The optical index G

The optical index is built up by the fluorescence intensity of the 470 and 430 nm emission wavelengths at 330 to 360 nm excitation wavelengths. After the process of normalization, we have found that both G(470,exci) and G(430,exci) have a descending trend with storage time. The range of G index is approximately from 1.5 to 0.5. However, it is different from fish to fish. The optical index value has a large standard deviation in both cobias and Seriola dumerili after several hours of storage time. In some fish cases, we do not get the clear descending result. We consider that there are many factors affecting the experiment result, e.g., the growing environment, temperature, slaughter method, and experimental error [27, 28, 31]. In summary, for the eight Seriola dumerili, on the whole, the value of G(470, exci) and G (430, exci) are negatively correlated to the K value within 24 hours. The correlation can be established only when it can be ensured that the fluorophores do not initially degrade. If the fluorophores initially degrade, the descending trend of the G index with storage time is less obvious, which causes a less negative correlation with K value.

40

However, the lower value of G means the rigor mortis and tenderization have occurred earlier than normal, and therefore the fish is less fresh. Thus, with the value of the G index, we can confirm the condition of the fish muscle, the postmortem process period that the muscle is in, and even the freshness within 24 hours after the fish has died.

4.3 The slaughter methods for spectra

It has been proven that as a fish loses energy, the fish quality declines and results in a higher rate of spoilage [32]. It has been proven that when fish are frightened or experience tension, NADH and collagen could degrade faster [27]. My results also confirm that fish freshness is affected by these conditions. Fig. 16 shows that when fish lose energy due to treatment with ice water, the fluorophores of NADH and collagen initially degrade faster, and the G index shows a less descending trend.

There are many factors that could affect the optical index, including temperature, harvest methods, season, aquaculture condition, and so on. Some of the factors relate to the natural condition of the fish; some depend on how the fish is treated after harvesting. The value can confirm the postmortem condition of the fish tissue and can be used to validate the quality of the fish.

41

Fig. 16 summarizes the fluorescence indexes of two Seriola dumerili that experienced tension due to low temperatures induced by ice measured with 330, 340, 350, and 360 nm wavelength light sources: (a) G(470,exci) in abdomen, (b) G(470,exci) in dorsum, (c) G(430,exci) in abdomen, and (d) G(430exci) in dorsum samples.

4.4 The changes of other fluorophores in fish tissue within 24 hours of refrigeration

In addition to the direct observation from EEM, the fluorophores are also confirmed by principle component analysis of eight Seriola dumerili. Moreover, this study tries to find other fluorophores that are related to freshness by using PCA to extract the main fluorophores.

Other studies have suggested that aromatic amino acid (tryptophan, tyrosine), ATP, NADH, and collagen are related to meat freshness [22]. However, for fresh fish, they needed to be investigated, which was appropriate for this study. The fluorophores of amino acid, ATP, collagen, and NADH were found by PC1 and PC2 with PCA with all the eight Seriola

42

dumerili. The three groups (330 nm, 280 nm, and 290 nm excitation wavelengths) of PC1 and PC2 account for more than 95 percent of all the components. Fig. 17 (a) and (b) show the main components of 470 and 430 nm emission wavelengths could be the fluorophores of NADH and collagen by 340 nm excitation wavelength, with all the fluorescence signals of Seriola dumerili loading in. Fig. 17 (c) and (d) depict the 390 nm emission wavelength, which could be the component of amino acid at 280 nm excitation wavelength, with all the fluorescence signals of Seriola dumerili loading in. In Fig. 17 (e) and (f), the 350 to 400 nm emission wavelengths could be ATP on the 290 nm excitation wavelength, with all the fluorescence signals of Seriola dumerili loading in.

Fig. 17 depicts eight Seriola dumerili analyzed by PCA: (a) and (b) are PC1 and PC2 excited by 280 nm; (c) and (d) are PC1 and PC2 excited by 340 nm; and (e) and (f) are PC1 and PC2 excited by 290 nm.

However, it is most important to confirm which components could change to the largest degree within a short time so that the spectrum can act as a freshness or quality index. From Fig. 18, the results of PC1 show that NADH and collagen sharply decrease within 24 hours.

However, at the 280 and 290 nm excitation wavelengths, it is hard to distinguish any

43

difference for amino acid and ATP within 24 hours. It could be that the levels of amino acid do not change a lot within a short time after the fish has died, or it could be that there are too many fluorophores in this area to balance the difference. In sum, we conclude that the fluorophores of NADH and collagen can be great biomarkers for freshness.

Fig. 18 shows the PC1 extracted by fluorescence from 0 to 24 hours storage time at (a) 340, (b) 280, and (c) 290 nm excitation wavelengths.

4.5 The white and red fish meats

Cobias and Seriola dumerili are white fish species, and glycolysis of the fish muscle is more severe for them than for red fish species. After white fish die, the muscles maintain the ATP concentration longer than red fish. The ATP concentration in white fish slows down the process of rigor mortis, tenderization, and even spoilage. Because of this, red fish become stale at a faster pace. In this study, cobias and Seriola dumerili were measured to analyze the spectra from tissue. However, we can speculate that the change of G index might be sharper

44

for red fishes because of their faster metabolism.

4.6 The problems in this study

In this study, actually, a total of 20 fish were tested. However, because of the unstable excitation power for the xenon lamp and because there were other spectra areas that we tried to test, the study was limited to eight fish subjects. In future work, we might use a laser as a power source for its stability and its concentration of energy. In addition, the delay for several minutes between slaughtering the fish and performing the initial measurements could have resulted in missed information related to the fish tissue.

Also, in this study, a Y type fiber was adopted to receive the fluorescence. Although fixing the fiber directly to the tissue surface might have prevented errors related to hand movements, the fluorescence received was only from a small point of fish tissue. This was done to reduce the non-homogeneous effect of fish tissue, but separating and measuring more parts of fish tissue could be beneficial in future research.

4.7 To develop a portable detector for fish freshness

4.7.1. The potential of nitrogen laser as excitation power

The 337 nm wavelength of a nitrogen laser would be a proper power source for a fish freshness detector. Nitrogen lasers have been proven to cause little destruction to tissue [10].

With the less invasive character of a nitrogen laser, it could be commercialized for the development of a portable instrument.

45

4.7.2. The usage and problem of G index for freshness

From my study, the sliced fishes were measured to analyze the spectrum. However, for now, it only can be applied as the freshness detector of sashimi or other fish fillets as a quick-detection method. The challenge is to investigate the change of spectra of fish muscles beyond spot measurements. By confirming the spectra change of muscle in the whole fish and developing a standard, the method can be applied as a faster and less invasive detection method.

46

V. Conclusion 5.1 The optical index

The fluorescence spectrum can present the concentration of type I collagen, type V collagen, and NADH. NADH is the energy source of the live body. When fish die, the fish muscles start glycolysis to maintain the energy of the fish muscles. After the fish muscle is acidified by the accumulation of many lactic acids from glycolysis, it stops generating NADH or ATP. At this point, the NADH may degrade. Also, it has been proven that collagen is related to the resolution of muscle. Collagen is a component of fish muscle structure and has been proven to participate in the tenderization of muscle. When collagen is transformed, the texture of fish may soften, which indicates the breakdown of muscle structure. NADH and collagen are the materials that maintain the normal metabolism and structure of muscle, and they may degrade soon after a fish dies. Thus, they play important parts in the early process of fish putrefaction. From this study, two species of fish and a total of 16 fish samples were analyzed. The G index value that we developed might descend from approximately 1.5 to 0.5 with 24 hours of refrigeration with an ascending K value. Therefore, based on the G index we developed from the fish, the condition of the fish can be confirmed. If the G index of the fish is higher, the fish could be assumed as a good quality or fresh fish which has longer expiry date. On the contrary, if the G index of the fish is lower, the fish can be assumed as a bad quality or “less fresh” fish which has shorter expiry date. In other words, we can use the G index to validate freshness in the early stages of fish putrefaction, and it can act as a measure of taste and freshness of sashimi or fish fillets.

5.2 The application of the index

47

For the light source for the optical index, the nitrogen laser has been commercialized for many years and is appropriate for our optical system. Also, because of the laser’s less destructive wavelength to tissue [10], the index has great potential for freshness detection [33].

The index can be used to monitor the quality of sashimi in restaurants or for quality control in the aquaculture market. Restaurants and the aquaculture market have strong operation performance standards and are capable of applying these indexes. With the G indexes, they can detect the freshness of fish faster.

48

VI. Future Work

Up to now, we have only used 1D signals to analyze the EEM signal. However, 2D matricess of EEM may contain more information that we have not found. A 2D algorithm has faster computation and is more appropriate for 2D signal processing. It is a method to analyze the EEM spectrum for other fluorophores. Also, with a 2D algorithm, maybe more target fluorophores can be discovered. Second, the amount of fish data is not enough to prove how the index works with other fish species. Accumulating more data of other fish species is necessary. Finally, if the optical index is to be applied in the food or restaurant industries, we need to find more useful indexes and combine them to build a valuable portable instrument.

49

VII. Reference

[1] K. Sveinsdottir, E. Martinsdottir, G. Hyldig, B. Jørgenson, and K. Kristbergsson,

“Application of quality index method (QIM) scheme in shelf-life study of farmed atlantic salmon (salmo salar)”, Sensory and Nutritive Qualities of Food, vol. 67, no. 4, pp.

1570-1579, May 2002.

[2] O.O. Cyprian, K. Sveinsdóttir, H. Magnússon and E. Martinsdóttir, “Application of quality index method (QIM) scheme and effects of short-time temperature abuse in shelf life study of fresh water arctic char (salvelinus alpinus)”, Journal of Aquatic Food Product Technology, vol. 17, no. 3, pp. 303-321, Oct. 2008.

[3] N. Hamada-Sato, K. Usui, T. Kobayashi, C. Imada, and E. Watanabe, “Quality assurance of raw fish based on HACCP concept”, Food Control, vol. 16, no.4, pp.

301-307, Apr. 2005.

[4] CNS1451，N6029，冷凍魚類檢驗法，經濟部標準檢驗局，台北，1991。

[5] T. Hattula and M. Kiesvaraa, “Breakdown products of adenosine triphosphate in heated fishery products as an indicator of raw materials freshness and of storage quality”, Food Science and Technology, vol. 29, no.1-2, pp. 135-139, 1996.

[6] J.M. Shewan, R.G. Macintosh, C.G. Tucker, and A.S.C. Ehrenberg, “The development of a numeric scoring system for the sensory assessment of the spoilage of wet white fish stored in ice”, Journal of the Science of Food and Agriculture, vol. 4, pp. 283-298, 1953.

[7] J.M. Ryder, “Determination of adenosine triphosphate and its breakdown products in fish muscle by high-performance liquid chromatography”, Journal of Agricultural and Food Chemistry, vol. 33, no. 4, pp. 678-680, 1985.

[8] T. Hatula, “Adenosine triphosphate breakdown products as a freshness indicator of some fish species and fish products”, Academic Dissertation, Department of Biochemistry in Helsinki University, 1997.

[9] N. Kato, M. Kunimoto, S. Koseki, S. Kitakami and K.I. Arai, “Freshness and quality of fish and shellfish”, Journal of the School of Marine Science and Technology, Tokai University, vol. 7, no. 2, pp. 87-99, 2009.

[10] A.A. Stratonnikov, V.S. Polikarpov, and V.B. Loschenov, “Photobleaching of endogenous fluorochroms in tissues in vivo during laser irradiation", Proceedings of SPIE, vol. 4241, 2001.

[11] L. Gil, J.M. Barat, I. Escriche, E. Garcia-Breijo, R. Martínez-Máñez, and J. Soto, “An electronic tongue for fish freshness analysis using a thick-film array of electrodes”, Microchim Acta, vol. 163, no. 1-2, pp. 121-129, 2008.

[12] N. Gokoglu and P. Yerlikaya, “Use of eye fluid refractive index in sardine (sardina portable electronic nose system to assess the freshness of Moroccan sardines”, Materials Science and Engineering C, vol. 28, no. 5-6, pp. 666-670, Jul. 2008.

[15] Y. Nanjyo and T. Yao, “Rapid measurement of fish freshness indices by an amperometric flow-injection system with a 16-way switching valve and immobilized enzyme reactors”, Analytica Chimica Acta, vol. 470, no. 2, pp. 175-183, 2002.

[16] J.B. Luten, J. Oehlenschläger, and G. Olafsdottir (Eds.), Quality of fish from catch to consumer: labeling, monitoring and traceability, Wageningen Academic Publishers, The

50

Netherlands, ISBN: 978-90-76998-14-5, pp. 211-224, 2003.

[17] H. Nilsen, M. Esaiassen, and F. Sigernes. “Visible/near-infrared spectroscopy: a new tool for the evaluation of fish freshness?”, Journal of Food Science, vol. 67, no. 5, pp.

1821-1826, Jun. 2002.

[18] S.C. Chu, T.C. Hsiao, C.Y. Wang, J.K. Lin, and H.H. Chiang, “Comparison of the performances of linear multivariate analysis method for normal and dyplasia tissues differentiation using autofluorescence spectroscopic”, IEEE Transactions on Biomedical Engineering, vol. 53, no. 11, pp. 2265-2273, Nov. 2006.

[19] Available from:

http://www.physik.unibas.ch/Praktikum/VPII/Fluoreszenz/Fluorescence_and_

Phosphorescence.pdf.

[20] S.P. Aubourg and I. Medina, “Influence of storage time and temperature on lipid deterioration during cod (gadus morhua) and haddock (melanogrammus aeglefinus) frozen storage”, Journal of the Science of Food and Agriculture, vol. 79, no. 13, pp.

1943-1948, Sep. 1999.

[21] G. Duflos, F.B. Le, V. Mulak, P. Becel, and P. Malle, “Comparison of methods of differentiating between fresh and frozen-thawed fish or fillets”, Journal of the Science of Food and Agriculture, vol. 82, no. 12, pp. 1341-1345, Sep. 2002.

[22] E. Dufour, J.P. Frencia, and E. Kane, “Development of a rapid method based on front-face fluorescence spectroscopy for the monitoring of fish freshness”, Food Research International, vol. 36, pp. 415-423, 2003.

[23] E. Dufour, A. Letort, A. Laguet, A. Lebecque, and J.N. Serra, “Investigation of variety, typicality and vintage of French and German wines using front-face fluorescence spectroscopy”, Analytica Chimica Acta, vol. 563, no. 1-2, pp. 292-299, Mar. 2006.

[24] R. Karoui, E. Thomas, and E. Dufour, “Utilization of a rapid technique based on front-face fluorescence spectroscopy for differentiating between fresh and frozen–thawed fish fillets”, Food Research International, vol. 39, no.3, pp 349-355, Apr.

2006.

[25] C.M. Anderson and J.P. Wold, “Fluorescence of muscle and connective tissue from cod and salmon”, Journal of Agricultural and Food Chemistry, vol. 51, no. 2, pp. 470-476, 2003.

[26] 蔡佳玲，「收穫後處理與包裝對海鱺、吳郭魚與鱸魚品質與 5℃儲藏期限之影響」，

碩士論文，國立臺灣海洋大學，2006。

[27] N.P.L. Tuckey, M E. Forster, and S.P. Gieseg, “Effects of rested harvesting on muscle metabolite concentrations and K-values in chinook salmon (oncorhynchus tshawytscha) fillets during storage at 15 ◦C”, Journal of Food Science, vol. 75, no. 5, pp. 459-464, 2010.

[28] K. Sato, S. Uratsuji, M. Sato, S. Mochizuki, Y. Shigemura, M. Ando, Y. Nakamura, and K. Ohtsuki, “Effect of slaughter method on degradation of intramuscular type V collagen during short-term chilled storage of chub mackerel scomber japonicas”, Journal of Food Biochemistry, vol. 26, no. 5, pp. 415-429, Nov. 2002.

[29] M. Ando, Y. Yoshimoto, K. Inabu, T. Nakagawa, and Y. Makinodan, “Post-mortem change of three-dimensional structure of collagen fibrillar network in fish muscle pericellular connective tissues corresponding to post-mortem tenderization”, Fisheries Science, vol. 61, no.2, pp. 327-330, 1995.

[30] M.D. SUA´ Rez, M. Abad, T. Ruiz-Cara, J. D. Estrada, and M. Garcia-Gallego,

“Changes in muscle collagen content during post mortem storage of farmed sea bream (sparus aurata): influence on textural properties”, Aquaculture International, vol. 13, no.

4, pp. 315–325, 2003.

51

[31] B. Roth, E. Slinde, and J. Arildsen, “Pre or post mortem muscle activity in atlantic salmon (salmo salar). The effect on rigor mortis and the physical properties of flesh”, Aquaculture, vol. 257, pp. 504-510, 2006.

[32] M. Tejada and A. Huidobro, "Quality of farmed gilthead seabream (sparus aurata) during ice storage related to the slaughter method and gutting", European Food Research and Technology, vol. 215, no. 1, pp. 1-7, 2002.

[33] C.W. Wu, T.C. Hsiao, S.C. Chu, H.H. Hu, and J.C. Chen, “Fibre optic fluorescence spectroscopy for monitoring fish freshness”, Proceedings of the SPIE, vol. 8220, 2012.

52

Appendix A

Abbreviation Full name

ADP Adenosine diphosphate

AMP Adenosine monophosphate

ATP Adenosine triphosphate

EEM Excitation emission matrix

HPLC High performance liquid chromatography

Hx Hypoxanthine

HxR Inosine

IMP Inosine monophosphate

NADH Nicotinamide adenine dinucleotide PCA Principle component analysis

PMT Photomultiplier tube

53

Appendix B

Function Description G(emi.,exci)

where F(emi, exci) denotes the fluorescence amplitude of certain excitation wavelength (exci nm) and emission wavelength (emi nm) and F(exci+50, exci) denotes the fluorescence amplitude of certain excitation wavelength (exci nm) and emission wavelength (exci+50 nm). The G index is developed as a optical index for freshness.

where [ATP], [ADP], [AMP], [IMP], [HxR], [Hx] mean the concentration of themselves. In most of fish, K values increase linearly during the first days of refrigeration storage and it is an index of freshness detection. When K value is higher, it means the less fresh of fish meat.

r value value is between -1 and 1 which can compare the relationship between x and y variables.

The value of recovery is used as the efficiency for our experiment process.

Based on the value, it can confirm the quality of our experiment method.

54

where x , S, and n denote as mean, standard deviation, and the number of {x_i}, respectively. RSD is used to confirm the precision of HPLC results.