibration (ng/mL) 0.05 0.2 0.3 1 10 20 50 200 400 1,000 Run 1 0.0486 0.183 0.297 1.000 10.324 20.601 51.183 200.032 405.977 1,022.503 Run 2 0.0510 0.179 0.304 0.949 10.326 21.343 50.758 188.363 396.803 941.911 Run 3 0.0512 0.205 0.302 1.109 10.179 19.918 52.995 193.679 385.151 929.260 Mean 0.0503 0.189 0.301 1.019 10.276 20.621 51.645 194.025 395.977 964.558 % Bias 0.51 –5.51 0.36 1.93 2.76 3.10 3.29 –2.99 –1.01 –3.54 % CV 2.92 7.41 1.15 8.02 0.82 3.46 2.30 3.01 2.64 5.24 Table 4. Tirzepatide calibration curve (n = 3) results summary. www.agilent.com For Research Use Only. Not for use in diagnostic procedures. RA250627.417 This information is subject to change without notice. © Agilent Technologies, Inc. 2025 Printed in the USA, August 5, 2025 5994-8529EN