SENSORY ANALYSIS OF A MODEL SYSTEM USING 5'-IMP AND CYSTEINE AT DIFFERENT pH

ANÁLISE SENSORIAL DE UM SISTEMA MODELO USANDO 5’IMP E CISTEÍNA À DIFERENTES pH

SENSORY ANALYSIS OF A MODEL SYSTEM USING 5'-IMP AND CYSTEINE AT DIFFERENT pH¹

Marta Suely MADRUGA^2,*, Donald S. MOTTRAM³

SUMMARY

Sensory analysis was used to get an overall flavour description of a reaction mixtures containing 5'-IMP and Cysteine. Ribose/cysteine systems were used as reference systems. Results from triangle and aroma profiling show a clear correlation between the terms used and the volatile analysis described in literature for these model systems. For instance reactions at pH 3.0 and 4.5 for 5'-IMP/cysteine systems, which were described as "meaty" and "boiled meat" by panellists, presented, in the literature, the higher number of "meaty" compounds in volatile analysis (1, 7, 8, 20).

Keywords: Sensory analysis, inosine-5’-monophosphate, cysteine, pH, flavour.

RESUMO

ANÁLISE SENSORIAL DE UM SISTEMA MODELO USANDO 5’IMP E CISTEÍNA À DIFERENTES pH. Análise sensorial foi realizada objetivando-se descrições aromáticas de misturas reactivas contendo 5’-IMP e cisteína. Sistemas modelos formados por ribose/cisteína foram utilizados como referência. Os resultados obtidos a partir do teste triangular e aroma descritivo mostraram uma clara correlação entre os termos utilizados e as análises de constituintes voláteis descritas na literaturas envolvendo estes sistemas. Por exemplo, sistemas modelos contendo 5’-IMP/ cisteína a pH 3,0 e 4,5, que foram descritos pelos panelistas como "cárneo" e "carne cozida", têm sido relatados na literatura como possuídores de elevados números de compostos de aroma de carne durante as análises de voláteis (1, 7, 8, 20).

Palavras-chaves: Análise sensorial, inosina-5’-monofosfato, cisteína, pH, aroma.

1 — INTRODUCTION

Amino acids, peptides and carbohydrates constitute the basis of meat flavour precursor. Ribonucleotides, especially Inosine-5-Monophosphate (5'-IMP) appear to be an important precursor. A model system contained a sugar source (5'-IMP) and an amino acid (Cysteine), at three different pH (3.0; 4.5; 6.0), was studied. 5'-IMP is the main source of ribose in dead muscle, ribose has been recognized as an important compound in the production of meat flavour [10,13], it has been presented as a good reacting agent in the Maillard reaction. Cysteine was chosen because its known importance in the formation of meat-like flavour compounds [11].

Sensory analysis were realized to assess the overall odour characteristics of the reaction products and to correlate description with volatiles analysis [7]. Through triangle and profile tests the main differences and general descriptions were picked up, related to the influence of pH on these systems.

The model system ribose/cysteine has been the subject of a number of studies related to meat flavour and therefore much is known about the volatiles produced by this system, as well as their flavour significance [1,2,3,13]. So, the sensory evaluation was undertaken to compare the two meat-like model systems.

2 — MATERIAL AND METHODS

2.1 – Reaction of 5'-IMP or Ribose with Cysteine

Reactions between 5'-IMP or ribose and cysteine were done at three different pHs, i.e. 3.0; 4.5 and 6.0. The range chosen relates to meat pH (6.0 - 5.9) and to 5'-IMP stability, which starts to thermally degrade at pH 4.5 [9, 14, 18]. Each reaction mixture was prepared under similar conditions to those used for the model system reactions reported by Madruga [6, 7]. After cooling the ampoules, each mixture was diluted to 25 times using distilled water. This concentration resulted from a panel consensus during preliminary trials, which selected this dilution as the best for presentation of samples to panellists. A total of six reaction mixtures were prepared as displayed in Table 1.

TABLE 1.
Reaction mixtures used in sensory analysis

Reactants (3ml of each)	PH	code
0.1M IMP + 0.1M cysteine	3,0	ICL
0.1M IMP + 0.1M cysteine	4,5	ICM
0.1M IMP + 0.1M cysteine	6,0	ICH
0.1M ribose + 0.1M cysteine	3,0	RCL
0.1M ribose + 0.1M cysteine	4,5	RCM
0.1M ribose + 0.1M cysteine	6,0	RCH

All six reaction mixtures were used for profiling test, but only four, i.e. ICM, ICH, RCM and RCH were assessed in the triangle test.

2.2 – Sensory analysis of the reaction system containing 5'-IMP or Ribose and Cysteine

Sensorial analysis of Model System were performed in two assessments: the triangle test and aroma profiling. In both, panels were formed from student and staff of the Food Studies Department (University of Reading), who where experienced in sensory evaluation of flavour. Results of the triangle test were used in the selection of assessor for profile test. Aroma profiling was used to provide a detailed and descriptive evaluation of the flavour attributes resulted from the model systems analyzed [5,16].

2.2.1 - Triangle test

Two way triangle tests [5,16] were carried out during four sessions, using 5'-IMP+Cysteine and Ribose+cysteine at pHs 4.5 and 6.0 as reaction mixtures.

The panel was formed by twenty one assessors. Panellists were presented with two set of three samples, the odd sample in the first judgement was used as two identical samples in the second judgement.

Diluted solutions (10ml) of the reaction mixtures were presented to the panellists in 60ml glass stoppered reagent bottles, covered with aluminium foil in order to avoid any visual bias, each one coded with three digit random numbers.

Each assessor was asked to shake the sample before removing the lid and to smell the sample in the appropriate order provided. They were asked to identify the odd sample and to describe the difference of odour sensation on a sheet.

Samples were presented in balanced order for each assessment in order to avoid positional bias and contrast effect [5]. Analysis of the results of the two-way triangle test was based on tables prepared by Roessler et al [17].

2.2.2- Aroma Profiling

Aroma Profiling was performed in two stages: the profile development and the quantitative descriptive analysis.

A panel of twelve assessors were selected from the triangle test for this test, assessors who had a high number of correct answers and ability to describe the difference of odour perception were chosen.

All six diluted mixture reactions (ICL, ICM, ICH, RCL, RCM, RCH), were presented in two sessions during the profile development stage. Each assessor was asked to describe the aroma of each reaction using their own terms and/or a list of terms obtained from the result of the triangle test.

Thirteen descriptive terms were selected for using in the quantitative descriptive analysis. During the two sessions all description was discussed by the panellists, lead by a panel leader, and agreement concerning the meaning of terms was obtained.

Two sets of three different samples (served in the same conditions of triangle test) was presented to each assessor in five sessions during the quantitative descriptive analysis. They were asked to smell and score the intensity of each attribute of each sample in order provided on a sheet. The second set was assessed after 5 to 10 minutes resting from the first judgment in order to avoid fatigue. Five replicates were realized in balance design in order to fulfil the statistical analysis [15], also the presentation was in systematic order to avoid the contrast effect.

Unstructured line, 100mm long, was used for quantitative descriptive analysis marked with "none" at the left and to "extreme" at the right end.

Analysis of the results of the aroma profiling was carried out on an IBM-compatible PC using a set of statistical programmes adapted for sensory analysis (SENPAK, Reading Scientific Service Ltd). The statistical analysis used included analysis of variance (ANOVA) and principal component analysis (PCA).

3 — RESULTS AND DISCUSSION

Triangle test was a good tool for training and select the panel, also differences among samples were previously detected. Heated reaction mixtures of 5'-IMP/cysteine and ribose/Cysteine both at pH 4.5 and 6.0 could be clearly differentiated from each other by the panel in this test (Table 2).

TABLE 2.
Results of two-way triangle test showing the proportion of correct judgements in the comparison of aromas of four reaction mixtures

	ICH	RCM
ICM	23/42**	23/42**
RCM	28/42**	37/42***

significance ** P< 0.01 (1%)

*** P< 0.001 (0.1%)

Profiling aroma gave an indication of how much the reaction mixtures differed from each other, a clear differences in the perception and description of the overall aroma of the six reactions: 5'-IMP/cysteine and ribose/cysteine, pH 3.0, 4.5, 6.0, was established.

Burnt sugar and fruity were characteristic attributes for reaction involving ribose at pH 6.0; egg-like was characteristic for 5'-IMP system at pH 6.0; and the boiled meat, sulphur and meaty attributes were used to describe both 5'-IMP and ribose systems at pH 3.0 and 4.5 (Figure 1). Farmer et al [1] reported "sulphurous, rubbery" description for the overall odour of reaction mixtures containing cysteine/ribose and cysteine/ ribose/phospholipid, also the system ribose/cysteine had some underlying meaty odour. Madruga [6] cited "meaty, roasted pork" and " rubbery, rotten eggs, sulphury", respectively for descriptions of reaction mixtures containing 5'-IMP/cysteine at pH 3.0/4.5 and pH 6.0.

FIGURE 1.
Aroma Averages of Model Systems using 5'-IMP/Cysteine and Ribose/Cysteine at pH 3.0; 4.5; 6.0.

Qualitative and quantitative differences in aroma descriptions were identified in both systems. Six attributes (boiled meat, burnt sugar, egg-like, fruity, meaty and sulphur), differentiated the 5'-IMP and ribose reaction systems at the three different pH. Of thirteen tested attributes, six could be used as discriminators of reaction systems, whereas the seven other attributes were only used as descriptors [16]. ANOVA tests shown that the F-ratio of these six attributes were highly significant between samples (Table 3).

TABLE 3.
Means and F-Ratio from the ANOVA results of six reaction mixtures

ATTRIBUTES	ANOVA MEANS (sample)	ANOVA F-RATIO (sample)
Boiled meat	17.52	5.48**
Burnt ham	2.65	0.86
Burnt meat	4.37	1.93
Burnt rubber	4.10	1.64
Burnt sugar	0.97	2.36*
Egg-like	11.07	7.07**
Fruity	7.97	5.78**
Meaty	19.73	7.37**
Onion	20.37	2.06
Rotten vegetation	7.78	1.13
Sickly	5.73	0.26
Sulphur	23.18	3.56**
Stale	4.95	0.37

Significance * P < 0.05 (5%)

** P < 0.01 (1%)

The profile of six sample was obtained by applying PCA to the data sets of the six main attributes. Figure 2 shows the correlation between the attributes and the sample scores. The profile of aroma generated by 5'-IMP and ribose (pH 3.0 and 4.5) mixture reactions was located in one quadrant of the first two PC plot, whereas 5'-IMP system (6.0); ribose system (pH 6.0) were placed separately in the other two different quadrants. In this case, 5'-IMP and ribose model reactions at pH 6.0 generated different aromas.

FIGURE 2.
PC plot showing the correlation of attributes and six reaction mixture

The high scores for boiled meat and meaty odours in ribose (pH 4.5) and 5'-IMP (pH 3.0) systems (Figure 2) is supported by the fact that decomposition of 5'-IMP is related to acidic conditions (9), also the Maillard reaction which is important route in producing aroma compounds is pH dependent [4, 12, 7]. Chemical analysis of 5'-IMP/cysteine aqueous systems at pH 3.0, 4.5 and 6.0, resulted in a higher production of volatile at pH 3.0 and 4.5, at these pH heterocyclic thiols were predominated [6, 7].

5'-IMP degradation also appears to play an important role in meat flavour formation and it is pH-temperature dependent [6, 14, 18]. Thermal degradation of 5'-IMP was more intensive at pH 3.0 and as consequence high amounts of aroma precursors such as 2-furfural were formed [6]. 2-furfural and/or 4-hydroxy-5-methyl-3(2H)furanone (HMFone) could react with H₂S (produced from cysteine), generating these meaty aroma compounds, as the route described by van den Ouwerland and Peer [19].

Terms with boiled meat, meaty and sulphur characteristics are well correlated with the positive direction of PC1. The negative direction of PC1 was characterized by burnt sugar and fruity. Boiled meat and burnt sugar are descriptors that were moderate correlate with the positive direction of PC2, while egg-like were negative well correlate. Figure 2 shows ribose and 5'-IMP systems (pH 3.0 and 4.5) located in the same quadrant. Whereas the 5'-IMP and ribose systems at pH 6.0 were placed separately in others quadrants. Therefore, it could be suggested that 5'-IMP and ribose systems, at pH 3.0 and 4.5 respectively, seems to have a similar paths-way for the generation of aroma compounds, although the intensity of overall odour impression deferred. The most notable fact on these model systems was the direct participation of 5’-IMP, ribose and low pH on the formation of ‘‘meaty" aromas. Inosine-5’-monophosphate has been recognized as a savoury compound which gives the "umami" taste to many foods. It is also believed to be a flavour potentiator which by itself, has no sensory property, but can enhance the flavour of other agents. So the results suggested that 5’-IMP and ribose, probably can enhance the meaty, boiled meat, sulphur characteristics, in meat products when processed at low pH.

4 — CONCLUSION

From these sensory chemical analysis, it has become clear that the composition and the intensity of aroma compounds produced by 5'-IMP is greatly influenced by pH in the range of 3.0 to 6.0. The optimum meaty aroma and the large number of meat-like compounds from 5'-IMP system resulted from reactions at pH 3.0.

5 — LITERATURE CITED

6 — ACKNOWLEDGEMENTS

The author wishes to thank Dr. D.S. Mottram for his supervision and assistance during this work and the assessors for the sensory analysis from the Department of Food Science and Technology (University of Reading), where the work was developed.

³ Dep. Food Science and Technology. University of Reading. Whiteknights, Reading, RG6 2AP, England - UK.

* A quem a correspondência deve ser endereçada.

• 1. FARMER, L.J.; MOTTRAM, D.S. and WHITFIELD, F.B. Volatile compounds in Maillard reactions involving cysteine, ribose and phospholipids. J. Sci. Food Agric., v. 49, p. 347-368, 1989.
• 2. FARMER, L.J. and MOTTRAM, D.S. Recent Studies on the formation of meat-like aroma compounds. In:BESSIERE, Y & THOMAS, A.F. Flavour Science and Technology, Chichester: Wiley, 1990, p. 113-116.
• 3. FARMER, L.J. The effect of lipid on the formation of volatile aroma compounds by the Maillard reaction Thesis, Univ. Bristol, 1990.
• 4. HODGE, J.E. (1967). Origin of flavor in foods. Nonenzymatic browning reactions. In:SCHULTZ, H.W.; DAY, E.A. & LIBBEY, L.M. Chemistry and Physiology of Flavors, AVI: Westport, 1967,p. 465-491.
• 5. LANMOND, E. Laboratory Methods for Sensory Evaluation of Food Food Research Institute, Canada. 1977
• 6. MADRUGA, M.S. Studies on some factors affecting meat flavour formation Thesis, Univ. Reading, 1994
• 7. MADRUGA, M.S. The effect of pH on the formation of volatile compounds produced by heating a model system containing 5'-IMP and cysteine. Journal of the Brazilian Chemical Society, 1998. in press
• 8. MADRUGA, M.S. Revisao: Formacao do aroma carneo. Bol. SBCTA, v. 31, p. 33-41, 1997.
• 9. MATOBA, T.; KUCHIBA, M.; KIMURA, M. and HASEGAWA, K. Thermal degradation of flavor enhancers, inosine-5'-phosphate and guanosine-5'-phosphate in aqueous solution. J. Food Sci., v. 53, p. 1156-1159,1170, 1988.
• 10. MAY, C.G. An Introduction to Synthetic Meat Flavour. Food Trade Review, v. 44, p. 7,9-10, 12-14, 1974.
• 11. MORTON, I.D.; AKROYD, P.; MAY, C.G. Flavouring substances. GB Patent 836694, 1960.
• 12. MOTTRAM, D.S. and LESEIGNEUR, A. The effect of pH on the formation of aroma volatiles in meat-like Maillard systems. In: BESSIERE, Y & THOMAS, A.F. Flavour Science and Technology, Chichester: Wiley, 1990, p. 121-124.
• 13. MULDERS, E.J. Volatile components from the non-enzymic browning reaction of the Cysteine/Cystine-Ribose system. Z. Lebensm. Unters. Forsch, v. 152, p. 193-201, 1973.
• 14. NGUYEN, T. T. and SPORNS, P. Decomposition of the flavor enhancers, monosodium glutamate,inosine-5'-monophosphate and guanosine-5'-monophosphate during caning. J. Food Sci., v. 50, p. 812-814,822, 1985.
• 15. O'MAHONY, M. Sensory Evaluation of Food: Statistical Methods and Procedures Marcel Dekker Inc, New York, 1986.
• 16. POWERS, J.J. Current practices and application of descriptive methods. In: PIGGOTT, J.R. Sensory Analysis of Foods, Elsevier Applied Science: London, 1988: p. 187-266.
• 17. ROESSLER, E. B.; WARREN, J. and GUYMON, J.F. (1948). Significance in triangular taste tests. In Lanmond, E., Laboratory Methods for Sensory Evaluation of Food Food Research Institute, Canada, 1977.
• 18. SHAOUL, O. and SPORNS, P. Hydrolytic stability at intermediate pHs of the common purine nucleotides in food, inosine-5'-monophosphate, guanosine-5'-monophosphate and adenosine-5'-monophosphate. J. Food Sci., v. 52, p. 810-12, 1987.
• 19. van den OUWELAND, G.A.M. and PEER, H.G. Components contributing to Beef Flavor. Volatile Compounds Produced by the Reaction of 4-hydroxy-5-methyl-3(2H)furanone and Its Thio Analog with Hydrogen Sulfide. J. Agric. Food Chem., v. 23, p. 501-505, 1975.
• 20. ITFIELD, F.B.; MOTTRAM, D.S.; BROCK, S.; PUCKEY, D.J. and SALTER, L.J. Effect of Phospholipid in the Formation of Volatile Heterocyclic Compounds in Heated Aqueous Solutions of Amino Acids and Ribose. J. Sci. Food Agric., v. 42, p. 261-72, 1988.

¹ Recebido para publicação em 10/01/98. Aceito para publucação em 21/10/98.

² Depto. Tec. Química Alimentos (DTQA). Universidade Federal da Paraíba (UFPB). Cidade Universitária. Campus I. CEP 58.059-900. João Pessoa - PB.

Publication Dates

• Publication in this collection
  24 May 1999
• Date of issue
  Oct 1998

History

• Accepted
  21 Oct 1998
• Received
  10 Jan 1998