Molecular effects of high-pressure processing on food studied by resonance Raman

Authors


Address for correspondence: Filip Tintchev, German Institute of Food Technology, Prof.-von-Klitzing-Str. 7, D-49610 Quakenbrück, Germany. f.tintchev@dil-ev.de

Abstract

Pressurization may cause unwanted side effects including color or texture changes of fish and meat. The color changes of poultry, pork, and smoked salmon were studied by CIE L*, a*, b* system, and resonance Raman (RR). High-pressure processing (HPP) of pork and chicken meat resulted in significant color modification at pressures higher than 270 and 280 MPa, respectively. RR spectra were taken after a high-pressure treatment of pork meat. According to the RR-data, deoxymyoglobin is the dominating myoglobin species in pork meat. High-pressure treatment causes conformational changes resulting in a stabile nonnative ferrous myoglobin species while the ferrous myoglobin state is maintained. High-pressure treatment causes a decrease of the relative RR intensities of astaxanthin by salmon as probed with 514 nm. RR spectra excited at 413 nm revealed a heterogeneous broadening of astaxanthin bands accompanied by the formation of deoxymyoglobin or deoxyhemoglobin. The broadening is interpreted as the degradation products of astaxanthin. Obviously, the high-pressure treatment of smoked salmon triggers redox processes of astaxanthin and the heme protein.

Introduction

The last decade, investigators have explored the possibilities of applying some novel technologies in the fight against pathogenic microorganisms in foodstuff. High-pressure processing (HPP) is one of these promising technologies. It is a nonthermal preservation technique that—depending on pressure, time, temperature, and product characteristics—allows microorganisms to be inactivated with fewer changes in flavors and vitamins as compared to conventional technologies.1–3 Many industrial high-pressure applications use a pressure level in a range of 400–600 MPa to achieve microbial inactivation at moderate temperatures. Such pressure can affect the color of raw meat, leading to discoloration of red muscle from pork and beef also at low temperatures of 5–10° C.4 This pressure-related discoloration of meat is caused by denaturation of myoglobin.5,6 In contrast7 observed no myoglobin denaturation before 500 MPa at 10° C, or before 570 MPa at 20° C.

Structural changes of Mb in solution induced by high pressure have been already studied using infrared and UV-Vs absorption spectroscopy.8 Alterations of the heme structure including a change in the spin and coordination state have been monitored by resonance Raman (RR) spectroscopy.9 Nevertheless, possible pressure-induced structural changes of Mb within the intact meat tissue have not yet been analyzed.

Color is also a very important quality parameter for salmonid products.10 Color of salmon is caused by carotenoids, mainly astaxanthin and to a smaller extent by other carotenoids like canthaxanthin and β-carotene.11 Negative effects on the color caused by HPP of smoked salmon could therefore limit the applicability of this new technology.

This work aims to elucidate the impact of pressure and its holding time on the color of poultry and pork meat, such as smoked salmon. Another aspect is to study the molecular mechanisms of the color changes and the reasons of the pressure stability or sensibility of meat and fish color.

RR spectroscopy provides information about the vibrations of molecules by excitation of an electronic transition. RR spectroscopy has found wide application to the analysis of natural pigments.12 Withnall et al.13 have used Raman spectra to detect carotenoids in natural products. An important advantage of RR is that pigments can be analyzed in complex matrices by collecting their Raman spectra under resonant condition, thereby selectively enhancing some of their Raman bands and not those of the surrounding matrix. Moreover, detection of Raman spectra is a quick analytical method.

Materials and methods

Materials

Pork meat from the muscle Longissmus dorsi was obtained from a local supermarket and used for RR spectroscopic experiments without and after a high-pressure treatment. To extract myoglobin (Mb) from untreated and pressurized samples, the meat was mixed with equal amounts of ice (demineralized water), chopped, and then homogenized with an Ultra-Turrax T 18 (IKA-Werke GmbH & Co. KG, Staufen, Germany). The meat mash was centrifuged (Cyrofuge 8000, Heraeus) at 35,000 g and 3° C for 15 min and the supernatant was filtered with a (MN 615) filter according to Bünnig and Hamm (1969). This supernatant solution, denoted as “meat extract,” was used without further purification for UV-vis absorption and RR spectroscopic measurements. As a reference, isolated horse heart Mb (Sigma-Aldrich GmbH, Steinheim, Germany) was used in an aqueous phosphate buffer at pH 7. Addition of sodium dithionite (Merck Chemicals, Darmstadt, Germany) to the solution led to the conversion of met-myoglobin (metMb) to deoxy-myoglobin (deoxyMb). Excess dithionate was removed by an Econo-Pac 10DG desalting column (Bio-Rad Laboratories GmbH, München, Germany). In contact with air, deoxyMb bound molecular oxygen to form oxy-myoglobin (oxyMb).

Cold smoked salmon, chopped and vacuum-packed of the brand “Friedrichs Wildlachs” was purchased from a supermarket for detection of Raman spectra in package. The packing foil consists of polyethylene (65 μm) and biaxial-oriented polyamide (15 μm). As standard solution Astaxanthin (Concentration e 92%, Sigma-Aldrich) was dissolved in chloroform with a concentration of approximately 1 mM.

High-pressure treatment

Mb and astaxanthin solutions, pork and fish meat extracts, and cylindrically sliced pork and fish meat pieces (d = 32 mm, l = 28 mm) were vacuum-packed and pressurized at 0.1–700 MPa at ambient temperature for up to 10 minutes. Due to the work of compression, the temperature increased by ca. 3° C per 100 MPa, resulting in a temperature of approximately 21° C at 700 MPa. A pressure buildup time of 15 sec was needed to reach 600–700 MPa. The high-pressure treatment was carried out in a custom-made lab-scale high-pressure system (High Pressure Research Center, Unipress Equipment Division, Sokolowska 29/37, Warsaw, Poland). Maximum design pressure for the system was 1000 MPa at an operating temperature range of −25 to 100° C. The volume of the sample holder was 0.75 L. A 1:1 mixture of water and propylene glycol (1,2 propanediol) was used as a pressure-transmitting medium. Additional experiments were carried out in a high-pressure processing pilot plant (NC-Hyperbaric, Burgos, Spain) with a volume of 55 L, allowing pressurization of up to 350 kg/h. A pressure of 600 MPa was built up within 3–4 min.

Spectroscopic measurements

The absorption spectra were recorded on a Unicam UV2 spectrophotometer with a spectral resolution of 0.5 nm and a Lambda 25 (PerkinElmer LAS GmbH, Rodgau-Jügesheim, Germany) spectrophotometer with a spectral resolution of 1 nm (pork meat) and 4 nm (smoked salmon). The path length of the detection cell was 10 mm and the scan speed 600 nm/min for all experiments.

RR spectra were measured with a 413-nm excitation line of a Kr+-laser (Coherent Innova 400; Coherent Innova, Santa Clara, CA USA) and a 514-nm line of a Ar+-laser (Coherent Innova 400) using a confocal Raman spectrograph (LabRam Jobin-Yvon/Horiba, Edison, NJ USA) equipped with an electronically cooled CCD camera. The spectral resolution was approximately 3 cm−1. The incident laser beam (3 mW) was focused on the sample that was placed in a rotating cell for measurements of proteins in solution and meat extracts. For measurements of meat pieces, a stationary sample holder was employed. Under these conditions, laser-induced damage of the samples, such as photodissociation of oxygen from oxyMb, can be ruled out since comparative measurements of meat contained in a rotating device revealed essentially the same results albeit with a lower signal-to-noise ratio.

Color measurements

Color measurements of meat have been performed using a colorimeter CR 110 from Minolta, employing in the specular component mode that includes the diffuse and specular reflection. In this way, the color can be evaluated independent of the surface structure. Illumination was performed by D65 (standard illuminant defined by the International Commission on Illumination), which is intended to represent average daylight and has a correlated color temperature of approximately 6500 K. The results are expressed within the L*a*b* color space that is based on the uniform distribution of colors and is very close to human perception of color. L* is the luminance or lightness component, ranging from 0 to 100, and the parameters a* (from green to red) and b* (from blue to yellow) are the two chromatic components, which range from −120 to 120.

Results and discussion

L*a*b* analysis

The effect of HPP on the appearance of the pork and poultry meat processed for 60 sec is shown in (Fig. 1A). A clear color gradient is visible from control to 500–600 MPa samples, respectively. Minor effects are already visible at 300 MPa for turkey and pork and 200 MPa for chicken meat. For pork meat these are presented with the help of L*a*b* color space system for L* (lightness), a* (redness), and b* (yellowness) values in (Fig. 1B). An increase in lightness (L*) for all three meat types and no significant effect on the redness (a* value) of pork meat is detected. It is well known that application of high pressure even at 5–10° C induces drastic changes in the color of red muscle from pork, beef, tuna, etc.14–16 Meat discoloration through pressure processing appears to result from (1) a “whitening” effect in the range 200–350 MPa, due to globin denaturation and/or to heme displacement or release, and (2) oxidation of ferrous myoglobin to ferric metmyoglobin, at or above 400 MPa.17 Both degradation phenomena appear to depend more on critical pressure thresholds than on time, since they took place already 2–5 min after the target pressure was reached.15 These could be seen also in the (Figs. 1A and 2B) where drastic discoloration of pork meat was detected after 1-sec treatment time at 400 MPa (up and down). A simultaneous increase of L*and A* values with no significant development during the holding time along with a decreasing b* value was seen. The color changes after HPP of smoked salmon is shown in (Fig. 3). Salmon color is more pressure stable than that of pork and poultry. The reason is the different color pigment, which in this case is astaxanthin. This pressure stability can be explained with a different molecular structure of the pigment astaxanthin for salmon and myoglobin for meat. But also a small whitening effect is detected by the salmon with increasing pressure, L* value increased and a* and b* values decreased (Figs. 3A and 4A). In earlier studies18 a significant reduction of redness was reported. However, in this study fresh salmon and processing times of 10, 30, and 60 min in comparison to 60 sec in our study have been used. A threshold value of acceptability for the lightness of smoked salmon has not been reported, but for fresh salmon a threshold value of 70 is considered as unacceptable.18

Figure 1.

Figure 1.

(A) Changes of native pork and poultry meat after high-pressure processing (HPP) at 0–600 MPa after 1-min treatment and 10° C start temperature (B) Color changes of pork meat after HPP at 400 MPa and 10° C and different treatment times.

Figure 1.

Figure 1.

(A) Changes of native pork and poultry meat after high-pressure processing (HPP) at 0–600 MPa after 1-min treatment and 10° C start temperature (B) Color changes of pork meat after HPP at 400 MPa and 10° C and different treatment times.

Figure 2.

Figure 2.

(A) Changes of the CIE color parameters L*, a*, b* of pork meat at 10° C and 0.1–700 MPa for 1 minute. (B) Changes of the CIE color parameters L*, a*, b* of pork meat at 10° C, 400 MPa, and different treatment times.

Figure 2.

Figure 2.

(A) Changes of the CIE color parameters L*, a*, b* of pork meat at 10° C and 0.1–700 MPa for 1 minute. (B) Changes of the CIE color parameters L*, a*, b* of pork meat at 10° C, 400 MPa, and different treatment times.

Figure 3.

Figure 3.

(A) Color changes in smoked salmon after HPP at 0–600 MPa, 1 min holding time and 20° C start temperature. (B) Color changes in smoked salmon after HPP at 600 MPa and different start temperatures, 1 min holding time.

Figure 3.

Figure 3.

(A) Color changes in smoked salmon after HPP at 0–600 MPa, 1 min holding time and 20° C start temperature. (B) Color changes in smoked salmon after HPP at 600 MPa and different start temperatures, 1 min holding time.

Figure 4.

(A) Changes of the CIE color parameter L*, a*, b* of smoked salmon by HPP 0–600 MPa at 20° C start temperature for 1 min holding time. (B) Changes of the CIE color parameters L*, a*, b* of smoked salmon at 600 MPa and different start temperatures and 1 min holding time.

Significant color changes were found after treatment at 500 MPa and 20° C starting temperature. To minimize the unwanted color changes lower starting temperatures were tested. As an optimal starting temperature 0° C has been identified, where similar results than for the control sample have been obtained (Figs. 3B and 4B). HPP at −10 and −5° C start temperature showed increased L* value and decreased b* value but no significant a* value changes. According to the results temperature can affect improve color retention and minimize undesired discoloration of the salmon, in opposite to (pork, turkey, and chicken) meat, where the temperature had no significant impact on the color. For better understanding of the discoloration mechanism on a molecular level RR spectroscopy was used.

Resonance Raman spectroscopy

The color stability of meat is determined by the maintenance of myoglobin in its native ferrous (2+) form. Oxidation to the ferric met-Mb is undesirable and takes place progressively during refrigerated storage. Factors increasing the oxidation to met-Mb are the partial oxygen pressure, rate of oxygen consumption by the tissues, presence of multivalent metal ions, temperature, light, pH, and the meat microflora. Oxygen consumption, met-Mb reducing enzymes, and the NADH pool can convert met-Mb to deoxy-Mb in meat.19 However, both enzyme activity and NADH pool continually deplete as time postmortem proceeds.

Heme proteins have been extensively studied by RR spectroscopy.12,20 By excitation of an electronic transition the vibrations of the heme can be selectively probed and provide information about the configuration, while the protein and food matrix is optically silent. The maxima of the Soret bands for the myoglobin species are between 409 and 434 nm, excitation at 413 nm causes therefore a strong enhancement of the heme modes. RR spectroscopy thus can inter alia differentiate between the myoglobin species: deoxymyoglobin (Deoxy-Mb), oxymyoglobin (Oxy-Mb), and metmyoglobin (Met-Mb) by the frequencies of the so-called marker bands in the region between 1300 and 1700 cm−1 (Fig. 5A). The coordination and oxidation state of the heme iron can be determined by the frequency of the mode ν4.12 In reduced deoxy-Mb the band is observed at 1356 cm−1 whereas in the oxidized oxygen-free form, met-myoglobin (met-Mb), this mode shifts up to 1370 cm−1. Both in deoxy-Mb and met-Mb, the heme iron is in a high-spin configuration with the imidazole residue of histidine 93 as the proximal ligand. The distal position remains vacant in the deoxy-Mb form whereas a water molecule occupies the sixth coordination site in met-Mb. These configurations are reflected by the spin- and coordination marker bands above 1450 cm−1, particularly of the mode ν3, which are found at 1473–1481cm−1 in deoxy-Mb and met-Mb, respectively. Oxygen binding to the ferrous state induces a low-spin configuration, in which the electron density is partially transferred to the empty antibonding orbitals of the ligand such that the ν4 frequency shifts up to 1377 cm−1. The other marker bands, ν3 and ν2, are found at higher frequencies such that, altogether, the analysis of this spectral region allows for an unambiguous identification of the Mb state.

Figure 5.

Figure 5.

(A) Resonance Raman (RR) spectra of deoxi-, oxi-, and metmyoglobin (B) RR spectra (a) before pressure treatment, and after pressurization (b) 600 Mpa and (c) 700 MPa, 10 minutes.

Figure 5.

Figure 5.

(A) Resonance Raman (RR) spectra of deoxi-, oxi-, and metmyoglobin (B) RR spectra (a) before pressure treatment, and after pressurization (b) 600 Mpa and (c) 700 MPa, 10 minutes.

After high-pressure treatment, distinct spectral changes are observed in the RR spectrum (Fig. 5B). The overall RR intensity decreases, accompanied by changes in the frequencies and intensities of the marker bands. The ν4 envelope shows a slight upshift of the maximum peak from 1356 cm−1 to 1359 cm−1 and 1363 cm−1 in spectra of meat previously pressurized at 600–700 MPa, respectively. Moreover, a closer inspection of the region between 1420 and 1620 cm−1 reveals an intensity decrease of the bands at 1473 (ν3) and 1564/cm (ν2) that are replaced by bands at 1493 and 1581 cm−1. These frequencies of the marker bands (ν4, ν3, and ν2) are typical for the modes of a ferrous heme in a six-coordinated low-spin configuration (6cLS). The nature of the second axial ligand is not clear a priori. A possible candidate is histidine 64, which is located in the distal part of the heme pocket. Indeed, the RR spectral parameters of the nonnative 6cLS ferrous Mb form resemble those of the bis-histidine-coordinated 6cLS ferrous cytochrome.21 Such a pressure-induced bis-coordinated heme species of Mb has in fact been previously discussed by Zipp & Kauzmann and Ognumola et al.6,22

The color of carotenoids is caused by an allowed π–π* transition of the conjugated carbon double bond system in the visible region. The maximum of the absorbance of the xanthophyll astaxanthin in chloroform is located at 491 nm. The laser line at 514 nm is thus close to the absorbance maximum. Excitation at 514 nm and at 413 nm was used in the RR experiments to study the interference of the chromophore with the background noise caused by the matrix of the salmon flesh (e.g., proteins and fat).

In Figure 6 the RR spectra of astaxanthin in chloroform (i) and smoked salmon flesh (ii) without pressure treatment, (iii) after high-pressure treatment at 600 MPa, and (iv) after high-pressure treatment at 700 MPa is shown. It is obvious that in these spectra the bands at 1155 cm−1 and 1518 cm−1 are strongly enhanced at an excitation at 514 nm caused by π–π* (S0–S2) electronic transition of carotenoids. These bands are related to in phase carbon–carbon stretching vibrations of the main polyene chain, ν3 having single bond and ν1 having double bond character. The enhancement is so strong that even the solvent vibration mode at 1215 cm−1 of chloroform is hardly visible in the spectra of astaxanthin in solution.

Figure 6.

RR spectra of (a) astaxanthin in chloroform excitation wavelength 514 nm; (b) smoked salmon, excitation wavelength 514 nm; (c) smoked salmon, pressurized at 600 MPa, 3 min; and (d) smoked salmon, pressurized 700 MPa, 1 minute.

Figure 7 shows the RR spectra of smoked salmon before and after a pressurization at 700 MPa recorded at 413-nm excitation wavelength. The position of ν3 kept nearly constant while ν1 is shifted by 2 cm−1 to 1520 cm−1 and has a broad shoulder toward higher wave numbers. At ambient pressure there is a small band around 1370 cm−1. By increasing pressure a band at 1357 cm−1 appeared. These bands are attributed to the ν4 mode of a heme protein like Mb or hemoglobin in different oxidation states, as this mode shows a strong enhancement upon excitation at 413 nm (vide supra). These findings are surprising since they suggest a redox process in pressure-treated salmon, which has not been observed for pressure-treated meat. It is thus reasonable to assume that the redox reactions involve astaxanthin. In fact, it is known that carotenoids bleach, that is, lose their color, when exposed to radicals or to oxidizing species. This process involves degradation of the conjugated double-bond system that is supported by the increased RR intensity on the high-frequency side of the 1520 cm−1 band, as this band is moved to higher frequencies for shorter polyene systems. The shorter polyene system should also move the absorption maximum, providing better resonance conditions for the oxidation products of astaxanthin upon 413-nm excitation as compared to 514-nm excitation. The expected smaller effect of astaxanthin oxidation on the C–C stretching is reflected by the less pronounced heterogeneous broadening of the 1155-cm−1 band. However, the mechanism for the oxidation degradation of astaxanthin and how the oxidation is coupled to the reduction of metmyoglobin/methemoglobin remains enigmatic.

Figure 7.

RR spectrum (excitation wavelength 413 nm) of (a) smoked salmon at ambient pressure; and (b) after pressurization at 700 MPa, 1 minute.

Conclusion

High-pressure treatment of pork, turkey, and chicken meat caused an unwanted discoloration, as shown on the L*,a*,b* values. This discoloration started at 200–300 MPa and caused a cooked appearance of the meat after 500–600 MPa for 1-min holding time. On the other hand, smoked salmon showed a higher pressure stability, significant discoloration started at a pressure level of 500 MPa. A minimum of undesired changes was found at lower treatment temperatures, at −10 and −5° C the lightness was lower than, at 0° C it was comparable to the untreated sample.

RR spectroscopy can provide molecular information about the changes of chromophores of (pressurized) food, which are responsible for the appearance as well as discoloration. It was found that astaxanthin oxidized during the high-pressure treatment of smoked salmon and its degradation is coupled to the reduction of metmyoglobin via a yet unknown mechanism. High-pressure treatment induces the formation of a nonnative ferrous Mb species. The nonnative Mb species contribute to the color appearance of the pork meat. This species should not be able to catalyze oxidation of the meat matrix (as iron is in the ferrous state).

Conflicts of interest

The authors declare no conflicts of interest.

Ancillary