MALDI-TOF MS Protein Profiling Combined with Multivariate Analysis for Identification and Quantitation of Meat Adulteration

doi:10.21203/rs.3.rs-1537510/v1

Download PDF

Research Article

MALDI-TOF MS Protein Profiling Combined with Multivariate Analysis for Identification and Quantitation of Meat Adulteration

https://doi.org/10.21203/rs.3.rs-1537510/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The problem of adulteration and mislabeling in meat products has raised the public concerns globally. An easy-operation, fast and robust method that applicable to routine inspections is urgently needed. This study showed that the MALDI-TOF MS protein profiling of four meat species (beef, chicken, duck and pork) combining with partial least squares discriminant analysis (PLS-DA) discovered 57 feature peaks for their unambiguous differentiation. Among them, 36 were identified in Uniprot. Based on the linear relation between the intensities of feature peaks, the partial least squares regression was successfully applied to build the prediction models for determining the adulteration ratios of beef meat mixtures containing one of the other three species. Blind tests were applied to evaluate the method and the average prediction accuracy at 94.7% was achieved. Taking duck meat as the adulterant, the detection sensitivity of the method could be down to 5%. Moreover, the method has also been successfully applied to analyze market samples and the results were in agreement with the PCR method, showing the potential of its practical application for qualitative and quantitative analysis of adulterated beef products.

Meat adulteration

MALDI-MS

Protein profiling

PLS-DA

PLS regression

Fraud of meat and meat products has become a prevalent issue around the world (Temisak et al. 2021). The adulteration or substitution of higher price meat with lower price meat and mislabeling are common practices (Abbas et al. 2018; Ballin et al. 2009; Black et al. 2016; Temisak et al. 2021), especially in processed meat products. These illegal activities not only cause the economic losses of consumers (Hossain et al. 2020), but also lead to healthy risks such as the exposure of pathogens and allergens (Ballin 2010; Velioglu et al. 2018). To some extent, meat fraud may cause religious problems, for example, pork and pork products consumption is banned according to Kosher and Halal food laws (Alamprese et al. 2013; Hossain et al. 2020; Ortea et al. 2012). Therefore, it is important to develop a simple, sensitive, accurate and high throughput meat authentication method (Ortea et al. 2012; et al. 2020; Wang et al. 2018) to assist governments and enterprises in monitoring the quality and safety of meat products.

Indeed, various analytical approaches have been used to assure the meat authenticity. Polymerase chain reaction (PCR) technique is the most representative DNA-based method widely used for the meat authentication, due to its sensitivity, accuracy and reliability (Alikord et al. 2018; Ren et al. 2017). However, its high sensitivity can also bring the possibility of false positives when the animal origin is contaminated with other meat species during the production (Rahmati et al. 2016; Ren et al. 2017; Kumar et al. 2014). Moreover, the technique has some shortcomings, like high cost, time-consuming and complicated procedures (Ding and Xu 1999; Scott and Knight 2009; Velioglu et al. 2018). Infrared spectroscopy (IRS) and Raman spectroscopy (RS) are commonly used for analyzing characteristic lipids in meat, because of their rapid, simple and noninvasive properties. However, these methods are sensitive to moisture and temperature, thus affecting the detection reproducibility and accuracy (Sørensen et al. 2012). Mass spectrometry (MS), a highly sensitive and specific analytical method, has been widely used in food analysis (Alikord et al. 2018). Moreover, liquid chromatography-mass spectrometry (LC-MS) is one of the most commonly used MS techniques applied to meat species classification (Gatmaitan et al. 2021). However, the involvement of tedious sample pretreatment (like protein digestion) and slow chromatographic separation process create some limitations for the application of LC-MS in the large-scale analysis (Li et al. 2018; Prandi et al. 2019).

Matrix-assisted laser desorption/ ionization time of flight mass spectrometry (MALDI-TOF MS) is regarded as a rapid and reliable MS method (Laakmann et al. 2013; Lagacé-Wiens et al. 2012). Its ultra-high throughput at the analysis rate of 10,000 samples per day (Raad et al. 2016) outperforms most analytical methods, including PCR (Bookout et al. 2006; Taboada et al. 2017), thus being particularly suitable for the large-scale screening of imported food products. In practical applications, it allows the direct analysis of characteristic proteins in complex biological samples, and has been widely utilized in the routine identification of microorganisms (Maier et al. 2006; Sandrin et al. 2012; Santos et al. 2016; Wieser et al. 2011) and discovery of diseases biomarkers (Aslani et al. 2018; Kriegsmann et al. 2014; Rodrigo et al. 2014). It was also used in food identification like detection of milk adulterations (Sassi et al. 2015) and determination of the geographical origin of honey (Wang et al. 2009). In meat identification, it has been applied to the determination of the origin of pork, beef, horse meat, mutton and chicken via the protein-based untargeted analysis (Flaudrops et al. 2015). Moreover, MALDI-TOF MS was combined with the cluster analysis to build a protein fingerprint database of common edible fishes for the analysis of meat adulteration and substitution (Stahl and Schröder 2017). Recently, Shi et al have demonstrated that the specific protein N-glycosylation profiles determined by UPLC and MALDI-TOF MS can be applied to the qualitative and quantitative analysis of meat adulteration (Shi et al. 2019). However, the complicated and tedious sample preparation procedures, including the separation of protein N-glycans, followed by PNGase F enzyme digestion and chemical derivation/labelling, could limit its practical applications. Therefore, the further development of MALDI-TOF MS to enhance its robustness for the qualitative and quantitative analysis of meat adulteration is required.

Herein, a quantitative method using MALDI-TOF MS protein profiling combined with Partial least squares discriminant analysis (PLS-DA) and partial least squares (PLS) model, aiming to determine the adulteration ratio of meat is reported. In this approach, the MALDI-TOF MS method was employed to obtain the protein profiles of four meat species, including chicken, duck, pork, beef, and their two-species binary mixtures containing the expensive meat “beef” adulterated with the other three species, according to Table S1. Principal component analysis (PCA) was used to reveal the differential clustering of the pure meat and the binary mixtures. PLS-DA was applied to select feature peaks to discriminate four meat species, which were then identified through the protein database (Uniprot). Based on the linear relationship of the ion intensities of the feature peaks in the different binary mixtures, PLS model was developed to quantify the adulteration ratios (Aliaño-González et al. 2019; Shi et al. 2019). This integrated approach was successfully validated with the blind tests and applied to the analysis of market samples, showing its potential in performing qualitative identification of meat adulteration and quantitative determination of adulteration ratios.

Urea (Aladdin, 99.5%); Thiourea (Aladdin, 99%); Tris (Aladdin, 99%); distilled water (Watsons); Bradford protein assay kit (Biosharp, BL524A); sinapic acid (Aldrich, 98%); trifluoroacetic acid (Alfa Aesar, 99.5%) were used as received.

Four species of meat, including chicken, pork, beef and duck, are common edible meat, were selected in this study. Totally, 103 pure meat samples and 81 mixed-meat samples were collected and the different tissue parts of each species shown in Table S1 were used. Pork, chicken and duck were chosen to mix with beef meat in a binary mixture at the mass percentage of 0.0%, 25.0%, 50.0%, 75.0% and 100.0% for mimicking beef meat adulterated with other species. Moreover, the ternary mixture samples containing 50% beef meat mixed with different percentage of chicken meat (50–0%) and pork meat (0–50%) were used for investigating the ability of the approach in differentiating the ternary mixtures. As the price of duck meat is the lowest among many different meat species, it is likely to partly substitute beef products with duck meat in adulteration. Therefore, duck meat was mixed with beef meat in 3% and 5% to evaluate the detection limit of the approach. All of these meats in fresh were purchased at local markets or supermarkets. Five market samples for testing were purchased in a supermarket and a hotpot restaurant. All the meat samples were stored at -20℃ until use.

For each sample, 5 g of muscle tissues was chopped into paste, fascia and fat were discarded. Then, 0.5 g of meat paste was suspended in 10 mL protein extraction reagent (6M Urea, 1M Thiourea, 50mM Tris, at pH8.2) (Bargen et al. 2014) and homogenized with a homogenizer operated at 8000rpm for 2 min. The homogenate was centrifuged at 12000g for 1 min at 4℃. The supernatant was then filtered with a 0.45µm (polyvinylidene fluoride, PVDF) membrane filter, followed by dialysis in distilled water for 8 hours.

To minimize the effect of non-biological factors (such as different extraction efficiency and the protein content of different individuals) on the protein profiles, the protein content normalization is applied (Wu and Li 2016). The protein concentration of each dialyzed sample was measured using Bradford method with a microplate reader (Synergy H1 H1MF, Agilent, USA). The sample was diluted with 0.1% trifluoroacetic acid solution to 0.35 mg/ml (protein concentration). Quality control (QC) samples were prepared by mixing 20uL protein solution of each diluted sample for evaluating the stability of this method. The MALDI matrix solution, saturated sinapic acid (SA) solution, was prepared by dissolving SA in 30% acetonitrile solution with 0.1% trifluoroacetic acid. The sample solution (15uL) was mixed with saturated SA solution in 1:1 ratio. Then, 2.5µl of each mixture was applied on the MALDI sample plate (n = 5) and dried at room temperature, which was then analyzed with a MALDI-TOF mass spectrometer. Triplicated analyses were performed to ensure the intraday and interday reproducibility.

The Autoflex Speed MALDI-TOF/TOF mass spectrometer (Bruker Daltonics, Germany) was employed in this study. All the analyses were conducted in the positive ion mode with a mass range of 3,000 − 22,000 Da in a linear mode. The MS conditions are as follows: laser power: 99.6%; frequency: 500Hz; shot number: 20000 operated at random walk mode; delayed extraction time: 350 ns.

The MALDIquant package (Gibb and Strimmer 2012) of R programming language was used for data preprocessing, including spectra smoothing, baseline correction, alignment, technical replicates averaging and peak detection. Finally, a feature matrix was created and then a normalization method (MS total useful signal, MSTUS) (Wu and Li 2016) built in Microsoft Excel was used to normalize this feature matrix.

Multivariate data analysis was performed with SIMCA-P14.1 software (version 14.1; Umetrics AB, Umeå, Sweden). PCA was applied to reveal the general clustering and grouping trends among four meat species. PLS-DA was used to maximize the separation between samples, and the feature peaks responsible for the separation were found based on the corresponding V + S-plots (VIP ≥ 1, |p(corr)| ≥ 0.5, |p| ≥ 0.05) of PLS-DA (Almeida et al. 2013; Bao et al. 2016; Li et al. 2019; Peerbhay et al. 2013). Then, these feature peaks of four meat species were further confirmed in the MALDI-MS raw spectra based on a sufficient signal-to-noise ratio (S/N > 3) and a good reproducibility, and then identified through Uniprot (the protein database). To build a linear regression prediction model for determining the adulteration ratios in binary mixtures, PLS method is applied, in which, samples among four meat species were separated into training (2/3) and validation set (1/3). (Table S2).

To evaluate the prediction capability of our approach, blind test and market-sample test were adopted. Seven binary mixture samples of beef mixed with chicken, duck or pork in random ratios (sample 1 − 7) were prepared as the blind samples, and five beef products (M1, M2, M3, M4 and M5) purchased from a supermarket and a hotpot restaurant were used as the market samples. The same sample preparation procedures and MS data acquisition conditions mentioned above were adopted for the sample analysis. Moreover, the same multivariate data analysis, including PCA, PLS-DA and the linear regression prediction model based on partial least squares (PLS) method were applied to the data analysis.

The representative mass spectra of four species of meat, including chicken, duck, pork and beef, are shown in Fig. 1a. The differential protein profiles shown in the mass range of 3000 to 22000 Da illustrated the different protein composition of the four species. However, the protein profiles of different tissue parts collected from the same species showed similarity (Fig. S1). In the PCA analysis of the processed ion peaks derived from the protein profiles, the score plot (Fig. 1b) revealed a partial differential clustering of four meat species, and the tight clustering of the QC samples illustrated the high stability of the method.

Encouraged by the partial differential clustering of the four meat species, PLS-DA was further applied to select feature peaks for the complete differentiation of each meat species in the binary mixtures, as shown in Table S3. Using the criteria (VIP ≥ 1, |p(corr)| ≥ 0.5, |p| ≥ 0.05), totally 85 important variables with significant contribution to the separation of four meat species were regarded as feature peaks. The quality of the PLS-DA models was assessed by three parameters (R², Q², RMSECV) supplied by the software. The accumulated explanatory power of the first two principal components on the original variable (R² all greater than 0.98) showed a goodness-of-fit and the accumulation of predictive power (Q² all greater than 0.97) showed good model-predictability (Jiang et al. 2017). The root mean square error from cross-validation (RMSECV all smaller than 0.07) indicated that the important features that derived from the PLS-DA models are reliable (Aliaño-González et al. 2019; Li et al. 2015). After checking with MS raw spectra to further confirm the presence of these feature peaks, a total number of 57 characteristic ion peaks were obtained (Table S5). 36 out of the 57 were identified in Uniprot and are summarized in Table 1. In the PCA analysis of the 57 feature peaks derived from the protein profiles, the score plot (Fig. 1c) revealed a complete separation of four meat species at the confidence level of 95% (Fig. S2).

Table 1

36 Feature peaks identified in Uniprot.
MW of Characteristic Ion Peak [Da]	Identified Proteins	MW of Protein in Uniprot [Da]	Accuracy	Species
5531.1	Uncharacterized protein	5,531	99.99%	Beef
6363.9	Uncharacterized protein	6,362	99.97%	Beef
6521.2	Uncharacterized protein	6,524	99.97%	Beef
6675.6	40S ribosomal protein S29	6,677	99.98%	Beef
6947.9	Beta-defensin	6,947	99.99%	Beef
6951.1	Tracheal antimicrobial peptide	6,954	99.96%	Beef
8300.8	Homeodomain-only protein (Odd homeobox protein 1)	8,299	99.98%	Beef
9062.4	Cytochrome c oxidase subunit 7A1, mitochondrial	9,063	99.99%	Beef
9463.5	40S ribosomal protein S27-like	9,463	99.99%	Beef
18914.2	CS domain-containing protein	18,912	99.99%	Beef
18925.4	Uncharacterized protein	18,926	99.99%	Beef
18936.3	Cytochrome P450, family 8, subfamily B, polypeptide 1	18,936	99.99%	Beef
7572.6	Uncharacterized protein	7,575	99.97%	Chicken
9144.8	G_PROTEIN_RECEP_F1_2 domain-containing protein	9,143	99.98%	Chicken
18289.6	Aminoacyl-tRNA hydrolase	18,290	99.99%	Chicken
5548.2	short transmembrane mitochondrial protein 1	5,545	99.99%	Duck
8359.4	ATP synthase subunit e, mitochondrial	8,357	99.97%	Duck
8772.7	serine protease inhibitor Kazal-type 6-like	8,775	99.97%	Duck
9542.8	Molybdopterin synthase sulfur carrier subunit	9,545	99.98%	Duck
15094.6	28S ribosomal protein S12, mitochondrial	15,903	99.99%	Duck
5458.7	KRAB domain-containing protein	5,461	99.96%	Pork
6296.0	HMA domain-containing protein	6,295	99.98%	Pork
6322.7	Uncharacterized protein	6,323	99.99%	Pork
6330.2	Uncharacterized protein	6,329	99.98%	Pork
6554.4	Cytochrome b-c1 complex subunit 10	6,553	99.98%	Pork
6627.3	SAFB like transcription modulator	6,624	99.95%	Pork
6857.4	Small integral membrane protein 11A	6,854	99.95%	Pork
7955.7	Uncharacterized protein	7,957	99.98%	Pork
8285.9	Endoplasmic reticulum-golgi intermediate compartment 1	8,287	99.99%	Pork
9034.5	Tachykinin precursor 4	9,038	99.96%	Pork
9312.1	Uncharacterized protein	9,312	99.99%	Pork
9446.0	Chromosome 4 C8orf82 homolog	9,445	99.99%	Pork
16570.1	40S ribosomal protein S15	16,571	99.99%	Pork
18068.0	PhoLip_ATPase_N domain-containing protein;	18,067	99.99%	Pork
18870.0	Ig-like domain-containing protein	18,870	100.00%	Pork
18890.7	MYL2	18,891	99.99%	Pork

The mass spectra of beef mixed with chicken at the proportion of 0% (beef only), 25%, 50%, 75% and 100% (chicken only) are shown in Fig. 2a, while mass spectra of the binary mixtures containing beef mixed with other species, like duck and pork at the different proportions, are depicted in Fig. S3. It is worth noting that the intensities of some feature peaks changed gradually with the mixing proportion. For instance, the feature peak at m/z 5442 for beef, m/z 5388 for chicken, m/z 6192 for duck and m/z 6333 for pork, changed in a nearly linear fashion with their mixing proportion, as shown in Fig. 2b. Indeed, the similar trends were also observed in the score plot of the binary mixtures (Fig. 2c) derived from the PCA analysis of the 57 feature peaks. In which, the binary mixtures were located between their corresponding pure species, and shifting toward the pure species as the corresponding proportion increased. Moreover, the ternary mixtures containing 50% beef and different percentage of chicken and pork could be well separated in the PCA score plot (Fig. 2d). In addition, at a fixed percentage (50%) of beef, the positions of the ternary mixtures shifted toward the position of 50% beef-pork binary mixture as the proportion of pork meat increased, while the positions shifted backward to the 50% beef-chicken binary mixture as the proportion of chicken meat increased. Moreover, in the determination of detection limit, the PCA score plot derived from the 57 feature peaks showed that 5% duck meat mixed in the beef-duck binary sample could be well separated at the confidence level of 95%, but 3% duck meat mixed in the binary mixture could not (Fig. S4). The linear trend provided the basis for building a PLS linear regression prediction model to predict the adulteration ratios in beefs. As shown in Fig. 3, the PLS linear regression prediction model of beef mixed with chicken (a1 and a2), duck (b1 and b2) or pork (c1and c2) at different percentage showed a linear trend with the correlation coefficients of R² all greater than 0.99 (for the training data set) and R² all greater than 0.99 (for the validation). Moreover, the root mean square error of estimation RMSEE were all smaller than 2.3 (for the training data set) and the root mean square error of prediction RMSEP were all smaller than 2.7 (for the prediction), indicating the good prediction potential of the PLS model (Aliaño-González et al. 2019; Li et al. 2015; Shi et al. 2019).

To verify the accuracy of the integrated approach, i.e., MALDI-TOF MS combined with the multivariate statistical analysis, for the identification of meat species and quantitation of adulteration ratios in the binary mixtures containing beef, seven blind-test samples (sample 1 to 7) were employed, and the results are shown in the PCA score plot (Fig. 4a). Taking the sample 1 as an example, the location of the data point between the chicken and beef implicated that the beef meat was adulterated with chicken. The adulteration was also confirmed with the presence of feature peaks of chicken and beef in the sample 1, as shown in Table 2. The adulteration ratio in the beef-chicken binary mixture determined by the PLS linear regression prediction model was 73.6% (chicken) which was consistent with the true ratio (75.0% chicken) at the accuracy of 98.1%. Other blind test results for the different proportions of chicken, duck and pork in beef samples are shown in Table 2. The prediction accuracies were in the range of 91.6–98.1%, showing the good accuracy and reliability of our developed method.

Table 2
The presence of feature peaks of different species in blind-test samples (1–7) and market samples (M1, M2, M3, M4 and M5), and determination of adulteration ratios using the PLS linear regression models. (B: beef, C: chicken, D: duck and P: pork)
Sample No.	Presence of feature peaks (m/z)	Binary mixtures model	Ture ratio	Predicted ratio	Accuracy
1	Beef: 5443, 9463, 18912 Chicken: 4939, 6110, 6249, 9143	Chicken-Beef Mixed Model	B:25.0% C:75.0%	B:26.4% C:73.6%	98.1%
2	Beef: 5443, 5531, 6359, 6677, 6947, 8299, 9463, 18925 Chicken: 4939, 6110	Chicken-Beef Mixed Model	B:70.0% C:30.0%	B:68.0% C:32.0%	93.3%
3	Beef: 5443, 9463, 18926 Chicken: 4939, 6110, 6249, 9143, 18290	Chicken-Beef Mixed Model	B:20.0% C:80.0%	B:16.5% C:83.5%	95.6%
4	Beef: 6362, 9463 Duck: 4850, 5342, 5545, 6192, 6656, 6969, 7029, 7160, 7547, 8357, 8775, 15093	Duck-Beef Mixed Model	B:10.0% D:90.0%	B:14.4% D:85.6%	95.1%
5	Beef: 6358, 6362, 9463, 18926 Duck: 4850, 5342, 6192, 6397,	Duck-Beef Mixed Model	B:60.0% D:40.0%	B:57.1% D:42.9%	92.8%
6	Beef: 5443, 9463, 18926 Pork: 5000, 5461, 6295, 6329, 8287, 9445, 16571, 18067, 18891	Pork-Beef Mixed Model	B:30.0% P:70.0%	B:28.1% P:71.9%	97.3%
7	Beef: 5442, 6358, 18926 Pork: 5000, 5461, 6295, 6329, 6553, 6624, 8287, 9445, 16571, 18891	Pork-Beef Mixed Model	B:20.0% P:80.0%	B:26.7% P:73.3%	91.6%
M1	Beef: 5443, 5531, 6359, 6677, 6947, 8299, 9463, 18926	---	---	B:100%	---
M2	Beef: 5443, 6359, 6677, 8299, 9463, 18926	---	---	B:100%	---
M3	Beef: 5443, 6359, 6677, 7327, 9463, 18926	---	---	B:100%	---
M4	Beef: 5443, 6359, 6677, 6947, 8299, 9063, 9463, 18926	---	---	B:100%	---
M5	Beef: 9063, 9463, 18926 Chicken: 4939,7575	---	---	B:77.5% C:22.5%	---

To further confirm the practical application of this approach, four fresh beef products (M1, M2, M3 and M4) collected from a hotpot restaurant and one beef ball cooked sample (M5) collected from a supermarket were analyzed. The PCA score plot (Fig. 4b) showed that the fresh beef product is classified into beef without mixing with the other species at 95% confidence level (Fig. S5), which was also confirmed with MS data showing the absence of corresponding characteristic ion peaks of chicken, duck and pork meats. It is worth noting that the data point of the beef ball sample shown in the PCA score plot was away from the cluster of beef samples and approached to the chicken samples. The adulteration of the beef ball sample mixed with chicken meat was confirmed with the presence of their corresponding feature peaks (7 peaks for beef and 2 peaks for chicken, as shown in Table 2). Moreover, the adulteration species was confirmed with the PCR method, as shown in Fig. S6, and the adulteration ratio determined by the PLS linear regression model was 22.5%.

In summary, a simple, fast and accurate method, namely MALDI-TOF MS protein profiling combined with multivariate statistical analysis, was developed in this study for qualitative and quantitative analysis of beef meat adulterated with chicken, duck and pork. A total of 57 feature peaks (with 36 identified in Uniprot) were discovered for the authentication of the four species. Moreover, PLS was successfully applied to achieve the prediction of different adulteration ratios of the binary mixtures (i.e., beef meat mixed with chicken, duck or pork). The R² > 0.99, RMSEE < 2.3 (for the data set) and RMSEP < 2.7 (for the prediction) of the three PLS models (i.e., the beef-chicken, beef-duck and beef-pork binary mixtures) indicated the robustness of the models. The average prediction accuracy at 94.7% in the blind tests confirmed the reliability of our developed method. More importantly, the method has been successfully applied to analyze five market samples in real life. One sample containing chicken meat was discovered and confirmed with the PCR method. Using the PLS regression model, the adulteration percentage of chicken in the sample was determined to be 22.5%, showing the practical application of our developed method for beef products screening. This approach is sensitive in detecting the presence of other meat species (duck, chicken, pork meats) in the targeted meat products (beef products), and its detection sensitivity could be down to 5% of adulterant (taking duck meat as the example). The success of this approach is mainly due to its simplicity of incorporating the characteristic feature peaks of targeted meat and adulterants in the simple PCA model, which enables the unambiguous differentiation if the adulterants are present in the products. Therefore, it is anticipated that the approach can be extended to more adulterant meat species when the characteristic feature peaks of other meat species are detected and incorporated into the model.

Supplementary InformationThe online version contains supplementary material available at

DOI: 10.1021/xxxx.

Author Contribution K. Pu designed and performed the experiments and data analysis, and wrote the manuscript; J. Qiu performed the experiments and data analysis, and wrote the manuscript; J. Li performed the experiments; W. Huang assisted the preparation of blind-test samples, and revised the manuscript; X. Lai assisted the data analysis, and revised the manuscript; C. Liu checked the processed data; Y. Lin revised the manuscript; K.-M. Ng contributed to the conceptualization, supervision of the study, and manuscript revision.

Funding K. Pu acknowledges National Students' platform for innovation and entrepreneurship training program (Grant No.: 202110560019). K.-M. Ng acknowledges Shantou University for granting the research fund (STU Scientific Research Foundation for Talents, Grant No.: NTF19043). Y. Lin acknowledges the funding support of National Natural Science Foundation of China (Grant No.: 82071973). The funding supported by 2020 Li Ka Shing Foundation Cross-Disciplinary Research Grant (Project Number: 2020LKSFG01F) is gratefully acknowledged.

Data Availability Data will be made available on reasonable request.

Ethics Approval This article does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent Not applicable.

Conflict of Interest Keyuan Pu declares that he has no conflict of interest. Jiamin Qiu declares that she has no conflict of interest. Jiaying Li declares that she has no conflict of interest. Wei Huang declares that he has no conflict of interest. Xiaopin Lai declares that she has no conflict of interest. Cheng Liu declares that he has no conflict of interest. Yan Lin declares that she has no conflict of interest. Kwan-Ming Ng declares that he has no conflict of interest.

Abbas O, Zadravec M, Baeten V, Mikuš T, Lešić T, Vulić A, Prpić J, Jemeršić L, Pleadin J (2018) Analytical methods used for the authentication of food of animal origin. Food Chem 246:6–17. https://doi.org/10.1016/j.foodchem.2017.11.007
Alamprese C, Casale M, Sinelli N, Lanteri S, Casiraghi E (2013) Detection of minced beef adulteration with turkey meat by UV–vis, NIR and MIR spectroscopy. LWT Food Sci Techno 225–232. https://doi.org/10.1016/j.lwt.2013.01.027
Aliaño-González JM, Ferreiro-González M, Espada-Bellido E, Palma M, Barbero FG (2019) A screening method based on Visible-NIR spectroscopy for the identification and quantification of different adulterants in high-quality honey. Talanta 203:235–241. https://doi.org/10.1016/j.talanta.2019.05.067
Alikord M, Momtaz H, Keramat J, Kadivar M, Rad HA (2018) Species identification and animal authentication in meat products: a review. J Food Meas Charact 12:145–155. https://doi.org/10.1007/s11694-017-9625-z
Almeida RM, Fidelis HVC, Barata ESL, Poppi JR (2013) Classification of Amazonian rosewood essential oil by Raman spectroscopy and PLS-DA with reliability estimation. Talanta 117:305–311. https://doi.org/10.1016/j.talanta.2013.09.025
Aslani N, Janbabaei G, Abastabar M, Meis JF, Babaeian M, Khodavaisy S, Boekhout T, Badali H (2018) Identification of uncommon oral yeasts from cancer patients by MALDI-TOF mass spectrometry. BMC Infect Dis 18:1–11. https://doi.org/10.1186/s12879-017-2916-5
Ballin NZ, Vogensen FK, Karlsson AH (2009) Species determination–Can we detect and quantify meat adulteration? Meat Sci 83:165–174. https://doi.org/10.1016/j.meatsci.2009.06.003
Ballin ZN (2010) Authentication of meat and meat products. Meat Sci 86:577–587. https://doi.org/10.1016/j.meatsci.2010.06.001
Bao J, Liu F, Zhang C, Wang K, Jia X, Wang X, Chen M, Li P, Su H, Wang Y, Wan JB, He C (2016) Anti-melanoma activity of Forsythiae Fructus aqueous extract in mice involves regulation of glycerophospholipid metabolisms by UPLC/Q-TOF MS-based metabolomics study. Sci Rep 6:1–13. https://doi.org/10.1038/srep39415
Bargen C, Brockmeyer J, Humpf UH (2014) Meat authentication: A new HPLC–MS/MS based method for the fast and sensitive detection of horse and pork in highly processed food. J Agr Food Chem 62:9428–9435. https://doi.org/10.1021/jf503468t
Black C, Chevallier OP, Elliott CT (2016) The current and potential applications of Ambient Mass Spectrometry in detecting food fraud. Trac–Trend Anal Chem 82:268–278. https://doi.org/10.1016/j.trac.2016.06.005
Bookout LA, Cummins LC, Mangelsdorf JD, Pesola MJ, Kramer FM (2006) High-throughput real‐time quantitative reverse transcription PCR. Curr Protoc Mol Bio 73:15–8. https://doi.org/10.1002/0471142727.mb1508s73
Ding BH, Xu JR (1999) Differentiation of beef and kangaroo meat by visible/near-infrared reflectance spectroscopy. J Food Sci 64:814–817. https://doi.org/10.1111/j.1365-2621.1999.tb15918.x
Flaudrops C, Armstrong N, Raoult D, Chabrière E (2015) Determination of the animal origin of meat and gelatin by MALDI-TOF-MS. J Food Compos Anal 41:104–112. https://doi.org/10.1016/j.jfca.2015.02.009
Gatmaitan NA, Lin QJ, Zhang J, Eberlin SL (2021) Rapid Analysis and Authentication of Meat Using the MasSpec Pen Technology. J Agr Food Chem 69:3527–3536. https://doi.org/10.1021/acs.jafc.0c07830
Gibb S, Strimmer K (2012) MALDIquant: a versatile R package for the analysis of mass spectrometry data. Bioinformatics 28:2270–2271.
Hossain MA, Uddin SM, Sultana S, Wahab AY, Sagadevan S, Johan RM, Ali EM (2020) Authentication of Halal and Kosher meat and meat products: Analytical approaches, current progresses and future prospects. Crit Rev Food Sci 1–26. https://doi.org/10.1080/10408398.2020.1814691
Jiang W, Gao L, Li P, Kan H, Qu J, Men L, Liu ZQ, Liu Z (2017) Metabonomics study of the therapeutic mechanism of fenugreek galactomannan on diabetic hyperglycemia in rats, by ultra-performance liquid chromatography coupled with quadrupoletime-of-flight mass spectrometry. J Chromatog B 1044–104:58–16. https://doi.org/10.1016/j.jchromb.2016.12.039
Kriegsmann J, Kriegsmann M, Casadonte R (2014) MALDI TOF imaging mass spectrometry in clinical pathology: a valuable tool for cancer diagnostics. Int J Oncol 46:893–906. https://doi.org/10.3892/ijo.2014.2788
Kumar A, Kumar RR, Sharma BD, Gokulakrishnan P, Mendiratta SK, Sharma D (2014) Identification of species origin of meat and meat products on the DNA basis: a review. Crit Rev Food Sci 55:1340–1351. https://doi.org/10.1080/10408398.2012.693978
Laakmann S, Gerdts G, Erler R, Knebelsberger T, Arbizu PM, Raupach MJ (2013) Comparison of molecular species identification for North Sea calanoid copepods (Crustacea) using proteome fingerprints and DNA sequences. Mol Ecol Resour 13:862–876. https://doi.org/10.1111/1755-0998.12139
Lagacé-Wiens PR, Adam HJ, Karlowsky JA, Nichol KA, Pang PF, Guenther J, Webb AA, Miller C, Alfa MJ (2012). Identification of blood culture isolates directly from positive blood cultures by use of matrix-assisted laser desorption ionization–time of flight mass spectrometry and a commercial extraction system: analysis of performance, cost, and turnaround time. J Clin Microbiol 50:3324–3328. https://doi.org/10.1128/JCM.01479-12
Li AP, Yang L, Zhang LC, He SS, Jia JP, Qin X M (2019) Evaluation of Injury Degree of Adriamycin-Induced Nephropathy in Rats Based on Serum Metabolomics Combined with Proline Marker. J Proteome Res 19:2575–2584. https://doi.org/10.1021/acs.jproteome.9b00785
Li Y, Zhang Y, Li H, Zhao W, Guo W, Wang S (2018) Simultaneous determination of heat stable peptides for eight animal and plant species in meat products using UPLC-MS/MS method. Food Chem 245:125–131. https://doi.org/10.1016/j.foodchem.2017.09.066
Li B, Wang H, Zhao Q, Ouyang J, Wu Y (2015) Rapid detection of authenticity and adulteration of walnut oil by FTIR and fluorescence spectroscopy: A comparative study. Food Chem 181:25–30. https://doi.org/10.1016/j.foodchem.2015.02.079
Maier T, Klepel S, Renner U, Kostrzewa M (2006) Fast and reliable MALDI-TOF MS–based microorganism identification. Nat Methods 3:i–ii. https://doi.org/10.1038/nmeth870
Ortea I, Pascoal A, Cañas B, Gallardo JM, Barros-Velázquez J, Calo-Mata P (2012) Food authentication of commercially-relevant shrimp and prawn species: from classical methods to foodomics. Electrophoresis 33:2201–2211. https://doi.org/10.1002/elps.201100576
Peerbhay YK, Mutanga O, Ismail R (2013) Commercial tree species discrimination using airborne AISA Eagle hyperspectral imagery and partial least squares discriminant analysis (PLS-DA) in KwaZulu–Natal, South Africa. Isprs J Photogramm 79:19–28. https://doi.org/10.1016/j.isprsjprs.2013.01.013
Prache S, Martin B, Coppa M (2020) Review: Authentication of grass-fed meat and dairyproducts from cattle and sheep. Animal 14:854–863.
Prandi B, Varani M, Faccini A, Lambertini F, Suman M, Leporati A, Tedeschi T, & Sforza S (2019) Species specific marker peptides for meat authenticity assessment: A multispecies quantitative approach applied to Bolognese sauce. Food Control 97:15–24. https://doi.org/10.1016/j.foodcont.2018.10.016
Raad MD, Fischer CR, Northen TR (2016) High-throughput platforms for metabolomics. Curr Opin Chem Bio 30:7. https://doi.org/10.1016/j.cbpa.2015.10.012
Rahmati S, Julkapli MN, Yehye AW, Basirun JW (2016) Identification of meat origin in food products – A review. Food Control 68:379–390. https://doi.org/10.1016/j.foodcont.2016.04.013
Ren J, Deng T, Huang W, Chen Y, Ge Y (2017) A digital PCR method for identifying and quantifying adulteration of meat species in raw and processed food. Plos One 12: e0173567. https://doi.org/10.1371/journal.pone.0173567
Rodrigo MA, Zitka O, Krizkova S, Moulick A, Adam V, Kizek R (2014) MALDI-TOF MS as evolving cancer diagnostic tool: a review. J Pharmaceut Biomed 95:245–255. https://doi.org/10.1016/j.jpba.2014.03.007
Sandrin RT, Goldstein EJ, Schumaker S (2012) MALDI TOF MS profiling of bacteria at the strain level: a review. Mass Spectrom Rev 32:188–217. https://doi.org/10.1002/mas.21359
Santos CI, Hildenbrand LZ, Schug A K (2016) Applications of MALDI-TOF MS in environmental microbiology. Analyst 141:2827–2837. https://doi.org/10.1039/C6AN00131A
Sassi M, Arena S, Scaloni A (2015) MALDI-TOF-MS platform for integrated proteomic and peptidomic profiling of milk samples allows rapid detection of food adulterations. J Agr Food Chem 63:6157–6171. https://doi.org/10.1021/acs.jafc.5b02384
Scott M, Knight A (2009) Quantitative PCR Analysis for Fruit Juice Authentication Using PCR and Laboratory-on-a-Chip Capillary Electrophoresis According to the Hardy-Weinberg Law. J Agr Food Chem 57:4545–4551. https://doi.org/10.1021/jf9002686
Shi Z, Yin B, Li Y, Zhou G, Li C, Xu X, Luo X, Zhang X, Qi J, Voglmeir J, Liu L (2019) N-glycan profile as a tool in qualitative and quantitative analysis of meat adulteration. J Agr Food Chem 67:10543–10551. https://doi.org/10.1021/acs.jafc.9b03756
Sørensen MK, Petersen H, Engelsen BS (2012) An on-line near-infrared (NIR) transmission method for determining depth profiles of fatty acid composition and iodine value in porcine adipose fat tissue. Appl Spectrosc 66:218–226. https://doi.org/10.1366/11-06396
Stahl A, Schröder U (2017) Development of a MALDI–TOF MS-based protein fingerprint database of common food fish allowing fast and reliable identification of fraud and substitution. J Agr Food Chem 65:7519–7527. https://doi.org/10.1021/acs.jafc.7b02826
Taboada L, Sánchez A, Sotelo GC (2017) A new real-time PCR method for rapid and specific detection of ling (Molva molva). Food Chem 228:469–475. https://doi.org/10.1016/j.foodchem.2017.01.117
Temisak S, Thangsunan P, Boonnil J, Yenchum W, Hongthong K, Boll HO, Yata T, Rios-Solis L, Morris P (2021) Accurate determination of meat mass fractions using DNA measurements for quantifying meat adulteration by digital PCR. Int J Food Sci Tech 56:6345–6358. https://doi.org/10.1111/ijfs.15375
Velioglu HM, Sezer B, Bilge G, Baytur SE, Boyaci IH (2018) Identification of offal adulteration in beef by laser induced breakdown spectroscopy (LIBS). Meat Sci 138:28–33. https://doi.org/10.1016/j.meatsci.2017.12.003
Wang GJ, Zhou GY, Ren HW, Xu Y, Yang Y, Guo LH, Liu N (2018) Peptide biomarkers identified by LC–MS in processed meats of five animal species. J Food Compos Anal 73:47–54. https://doi.org/10.1016/j.jfca.2018.07.004
Wang J, Kliks MM, Qu W, Jun S, Shi G, Li QX (2009) Rapid determination of the geographical origin of honey based on protein fingerprinting and barcoding using MALDI TOF MS. J Agr Food Chem 57:10081–10088. https://doi.org/10.1021/jf902286p
Wieser A, Schneider L, Jung J, Schubert S (2011) MALDI-TOF MS in microbiological diagnostics—identification of microorganisms and beyond (mini review). Appl Microbiol Biot 93:965–974. https://doi.org/10.1007/s00253-011-3783-4
Wu Y, Li L (2016) Sample normalization methods in quantitative metabolomics. J Chromatog A 1430:80–95. https://doi.org/10.1016/j.chroma.2015.12.007

No competing interests reported.

SupplementaryInformationKM.pdf

Download PDF

Editorial decision: Major revision
13 Jun, 2022
Reviews received at journal
29 Apr, 2022
Reviewers agreed at journal
19 Apr, 2022
Reviewers invited by journal
15 Apr, 2022
Submission checks completed at journal
13 Apr, 2022
Editor assigned by journal
13 Apr, 2022
First submitted to journal
08 Apr, 2022

You are reading this latest preprint version

MALDI-TOF MS Protein Profiling Combined with Multivariate Analysis for Identification and Quantitation of Meat Adulteration

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results And Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1