Assessment of tenderness of aged bovine gluteus medius muscles using Raman spectroscopy
Introduction
Flavor, tenderness and juiciness are the most important quality traits for the overall liking of beef (Hunt et al., 2014). Studies show that customers are willing to pay a higher price for beef cuts with guaranteed tenderness (Miller, Carr, Ramsey, Crockett, & Hoover, 2001) and, accordingly, that tenderness is ranked higher than price when it comes to the purchasing decision (Reicks et al., 2011). Consequently, the management of tenderness is an issue especially for the red meat production due to the storage time required for aging. On average, beef is aged for 28 days before retail sale in the U.S. but storage times up to 67 days are used in practice (Guelker et al., 2013). During this time, the tenderness of meat is changed by proteolytic degradation of proteins and by protein oxidation (Huff Lonergan, Zhang, & Lonergan, 2010).
While tenderness can only be directly assessed by human panelists this procedure is expensive and yields subjective results. Therefore, mechanical methods are used as objective measures for tenderness but again these methods are time-consuming, destructive and laborious. The Warner-Bratzler shear force (WBSF) test and the slice shear force test are the most often applied mechanical methods. A Warner-Bratzler apparatus is used to measure the force which is required to cut the cooked sample with a V-shaped blade in relation to the distance the blade has cut into the meat. Conventionally, the maximum force recorded during the cutting process is reported. The results of this WBSF measurement usually show discrepancies with tenderness values estimated by a trained sensory panel with correlation coefficients between − 0.2 and − 0.9 (Destefanis et al., 2008, Peachey et al., 2002, Shackelford et al., 1999). This discrepancy may be explained by the finding that tenderness is mainly evaluated subjectively (Wezemael, Smet, Ueland, & Verbeke, 2014). Despite its disadvantages, the WBSF is still an accepted objective measurement for meat tenderness (Lorenzen et al., 2010).
Several publications were addressing a replacement for the WBSF method (Damez and Clerjon, 2008, Prieto et al., 2009, Xiong et al., 2014). Table 1 shows an overview of – mostly spectroscopic – studies which used beef or lamb cuts and at least 40 samples. On the one hand, good results are reported in some studies using near-infrared (NIR) reflectance spectroscopy with R2's during calibration ranging from 0.5 to 0.8. On the other hand, only moderate correlations or no useful predictions are reported in other studies. A change from reflectance to transmittance spectra does not significantly improve the correlations (Leroy et al., 2003). For visible-NIR hyperspectral imaging (496–1036 nm) correlations with R2 = 0.45 are shown to predict cooked beef tenderness (Cluff et al., 2008).
As NIR spectra revealed potential for the identification of tender beef according to WBSF (Shackelford et al., 2004, Shackelford et al., 2005), a portable near-infrared reflectance (NIR-R) instrument was developed for beef carcasses (Rust et al., 2008). This and a similar instrument are shown to discriminate tender from tough beef ribeye (n = 1155) (Shackelford, King, Wheeler, & Koohmaraie, 2012a), strip loin (n = 467) (Shackelford, Wheeler, & Koohmaraie, 2012b) and longissimus lumborum (LL) samples (n = 768) (Rust et al., 2008),
Correlations of up to R2 = 0.46 are reported for non-spectroscopic techniques such as electrical impedance and conductivity with WBSF of bovine longissimus dorsi (LD) muscles (n = 47) (Byrne, Troy, & Buckley, 2000).
Promising results are reported with Raman spectroscopy. This method gives insight into the molecular composition of a sample by means of vibrational spectroscopy. The scattered light can reveal biochemical and structural information. PLSR (partial least squares regression) correlations of Raman spectra of roasted beef silversides with shear force (SF) are promising with R2 = 0.75 and root mean squared error of cross-validation RMSECV = 6.3 N (Beattie, Bell, Farmer, Moss, & Patterson, 2004). In a more recent study, a portable Raman scanner with potential for the application in processing plants is shown to measure shear force and cooking loss of frozen and thawed sheep muscle (Schmidt, Scheier, & Hopkins, 2013). The Raman spectra of 140 LT (M. longissimus thoracis) and LL samples from two different origins correlate well with the SF yielding R2 = 0.79 and 0.83 for the two data sets separated according to origin. However, while freezing/thawing was required for the transport of the samples this is not reflecting the typical process flow of meat production. Therefore, subsequent work is evaluating the ability of Raman spectroscopy for predicting SF from Raman spectra of meat cuts that have never been frozen. A study with semimembranosus (SM) muscles, shows only a limited ability for the prediction of SF of aged lamb from Raman measurements performed on day 1 post mortem (p.m.) with cross-validated coefficient of determination R2 = 0.27 whereas no correlation is found for the LL (Fowler et al., 2014a, Fowler et al., 2014b).
While clear potential of Raman spectroscopy for non-destructive and rapid determination of SF has been shown for roasted beef and for frozen and thawed lamb, data on beef that has not been frozen or grilled is lacking. Therefore, the aim of this study is to evaluate whether the Raman spectra measured on wet aged, raw beef samples can predict the measured WBSF.1
Section snippets
Meat samples
In total, samples from 175 young bulls were collected from commercial abattoirs over two periods of 5 months each (one for calibration and one for validation, both separated by two years). Ten samples were measured on average per day every two weeks which required ten sampling days for calibration and eight for validation. The animals had an age of 18–24 months, a slaughter weight of 360–400 kg and they were from different origin (90% Simmentals and 10% mixed origin, mostly brown cattle). Three to
Shear force results
The mean SF values of the calibration and validation (see Table 3) are in agreement with values of 36.3 N found in the literature (Belew, Brooks, McKenna, & Savell, 2003). The samples aged at − 1 °C exhibit a slightly increased shear force compared with the samples aged at 7 °C, but this difference is non-significant (α < 0.01) for calibration and validation samples (1.9 N and 1.4 N, respectively). There is furthermore no significant difference (α < 0.01) between the SF values of the calibration and of
Conclusion
To our knowledge, this is the first study showing the principal feasibility of a non-invasive assessment of the tenderness of aged, bovine gluteus medius muscles with a portable Raman device. The aging at 7 °C yielded on average smaller shear force values than the aging at − 1 °C (1.9 N and 1.4 N differences for calibration and validation samples, respectively). However, this small improvement hardly justifies the increased risk of microbial spoilage when aging at higher temperature.
The PLSR
Acknowledgment
The Research Centre of Food Quality is funded by the European Regional Development Fund (ERDF). We thank Michael Pabst for excellent cooperation and Kaufland Fleischwaren SB GmbH & Co. KG for providing the beef samples. Johannes Linder, Lars Philipp and Christian Schütz as well as Christopher Berg, Oliver Miehlich and Adrian Beck deserve our appreciation for performing the shear force measurements and Thomas Kador for assistance with the Raman measurements.
References (41)
- et al.
Extended multiplicative signal correction in vibrational spectroscopy, a tutorial
Chemometrics and Intelligent Laboratory Systems
(2012) - et al.
Preliminary investigation of the application of Raman spectroscopy to the prediction of the sensory quality of beef silverside
Meat Science
(2004) - et al.
Warner-Bratzler shear evaluations of 40 bovine muscles
Meat Science
(2003) - et al.
Postmortem changes in muscle electrical properties of bovine M. longissimus dorsi and their relationship to meat quality attributes and pH fall
Meat Science
(2000) - et al.
Relationship between collagen characteristics, lipid content and raw and cooked texture of meat from young bulls of fifteen European breeds
Meat Science
(2011) - et al.
Meat quality assessment using biophysical methods related to meat structure
Meat Science
(2008) - et al.
Relationship between beef consumer tenderness perception and Warner-Bratzler shear force
Meat Science
(2008) - et al.
Predicting tenderness of fresh ovine semimembranosus using Raman spectroscopy
Meat Science
(2014) - et al.
Raman spectroscopy compared against traditional predictors of shear force in lamb m. longissimus lumborum
Meat Science
(2014) - et al.
A bootstrap-vip approach for selecting wavelength intervals in spectral imaging applications
Chemometrics and Intelligent Laboratory Systems
(2010)
A comparison of technical replicate (cuts) effect on lamb Warner-Bratzler shear force measurement precision
Meat Science
Biochemistry of postmortem muscle — Lessons on mechanisms of meat tenderization
Meat Science
Consumer assessment of beef palatability from four beef muscles from USDA choice and select graded carcasses
Meat Science
Efficacy of performing Warner-Bratzler and slice shear force on the same beef steak following rapid cooking
Meat Science
Relationships between sensory and objective measures of meat tenderness of beef m. longissimus thoracis from bulls and steers
Meat Science
Application of near infrared reflectance spectroscopy to predict meat and meat products quality: A review
Meat Science
Demographics and beef preferences affect consumer motivation for purchasing fresh beef steaks and roasts
Meat Science
Prediction of beef quality attributes from early post mortem near infrared reflectance spectra
Food Chemistry
Identification of the early postmortem metabolic state of porcine M. semimembranosus using Raman spectroscopy
Vibrational Spectroscopy
Preliminary investigation on the relationship of Raman spectra of sheep meat with shear force and cooking loss
Meat Science
Cited by (39)
Raman spectroscopy for rapid fingerprint analysis of meat quality and security: Principles, progress and prospects
2022, Food Research InternationalCitation Excerpt :Hence, a series of sample pre-processing is carried out to solve this problem. To date, the meat detection forms of Raman spectroscopy mainly include complete muscles (Bauer et al., 2016), meat batters (Shao, Zou, Xu, Wu, & Zhou, 2011), and individual components separated from meat (e.g. proteins and lipids) (Lee et al., 2018; Wang et al., 2020). In view of different forms and purposes, various Raman techniques have been applied for meat detection.
Advances in the understanding and measurement of meat texture
2022, New Aspects of Meat Quality: From Genes to Ethics, Second EditionRaman spectroscopic techniques for nondestructive analysis of agri-foods: A state-of-the-art review
2021, Trends in Food Science and TechnologyCitation Excerpt :The spectra were obtained with a laser power of 80 mW and an integration time of 2.5 s. By employing ACO variable selection and multiplicative scatter correction (MSC) pre-processing, R2 values between 0.85 and 0.90 of the PLSR model were obtained, indicating the significant potential of the proposed method for prediction of pH in porcine meat. The combination of RS with chemometric analysis was an indirect method for pH calibration, and it could be extended to predict other quality attributes such as juiciness, tenderness, chewiness, shear force, fattiness, aroma of various meats as well (Fowler, Schmidt, van de Ven, & Hopkins, 2018; Bauer, Scheier, Eberle, & Schmidt, 2016; Zhao et al., 2018; Andersen, Afseth, Gjerlaug-Enger, & Wold, 2021). Unfortunately, due to the high background noise and inherent fluorescence effect of biological tissues, most of the established prediction models only showed weak corrections (R2 < 0.8 or even <0.5), especially for the Raman spectra collected by portable Raman system.
Rapid discrimination of intact beef, venison and lamb meat using Raman spectroscopy
2021, Food ChemistryCitation Excerpt :Over the last decade, vibrational spectroscopy has garnered significant attention as an applicable alternative for meat assessment. Vibrational spectroscopy (Infrared and Raman spectroscopy) posit great advantages over traditional methods for meat discrimination and meat quality assessments (Andersen, Veiseth-Kent, & Wold, 2017; Argyri et al., 2013; Bauer, Scheier, Eberle, & Schmidt, 2016; Berhe, Lawaetz, Engelsen, Hviid, & Lametsch, 2015; Fowler, Schmidt, van de Ven, Wynn, & Hopkins, 2014; Tomasevic, Nedeljković, Stanišić, & Pudja, 2016). These advantages include: minimal sample preparation; rapid measurement/analysis time; non-destructive; non-invasive; cost-effective; increased sensitivity, as well as potential for inline integration (Wang, Peng, Sun, Zheng, & Wei, 2018).