Online trypsin digestion coupled with LC-MS/MS for detecting of A1 and A2 types of β-casein proteins in pasteurized milk using biomarker peptides

Abstract

Bovine A1-or A2-type β-caseins have attracted a growing interest due to their variation in beta-casomorphin-7 (BCM-7) formation, which may affect health. In the present work, identification and quantification of A1 and A2 types of β-casein proteins at the peptide level was achieved for the first time. An automated and online immobilized trypsin digestion system was employed for high throughput digesting of proteins into peptides. Tryptic peptides were separated and analyzed subsequently by liquid chromatography coupled to mass spectrometry platform. Two specific peptides ranging from the position of 49 to 97 in the peptide chain were selected for the identification and quantification of A1 and A2 β-casein, which covered the different amino acids between them. Synthetic isotopically labeled winged peptides were used for absolute quantification. Compared with traditional in-solution digestion, online digestion shortens digestion times from 2 to 24 h to 4 min. The limits of quantification (LOQ) of A1 and A2 β-casein in pasteurized milk are 0.8 and 2.4 µg/g, respectively. To further demonstrate the applicability of the proposed method, commercial pasteurized milk tests were performed with satisfactory results.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13197-022-05376-6.

Introduction

Bovine milk is a highly consumed food product due to its content of valuable nutrients. Three major types of α, β and κ-caseins comprise approximately 80% of milk proteins. Among them, β-casein constitutes about 40% of the total casein in milk (Nguyen et al. 2018; Giglioti et al. 2020). β-casein is a highly polymorphic protein. To date, a total of 12 different β-casein variants have been identified, including A1, A2, A3, B, C, D, E, F, G, H1, H2, and I (Caroli et al. 2016). Those β-casein variants can be grouped into two types: A1 type including A1, B, C, F, and G variants with the presence of histidine at position 67 of the sequence, A2 type including A2, A3, D, E, H1, H2, and I with proline at position 67 of the sequence (De Poi et al. 2020). This difference led to a key conformational change in the secondary structure of β-casein (Fuerer et al. 2019). A1 β-casein can release an exorphin β-Casomorphin-7 (BCM-7) during digestion by proteases, while A2 β-casein releases much less or even none of BCM-7 after digestion due to the high enzymatic resistance between proline and isoleucine (Givens et al. 2013; Asledottir et al. 2018).

Some epidemiological studies have suggested that β-CM7 and other related β-CMs may be associated with increased risk of diseases, such as Type 1 diabetes, heart disease, autism, and schizophrenia (Thiruvengadam et al. 2021; Kay et al. 2021). Despite the European Food Safety Authority concluded that “a cause-effect relationship between the oral intake of BCM-7 or related peptides and etiology or course of any suggested non-communicable diseases cannot be established.” (EFSA 2009), there is growing evidence of the potential relationship between BCM-7 and the increase in digestive discomfort and gastrointestinal issues (Caroli et al. 2016; Jarmołowska et al. 2019; Giglioti et al. 2020).

A1 and A2 β-casein gained significant attention not only from the scientific community but also from dairy consumers and producers. “A2” pasteurized milk is promoted commercially with much higher prices (approximately 3 times the price of common pasteurized milk) in a range of countries in recent years, including Australia, New Zealand, China, etc. A2 milk claimed that it is easier to digest for people who suffer from milk intolerances. Thus, there is a great interest in developing rapid and reliable analytical methods to certify and quantify A1 and A2 β-casein proteins.

Currently, there have been two kinds of methods for A1 and A2 analysis, at the gene level and at the protein or peptide level. For genetic polymorphism analysis of β-casein, the polymerase chain reaction (PCR) and the real-time PCR are the most frequently used tools (Cieślińska et al. 2019; Sebastiani et al. 2020). Methods at the protein level, include polyacrylamide gel electrophoresis (PAGE) (Duarte-Vázquez et al. 2018), isoelectric focusing (IEF) (Caroli et al. 2016), capillary electrophoresis (CE) (Feng et al. 2020), high-performance liquid chromatography (HPLC) (Nguyen et al. 2018), and high resolution HPLC-mass spectrometry (HPLC-MS) (Givens et al. 2013; Fuerer et al. 2019; Nguyen et al. 2020). Among those methods, mass spectrometry is by far the most reliable detection technology due to its high sensitivity and the ability to provide structural information (Valletta et al. 2021). Compared to intact protein detection, the MRM method at the peptide level provides a much easier analysis approach with higher sensitivity. Specific peptides of unique sequences could provide information as useful as proteins for clear-cut identification at the level of the amino acidic sequence (Tedeschi et al. 2018). This approach is very meaningful especially for the identification of homologous proteins with only a few different amino acids among hundreds, such as A1 and A2 β-casein proteins, since discrimination of short peptides with the single different amino acid is much easier. However, to the best of our knowledge, no method is currently available for simultaneously identifying and quantifying A1 and A2 types of β-casein at the peptide level in bovine milk.

Trypsin is the most widely used enzyme in proteomic research, which can specifically cleave peptide bonds after arginine and lysine residues. Although the in-solution digestion is predominantly used, it has several drawbacks, such as time consuming, trypsin autolysis, and intolerance to high temperatures or organic solvents (Šlechtová et al. 2017). Enzyme immobilization facilitates simple and automated operation, and increases enzyme stability and reusability (Toth, & Nguyen, 2017).The aim of this work was to establish a robust and high throughput method for identification and quantification of A1 and A2 types of β-casein proteins using online trypsin digestion coupled with LC-MS/MS platform. This approach is very suitable for discrimination of homologous proteins or authentication of closely related species.

Materials and methods

Materials and reagents

All solvents and chemicals used were of reagent grade unless otherwise mentioned. Water was obtained with a MilliQ element A10 System (S. Francisco, CA, USA). β-casein standard material (BioUltra, ≥98%), acetonitrile (HPLC purity), Tris base, hydrochloric acid, formic acid (analytical reagent grade), dithiothreitol (DTT), iodoacetamide (IAA), calcium chloride dihydrate, isopropanol and ammonium bicarbonate were purchased from Sigma-Aldrich (St. Louis, USA). Peptides IHPFAQTQSLVYPFPGPIHNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK, IHPFAQTQSLVYPFPGPIPNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK, internal standard materials DELQDKIHPFAQTQSL^*(¹³C₆, ¹⁵ N)VYPFPGPIHNSLPQNIPPLTQTPVVVPPFLQPEVMGVSKVKEAM (ISM-A1), DELQDIHPFAQTQSL^*(¹³C₆,¹⁵ N)VYPFPGPIPNSLPQNIPPLTQTPVVVPPFLQPEVMGVSKVKEAM (ISM-A2) were synthesized by Shanghai GLbiochem Co., Ltd. Protein LoBind tube (1.5 mL), low adsorption pipette tip (10-100 µL, 100-1000 µL), and acetate cellulose membrane (0.45 μm) were obtained from Eppendorf, Brand and Agela Techologies, respectively. Modified trypsin (sequencing grade) was obtained from Promega (Madison, USA). Mobile Phase Kit Dedicated for Perfinity iDP System including “Wash Solution Concentrate” kit, “Digest Buffer Concentrate” kit, and “Re-equilibration Concentrate” kit were purchased from Shimadzu (Shanghai) Global Laboratory Consumables Co., Ltd.

Commercial cow’s milk were legal inspection and quarantine commodities of import and export (see Table 2 for details). A1 milk was collected from cows that were genetically profiled as A1/A1 in Australia. Goat milk bought from import and export commodity supermarket were used as blank matrix after confirming that the two selected biomarker peptides of bovine A1and A2 β-casein were not detected in it.

Table 2.

Label claim contents and detected contents of A1 and A2 β-casein in pasteurized milk by online digestion and in-solution digestion coupled to HPLC-MS/MS analysis

Sample		Country	Label claim (β-casein, g/100 mL)	Detected content (g/100 mL)
				Online digestion		In-solution digestion
				A2	A1	A2	A1
A2 pasteurized milk	S-1	Australia	0.9	0.81	/	0.84	/
	S-2	New Zealand	0.9	0.83	/	0.87	/
	S-3	China	0.9	0.86	/	0.91	/
	S-4	Australia	1.0	0.93	/	1.01	/
	S-5	China	0.9	1.03	/	1.08	/
Ordinary pasteurized milk	S-6	New Zealand	/	0.57	0.39	0.61	0.40
	S-7	China	/	0.55	0.37	0.58	0.39

Preparation for online digestion

Mobile Phase A: mix of 490 mL water, 1 mL acetonitrile and 0.5 mL formic acid; Mobile Phase B: mix of 50 mL water, 450 mL acetonitrile and 0.5 mL formic acid; Mobile Phase C-A: “Wash Solution Concentrate” kit was dissolved in water and diluted to 1000 mL; Mobile Phase C-B: “Digest Buffer Concentrate” kit was dissolved in water, mixed with 20 mL isopropanol and diluted to 1000 mL (pH 8.4); Mobile Phase C-D: “Re-equilibration Concentrate” kit was dissolved in water, mixed with 250 mL isopropanol and diluted to 1000 mL.

Preparation of calibration standards

Commercial bovine β-casein was used as a standard material. According to previous reported, this standard material is a mixture of A1, A2 and B variants. Both of B and A1 variants belong to the A1 type with the presence of histidine at the position 67 of the sequence (De Poi et al. 2020). Thus, the A1 and A2 types of β-casein in the standard material were quantified using synthetic marker peptides at first. The result showed that the β-casein standard material contains approximately 60% of the A2 type β-casein and 40% of the A1type β-casein, which was consistence with previously reported (Keith 2020). β-casein protein standard is diluted with water to generate a series of seven standard solutions, which containing 0.8, 1.6, 4, 8, 20, 40 and 80 µg/mL A1 β-casein and 1.2, 2.4, 6, 12, 30, 60 and 120 µg/mL A2 β-casein, repectively.

Sample preparation

5 g of commercial pasteurized milk were dissolved and diluted to 50 mL using 50 mM Tris-HCl (pH=8.0 ± 0.2), with the final concentration of total protein at approximately 3 mg/mL determined by Bradford assay (Sigma-Aldrich, St. Louis, USA). Aliquots of 500 µL solution were spiked with 50 µL 100 ug/mL internal standard material. The mixtures were reduced and alkylated as previously reported (Gu et al. 2018) for subsequent online digestion.

Online trypsin digestion and HPLC-MS/MS analysis

The integrated Digestion Platform (iDP) included mainly a LC-20AD pump with one quaternary low pressure gradient proportional valve, two LC-20ADXR pumps, one autosampler, one column oven compartments and four 2-position 10-port valves. The LC-20AD pump was used for the delivery of the Perfinity trypsin digestion buffer and column regeneration buffer. The two LC-20ADXR pumps delivered 0.1% FA in water (eluent A) and 0.1% FA in acetonitrile (eluent B) for binary gradient elution.

Perfinity trypsin column (2.1 mm × 30 mm) was operated at 40 °C with a flow rate of 50 µL/min. An additional 4 min were allotted to purge the Perfinity trypsin column and connection lines. The unreacted proteins, cleavage products and internal standards were carried together directly to the desalting column (Halo® C18 4.6 mm × 5 mm, 2.7 μm guard column). After the 4 min digestion period, the desalting column was idled between closed lines while the Perfinity trypsin column was regenerated. During the consecutive sample run, cleavage products from the previous run were eluted from the desalting column to the analytical column (A C18 Kinetex 100 mm × 2.1 mm, 1.7 μm particles, Phenomenex, CA, USA), while the second desalting column was used for handling the newly injected sample. In Run 1, the peptide is eluted from the desalting column 1 onto the analytical column. In Run 2, freshly cut peptides from the trypsin column are loaded onto the desalting column 2 in parallel. The analytical separation was performed at 50 °C with a flow rate of 400 µL/min over a stepwise gradient ranging from 5 to 95% acetonitrile containing 0.1% formic acid. Gradient procedure: 4.1 min, 5% B; 6.0 min, 5% B; 14 min, 95% B; 16 min, 95% B; 16.1 min, 5% B, 18.54 min 5% B.

Multiple reaction monitoring (MRM) experiments were performed on a HPLC-MS/MS 8060 (Shimadzu, Japan) operated with a heated electrospray ionization probe in positive ion mode using the following settings: Ion source interface voltage 1500 V, Ion source interface temperature 300 °C, heating gas 10 L/min, drying gas 10 L/min, nebulizing gas 3.0 L/min, DL tube temperature 200 °C, heating module temperature 400 °C. The native and isotopically labeled internal standard peptides chromatograms were acquired by multiple reaction monitoring (MRM) mode with mass transitions summarized in Table 1.

Table 1.

List of marker peptides selected for the MRM acquisition method with the relevant three-most intense transitions

Protein	Peptide sequence	Precursor(charge state) [m/z]	Product ions (fragments) [m/z]	Collision energy (eV)
A1 β-casein	IHPFAQTQSLVYPFPGPIHNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK	1072.2(5+)	714.9(y132+)*	28.0
			1428.8(y13)	24.5
			846.5(y8)	42.0
	IHPFAQTQSL(13C6,15 N)VYPFPGPIHNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK	1073.6(5+)	714.9(y132+)*	28.0
			1428.8(y13)	26.5
			846.5(y8)	41.0
A2 β-casein	IHPFAQTQSLVYPFPGPIPNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK	1064.2(5+)	714.9(y132+)*	28.0
			1428.8(y13)	24.0
			846.5(y8)	42.0
	IHPFAQTQSL(13C6,15 N)VYPFPGPIPNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK	1065.6(5+)	714.9(y132+)*	28.0
			1428.8(y13)	27.0
			846.5(y8)	39.5

Detection of A1 and A2 β-casein in milk

The detection limits (LOD) and the quantification limits (LOQ) were defined as the minimum concentration giving a signal-to-noise ratio (S/N) of 3 and 10 for the most intense MRM transition of the marker peptide. Linear dynamic range was investigated starting from 0.8 to 80 µg/mL for A1 and 1.2-120 µg/mL for A2. The intra-day repeatability was evaluated by performing three independent pretreatments and three HPLC-MS/MS injections for each enzymatic hydrolysate in the same day. The inter-day repeatability was calculated on 3 days by performing three independent pretreatments of the same pasteurized milk as above and three HPLC-MS/MS injections for each enzymatic hydrolysate. Recovery of the whole method was evaluated by analyzing in triplicate a matrix fortified sample (goat milk) at three concentration levels.

Traditional in-solution tryptic digestion

For optimization of mass spectrometry conditions and comparation of quantitative results, the samples were also analyzed by an off-line tryptic digestion coupled HPLC-MS/MS method. Briefly, aliquots of 200 µL diluted sample (2.4) mixed with 50 µL 100 ug/mL internal standard material were reduced with 10 µL DTT solution (500 mM) for 30 min at 75 °C. After cooling the sample at room temperature, an alkylation was performed by adding 30 µL of IAA solution (500 mM). The alkylation was left in the dark for 30 min at room temperature. Subsequently, NH₄HCO₃ (150 µL, 500 mM), CaCl₂ (10 µL, 100 mM) and trypsin (50 µL, 500 µg/mL) were added and incubated for 16 h at 37 °C. The digestion was stopped by adding 10 µL formic acid and adjusted to 1.0 mL by water. Samples were injected into the HPLC-MS/MS after filtered with 0.22 μm cellulose acetate membrane.

Statistical Analysis

All experiments were performed at least twice. Mass spectrometry measurements were repeated in triplicates for each sample, and recovery tests were carried out in six replicates. Experimental results were recorded as mean ± standard deviation.

Results and discussion

Online protein digestion

In-solution trypsin digestion followed by mass spectrometry identification is predominantly used in proteomic studies. The standard procedure of in-solution trypsin digestion typically requires 2–24 h (Toth et al. 2017), which is very time-consuming. As a solution to this problem, the Perfinity trypsin column containing immobilized trypsin was utilized in the present work. The use of an immobilized enzyme column can prevent autodigestion, increase efficiency, and shorten digestion times compared with in-solution digestion (Frejie et al. 2005).

The schematic illustration of the online immobilized column trypsin digestion and LC-MS/MS analysis is shown in Fig. 1. During the first half of the run, the digestion and trypsin column wash period are carried out, and the LC-MS/MS system was idle. During the second half of the run, the LC-MS/MS data acquisition was performed and the trypsin column was idle. The addition of an extra ten-port valve with 2 trapping columns allowed simultaneous digestion and LC-MS/MS analysis. Each consecutive sample was digested and trapped on one desalting column while the previously digested and desalted sample was eluted from the other desalting column and analyzed by LC-MS/MS. The diluted milk sample after reduction and alkylation was directly added to the vial for subsequent digesting. After digestion, the enzymatic hydrolysates were desalted through a desalting column, and then subjected to a reversed-phase column for chromatographic separation. Finally, the separated peptides were analyzed by mass spectrometry. It only takes 18.54 min to complete the entire analysis process including online digestion. The employment of the immobilized trypsin column enhanced enzyme stability and activity, facilitated operation, and improved the automation and throughput of the analysis.

Fig. 1.

Schematic representation of the immobilized trypsin column digestion and LC-MS/MS platform

Optimization of protein digestion

Optimization of digestion conditions was used the Perfinity iDP by varying parameters within the software. Digestion temperature and time are two important factors affecting online immobilized digestion efficiency. At low temperature, enzyme-substrate binding rate decreased and enzyme activity was reduced, while at high temperature, the enzyme was denatured and the enzyme would lose its activity (Cheison et al. 2010). Effect of temperature on mass signal response ranging from 30 to 50 °C with the digestion buffer flow rate of 0.025 mL/min and reaction time for 4 min was investigated. As shown in Fig. 2A, when increasing the digestion temperature from 30 to 40 °C, it was observed that the signal response of A1 and A2 biomarker peptide increased and then decreased with higher temperature. Hence, 40 °C were chosen for subsequent experiments. Digestion time ranging from 1 to 8 min were studied at 40 °C with the digestion buffer flow rate of 0.025 mL/min. The signal response gradually increased from 1 to 4 min, and then remained almost constant between 4 and 8 min (Fig. 2B). Therefore, a digestion time of 4 min was selected for subsequent experiments.

Fig. 2.

Ion abundance of A1 and A2 biomarker peptide at A different digestion temperatures and B with different digestion times

Selection and validation of peptide markers

In this work, we expected to establish an online trypsin digestion method for identifying and quantifying A1 and A2 β-casein based on marker peptides. Since A1 and A2 β-casein differ only at amino acid position 67, the marker peptides must contain the amino acid at position 67. Owing to the trypsin digestion sites are lysine and arginine, according to the amino acid sequences of A1 and A2 β-casein, it is speculated that there is only one possible marker peptide, IHPFAQTQSLVYPFPGPIHNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK for A1 β-casein, and IHPFAQTQSLVYPFPGPIPNSLPQNIPPLTQTPVVVPPFLQPEVMGVSK for A2 β-casein. Both of the peptides were fragments 49 to 97 of the mature protein. To facilitate writing and reading, they were referred to as A1 β-casein f (49->97) and A2 β-casein f (49->97) in the following text.

The uniqueness of these two peptides was verified using basic local alignment search tool (BLAST) analyses against the UniProt databases. For A1 β-casein f (49->97), the Blast results return 100% identity only to β-casein of the cow (Bos Taurus), indicating that it can be used as a biomarker peptide of bovine A1 β-casein. For A2 β-casein f (49->97), the Blast results return 100% identity to β-casein of a cow (Bos Taurus) and wild yak (Bos Mutus) (as shown in Figure S1 and S2 in the Supplementary Material). It is well known that wild yak milk is more rare and precious, it is very unlikely to label wild yak milk as bovine milk or mix it with bovine milk. Therefore, A2 β-casein f (49->97) can be used as a biomarker peptide of bovine A2 β-casein in the testing of products labeled as bovine milk.

Specificity and robustness of selected peptide biomarkers

The MRM acquisition method was established using these two biomarker peptides to verify their specificity and robustness, as shown in Table 1. Precursor ion masses, product ion masses, and fragmentation potentials were generated using Skyline software version 20.1.0.31 (https://skyline.gs.washington.edu). Pasteurized milk samples are complex and contain multiple proteins and other components. Therefore, the specificity, sensitivity and repeatability of the developed method should be checked in processed products. Heat treatment will induce some glycation of proteins caused Maillard reaction (Arena et al. 2017). This may block the lysines which are also the major cleavage site by trypsin. In order to investigate the influence of glycation on peptides, enzymatic digests of conventional commercial pasteurized milk processed bovine milk were anaylsed using ultra-performance liquid chromatography-quadrupole/electrostatic field orbitrap high resolution mass spectrometer (UPLC-Q Exactive HRMS) system (Thermo Fisher, USA). The collected full MS/dd-MS² scan data sets were used for peptide identification with ProteinPilot Software (see Supplementary Material for details). Results of peptide matching indicated that glycation occurred on neither of the biomarker peptides (as shown in Figure S3 and S4). In addition, since fragment ions of glycated peptides are associated with poor MS intensity (Arena et al. 2010), the results showed that the fragment ions of both selected peptides have good response intensity (see details in Figure S5 and S6), suggesting that glycation has not occurred at lysine within the two peptides.

Optimization of MS parameters

The mass spectrometry parameters need to be optimized to improve the sensitivity and stability of MS signals. Collision energy is one of the most important affecting factors. Skyline software was used to build methods for optimizing collision energy. Using the ion channel 1072.2⁵⁺→714.9²⁺ of the A1 biomarker peptide f (49->97) as an example, the optimization of collision energy is centered at -30 eV with a step size of 0.5 eV and 6 steps left and right. Thus, a total of 13 different collision energies were applied. As shown in Fig. 3 A, the maximum peak area was obtained at -28.0 eV. The optimization results of the A2 biomarker peptide are shown in Fig. 3B.

Fig. 3.

Optimization of collision energy for ion channel 1072.2⁵⁺→714.9²⁺ of the A1 biomarker peptide (A) and ion channel 1064.2⁵⁺→714.9²⁺ of the A1 biomarker peptide (B). The optimization of collision energy is centered at -30 eV with a step size of 0.5 eV and 6 steps left and right

Interface Setting Support software (Shimadzu, Japan) was used to optimize mass spectrometry parameters such as nebulizing gas flow, heating gas flow, dry gas flow, DL temperature, interface voltage, and interface temperature. The nebulizing gas flow rate ranging from 1.5 to 2.5 L/min, heating gas flow rate and drying gas flow rate ranging from 8 to 12 L/min were optimized by cross combination. A total of 12 sets of gas flow rates (1.5, 8, 8; 1.5, 8, 10; 1.5, 8, 12; 1.5, 10, 8; 1.5, 10, 10; 1.5, 12, 8; 2.5 ,8, 8; 2.5, 8, 10; 2.5, 8, 12; 2.5, 10, 8; 2.5, 10, 10; 2.5, 12, 8 mL/min) were optimized. A total of 8 conditions combined of DL tube temperature (150, 200, 250 °C) and ion source interface temperature (200, 250, 300, 350 °C) were investigated. Four interface voltages (0.5, 1, 2, 3, and 4 kV) were examined. The results showed that the optimized nebulizing gas, heating gas, drying gas flow rate, DL tube temperature, ion source interface temperature and interface voltage were 2.5, 8.0, 10.0 L/min, 200 °C, 350 °C and 2 kV, respectively.

Under optimized experimental conditions, the MS parameters of each peptide are listed in Table 1, and the extracted ion chromatograms of the marker peptides for A1 and A2 types of β-casein are shown in Fig. 4.

Fig. 4.

Extracted ion chromatograms of the marker peptides for A1 (A) and A2 (B) β-casein

Evaluation of the analytical method performances

In order to correct the recovery losses caused by ionization efficiency, sample preparation and protein digestion, a proper internal standard (preferably isotope-labeled protein) should be used. However, the full length isotope-labeled protein is hard to obtain and expensive to produce. The extended peptide precursor which contains the internal standard peptide and the same tryptic cleavage site with the analyzed protein is an alternative option (Benesova et al. 2021). The extended peptide precursors were designed as internal standard materials ISM-A1, ISM-A2 (as shown in Sect. 2.1).

A standard curve was established with the mass concentration as the abscissa (X) and the peak area ratio of the marker peptide and the internal standard peptide as the ordinate (Y). The results were as shown in table S1 in supplementary materials. The standard curve of each peptide showed a good linear relationship (R²≥0.99). It is finally determined that the LOQ of A1 and A2 β-casein are 0.8 and 1.2 µg/g, respectively. To the best of our knowledge, this is the first quantitative study of A1 and A2 β-casein using marker peptides by online digestion.

Recovery of the whole analytical method was assessed using goat milk as a blank matrix. Nine portions of goat milk were spiked with A1 and A2 β-casein standard at three concentration levels with three portions per concentration level. The results showed that the proposed method had a good recovery and precision even at low spiked levels.

The reproducibility of the online digestion method is an issue of great concern, since the immobilized enzyme column needs to be washed, stored, and activated every time after use. To monitor system reproducibility, the β-casein standard was interspersed between experimental samples, at approximately every 10 samples. Good results were obtained also for method reproducibility with intra-day reproducibility ranging from 2.1 to 8.7% (n=3) and inter-day reproducibility lower than 15% (n=3).

Commercial A2 pasteurized milk analysis

Commercial pasteurized milk samples from different regions were tested to further verify the applicability of the established method. In order to evaluate the reliability of online digestion, all the samples were analyzed using both two methods, online trypsin digestion coupled to HPLC-MS/MS and in-solution digestion coupled to HPLC-MS/MS. The results showed that only A2 β-casein f (49->97) was detected in A2 pasteurized milk, while both of A1 β-casein f (49->97) and A2 β-casein f (49->97) were detected in ordinary pasteurized milk (typical results were shown in Figures S7, S8). The contents of A1 and A2 β-casein were determined, as shown in Table 2. The detected contents by online digestion coupled to HPLC-MS/MS were in good agreement with the label claim values. Although the results of in-solution digestion are closer to the label value, the online digestion method has clear advantages of being much faster regarding sample preparation and analysis. These results demonstrate that the present online digestion coupled with LC-MS/MS method based on analysis of tryptic peptides can be reliably used for A2 β-casein quantification.

Conclusions

Automated online trypsin digestion coupled with LC-MS/MS method for identification and detection of A1 and A2 β-casein proteins at the peptide level was developed. The tryptic fragments A1 β-casein f (49->97) and A2 β-casein f (49->97) were selected as biomarker peptides for A1 and A2 β-casein, respectively. The proposed method requires only 4 min for online trypsin digestion, greatly improving the analysis throughput. An additional advantage of the present method was the use of extended peptide precursors as internal standard material for precise quantification. Finally, it was successfully applied to determine the content of A1 and A2 β-casein in commercial pasteurized milk with comparison to the traditional in-solution digestion. Compared with the previous methods for β-casein protein analysis, the proposed method offered simple and automated operation, high-throughput analysis and high sensitivity with satisfactory accuracy and reproducibility.

Supplementary information

Below is the link to the electronic supplementary material.

Funding

National Key Research and Development Plan Project of China (No. 2018YFC1603603); Shanghai Agricultural Science and Technology Project (No. 19,391,901,500); Science and Technology Joint Project of the Yangzte River Delta (19,395,810,100); Shanghai Professional Technical Service Platform Project (20DZ2291900) are gratefully acknowledged.

Data availability

Data transparency.

Code availability

Software application or custom code.

Declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Consent to participate

All authors have read and approved the MS; and, that all are aware of its submission to JFST.

Consent for publication

All authors are aware and consent of its publication in JFST, including publishing an individual’s data or image.

Ethics approval

Not Applicable.