Doi:10.1016/j.foodchem.2006.05.028

Relation among taste-related compounds (phenolics and caffeine) and sensory profile of erva-mate (Ilex paraguariensis) Nivia Maria Streit a,*, Luisa Helena Rychecki Hecktheuer a, Marta Weber do Canto a, Carlos Augusto Mallmann b, Letı´cia Streck a, Thaylise Vey Parodi a, Liana Pedrolo Canterle a a Departamento de Tecnologia e Cieˆncia dos Alimentos, Centro de Cieˆncias Rurais, Universidade Federal de Santa Maria, Camobi, b Departamento de Medicina Veterina´ria Preventiva, Centro de Cieˆncias Rurais, Universidade Federal de Santa Maria, Camobi, Received 22 October 2005; received in revised form 14 February 2006; accepted 19 May 2006 Erva-mate (Ilex paraguariensis St. Hil.) is a native species from temperate regions of South America, such as Brazil, Paraguay, and Argentina, that is consumed as a beverage known as mate. The objective of this research was to determine the content of caffeic acid,catechin, chlorogenic acid, caffeine, and gallic acid in mate to explain their influence in beverage taste and sensory differences betweennative and reforested plants, as well as between beverages from plants of different regions of Brazil (Santa Catarina and Rio Grande doSul states). Compounds were determined by HPLC and results were related to a sensory evaluation performed by trained tasters. Tastersconsidered the beverage from reforested plants to be more bitter than the beverage from native plants. Beverages from reforested plantshad significantly higher caffeic acid and lower catechin, chlorogenic acid, caffeine, and gallic acid content than native plants. Beveragesfrom plants of Santa Catarina state had significantly higher catechin, caffeine, and gallic acid content than plants from Rio Grande doSul state.
Ó 2006 Elsevier Ltd. All rights reserved.
Keywords: Ilex paraguariensis; Mate; Sensory analysis; Chromatography America, in particular, Ilex paraguariensis Erva-mate may show variations Ilex paraguariensis Saint Hilaire (Aquafoliaceae) is in quality and physico-chemical characteristics due to the known popularly as erva-mate. It is native to temperate cli- influence of some factors such as age of the tree and leaves, mate regions, resists to low temperature, and its natural harvest time, kind of herb (native or reforested), cultivation occurence area is restricted to Brazil, Paraguay, and Argen- system, producer region, processing, and storage lant beverage similar to green tea and is consumed Polyphenols are the most abundant compounds in the traditionally in South America. This beverage is prepared tea leaves. Among them, flavanols (catechins) constitute by the infusion of green or dry Ilex leaves from South above 30%, and have an important contribution especiallyto the bitter and astringent taste of green tea (). Catechins are found in green * Corresponding author. Address: Laborato´rio de Biofarmacote´cnia – tea together with chlorogenic acids, such as 5-caffeoylqui- Biofar Cidade Universita´ria ‘‘Armando de Salles de Oliveira,’’ Departa- nic acid (5-CQA), which is the major phenolic found in cof- mento de Farma´cia Av. Prof. Lineu Prestes, 580 – Bloco 15, Te´rreo Sa˜o Paulo, SP, Brazil. Tel./fax: +55 11 11 3815 8484.
Caffeic acid is a bitter taste compound, that is usually 0308-8146/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2006.05.028 Please cite this article as: Nivia Maria Streit et al., Relation among taste-related compounds (phenolics and caffeine) and sensoryprofile of erva-mate (Ilex paraguariensis), Food Chemistry (2006), doi:10.1016/j.foodchem.2006.05.028.
N.M. Streit et al. / Food Chemistry xxx (2006) xxx–xxx found in small quantities in the processed arabic coffee Ilex paraguariensis is known for its characteristic bitter taste. Nowadays, it is used in commercial herbal prepara-tions such as a tonic, against cellulite and aging. Some of The erva-mate (Ilex paraguariensis) samples harvested in these pharmacological activities are attributed to the high June 2004 were obtained from two regions of South Brazil: Ilo´polis, in the state of Rio Grande do Sul, and Xaxim, in the state of Santa Catarina. Samples from both regions a significant content of various chlorogenic acids (CGA) were divided into native or reforested according to the cul- tural practice used. Samples were grounded, dried packed Some flavonoids are commonly found as polymers in aluminum packages and polyethylene film, and stored such as tannins. While monomeric flavonoid compounds at À20 °C to protect against humidity and keep their phys- are more bitter than astringent, the polymeric compounds ical, chemical, and microbiological characteristics.
Infusion of erva-mate was prepared according to ). Tannins are compounds of high molecular weight, which give the sensation of astringency to food, and are modification. A 5 g sample was mixed with 80 ml of dis- classified into two groups based on their structural types: tilled tri-deionized water at 80°C for 5 min. After cooling, hydrolysable tannin and condensed tannin. The first the volume was adjusted to 100 ml and filtered. For all group has a glucose central nucleus or a polyhydric alco- the analysis samples were run in triplicate.
hol esterified with gallic or elagic acid, and it is easilyhydrolyzed by acids, bases, or enzymes. The second group includes catechin polymers and/or leucoanthocyanidinsthat are not easily hydrolyzed by acid treatment Caffeic acid, (±)-catechin hydrate, chlorogenic acid, caf- feine, and gallic acid standards were purchased from Caffeine is neutral compound when in water solution, Sigma–Aldrich (St. Louis, MO, USA). HPLC-grade meth- and its molar sensory perception is 7.0 · 10À4 anol, acetic acid and acetonitrile were from Tedia (Fair- alkaloid with a distinct bitter taste. Several studies triedto establish a correlation between the bitter characteristics of coffee as a beverage and its caffeine content, but with lit-tle success. As a matter of fact, it has been found that caf- Analyses were performed using an HP 1100 series (Agy- feine contributes only with a small proportion (ranging lentÒ) liquid chromatograph system, equiped with quater- from 10% to 30%) of the detected bitter taste of coffee nary pump, auto sampler, and thermostatted column compartment (35 °C). Compounds were separated using a Under stress conditions, some plants have the capability C18 reverse phase column (5 lm, 150 · 4.6 mm ID) LUNAÒ, protected by a PhenomenexÒ C18 guard column Moreover, an increasing trend on alkaloid (4 L · 3.0 mm ID). Detection was carried out at 274 nm.
production according to the degree of shade was observed The gradient elution consisted of 0.5% aqueous acetic acid (v/v) (solvent A), methanol (solvent B), and acetoni- The increase of defense substances against defoliating trile (solvent C). The gradient conditions were: 0–11 min, insects and fungi (ex. caffeine and theobromine) on plants 70% A, 30% B; 11–14 min, 100% C; 14–20 min, returning under a higher degree of shade may be a way to guarantee to 70% A, 30% B. The flow rate was 0.5 ml/min and the leaves longevity. Therefore, the leaves would stay on the plant for longer to perform photosynthesis, thus compen- Calibration curves were constructed for the compounds sating the biologic investment necessary to the construction evaluated (caffeic acid, catechin, chlorogenic acid, caffeine, and gallic acid) at five concentrations ranging from 2 to The objective of the present research was to evaluate 5000 lg mlÀ1. Curves obtained had r2 greater than 0.999.
taste-related compounds (caffeic acid, catechin, chlorogenic Quantification limit to the method for caffeic acid, cate- acid, caffeine, and gallic acid) and the sensory profile of chin, chlorogenic acid, caffeine, and gallic acid were 0.8, erva-mate infusions (Ilex paraguariensis) prepared using 1.0, 1.0, 0.8 and 0.6 lg mlÀ1, respectively. Recoveries from plants from different regions of Southern Brazil (Rio erva-mate samples spiked with known amounts of these Grande do Sul and Santa Catarina states) and different cul- compounds ranged between 93% and 110%.
tural practices (native and reforested). Results of the chem-ical and sensory analysis were correlated in order to identify the main compounds possibly related to the bittertaste, and explain the differences of taste between native Sensory tasters were recruited using questionnaires about their alimentary habits and healthy state. They were Please cite this article as: Nivia Maria Streit et al., Relation among taste-related compounds (phenolics and caffeine) and sensoryprofile of erva-mate (Ilex paraguariensis), Food Chemistry (2006), doi:10.1016/j.foodchem.2006.05.028.
N.M. Streit et al. / Food Chemistry xxx (2006) xxx–xxx Table 1Phenolic compounds and caffeine in erva-mate beverage from samples harvested in the south of CV = coefficient of variation. Different letters in the same line indicate significant differences (p 6 0.05).
A Results are expressed on a dry matter basis and are the average concentration ± standard deviation of four samples analyzed in June 2004.
trained using ordering tests for basic tastes and triangular discriminative tests. Triangular test was used in the trainingfor bitterness perception. Training was done using the shows the content of caffeic acid, catechin, extract of mate beverage fortified with 0.0025%, 0.005% chlorogenic acid, caffeine, and gallic acid in erva-mate sam- and 0.01% caffeine. The extract was prepared by mixing ples from the south of Brazil. ANOVA revealed significant erva-mate sample with 70 °C water (1:50, w/v) for 3 min effects of the regions and the cultural practices on the levels (). The beverage was put into thermo flasks of chemical compounds in erva-mate extracts. Catechin to keep temperature at 60 °C. Thirty-two candidates were was the compound found at the highest level in the infusion recruited, but due to personal schedule limits, only 20 peo- of Ilex paraguariensis (Caffeic acid and caffeine ple participated in the sensory training program, and 15 concentrations of erva-mate samples from the south of were selected to take part in the sensory evaluation of Brazil were in the range previously reported in samples from Argentina (0.023% and 1.92 g%, respectively) Erva-mate extract for sensory evaluation was prepared as described above, except that it was not fortified with caf- feine. Sensory evaluation was carried out employing had significantly higher caffeic acid and lower catechin, trained tasters (15). For Triangular discriminative test sam- chlorogenic acid, caffeine, and gallic acid content as com- ples of native (N) and reforested (R) erva-mate from Rio pared to the beverage from native plants. Beverage from Grande do Sul (RS) and Santa Catarina (SC) were pre- plants of Santa Catarina state had significantly higher cat- sented into six combinations (RS–N versus RS–R, SC–N echin, caffeine, and gallic acid content than from Rio versus SC–R, RS–N versus SC–N, RS–N versus SC–R, RS–R versus SC–R, and RS–R versus SC–N). In each Triangular sensory testing revealed differences combination, samples were displayed in two different in the taste of beverages brewed from native and reforested sequences, for example: (a) RS–N, RS–N, and RS–R, or plants as well as brewed from plants from the state of Rio (b) RS–N, RS–R, and RS–R; so that 50% of the tasters Grande Sul and state of Santa Catarina. shows the could analyze sequence ‘‘a’’ and the other 50% sequence results of the ordering sensory test. Ordering test revealed ‘‘b’’. Each combination was analyzed in one session, in that samples from reforested areas of Rio Grande do Sul adequate schedule, totaling three days.
state were considered the bitterest comparing to the other The ordering test was carried out to determine the bit- The results of sensory analysis may be explained by the levels of catechin, caffeic acid, and chlorogenic acid found received the four samples of erva-mate infusion (RS–N, in the extracts indicating the role of these substances on the RS–R, SC–N, and SC–R) and were asked to order samples according to the increase of the bitter taste, so that the bit-terest sample received the higher score.
Table 2Sensory score of triangular test for the taste of erva-mate beverages Results of chemical assays were analysed by two-way analysis of variance: Two regions (RS and SC) · two cul- tural practices (native and reforested), followed by Dun- can’s test. Analyses were performed using ‘STATISTICA’ 6.0. Results of the triangular test were analysed using the chi-square test in accordance to Results of the ordering test analysis were evaluated based on the New- a Sample codes: N: native; R: reforested; RS: Rio Grande do Sul; SC: Santa Catarina. Test evaluate by 15 sensory tasters.
Please cite this article as: Nivia Maria Streit et al., Relation among taste-related compounds (phenolics and caffeine) and sensoryprofile of erva-mate (Ilex paraguariensis), Food Chemistry (2006), doi:10.1016/j.foodchem.2006.05.028.
N.M. Streit et al. / Food Chemistry xxx (2006) xxx–xxx plants had the highest chlorogenic acid concentrations, Sensory score of the ordering test according to the bitterness of erva-mate which probably helped to diminish the bitter taste of these samples when compared to the reforested ones.
Native plants are found within woods and therefore are submitted to a higher degree of shade. According to The critical value for the 15 tasters and 4 samples, at a level of significance of 5%, according to Newell and Mc Farlane’s Table (is intensity and stress conditions, increase the production of 19. Different letters indicate significant differences (p 6 0.05).
defence substances, such as caffeine. The increase of these A Sample codes: N: native; R: reforested; RS: Rio Grande do Sul; SC: protective substances against defoliating insects and fungi can be a way to guarantee longevity to leaves. Accordingly,native plants had higher caffeine content than reforestedones.
Catechin is a monomeric flavonoid characterized by Since caffeine is a bitter substance, its higher concentra- tion in native samples, which were less bitter, was unex- Astringency is usually considered to be a tactile sensation pected. All erva-mate samples had significant levels of resulting from the precipitation of salivary proteins and glucose (data not shown). Hence, a possible explanation for this unexpected result is that caffeine bitter taste while bitterness is a taste mediated by sensory receptors may have been masked by a sweet taste. Taste may be ). Catechin is not chemically defined as influenced by the molecular packing of taste molecules astringent, since poli-flavonoids with molecular weight close to the sensory receptor proteins that are inserted below 500 usually do not precipitate proteins ( in the cell membrane (an hydrophobic phase) Nevertheless, the astringency of this small flavonoid molecule has been attributed to the precipitation or a taste was found to be suppressed by sweeteners such as strong link with proteins, due to the presence of dihy- droxy-1,2 or trihydroxy-1,2,3 groups, which can explain As a general rule, sweeteners are rather hydrophilic the astringency property in flavan-3-ols monomers and bitter molecules have predominantly a hydrophobic character. The inhibition of the bitter taste by sweet sub- ated the interaction between bitter and astringent tastes.
stances has been attributed to hydration and surface prop- the omission of catechin from a biomimetic reconstitution organization of bitter (caffeine) and sweet (sucrose) sub- of tea taste decreases both astringency and bitterness. In stances in an aqueous solvent near a modeled hydropho- contrast, we found higher catechin levels in beverages from bic phase revealed that caffeine molecules form an native erva-mate samples which were rated as less bitter adsorption layer, whereas sucrose may induce the dissoci- than the reforested ones. Therefore, we can propose that ation of some caffeine aggregates and slow down their catechin may not be a key compound in the bitterness of erva-mate. Besides, the higher catechin content of native study was found on the taste interaction between caffeine erva-mate extract could have masked the perception of bit- and glucose, it is possible that caffeine–glucose interaction terness of other substances, due to their astringency.
could mask the bitter taste of caffeine in erva-mate Caffeic acid is a bitter taste compound that is normally found in trace quantities in arabic coffee, and it may Polymeric phenolic compounds, like gallic acid are more increase when chlorogenic acid is hydrolyzed in the bitter erages from native plants gallic acid concentration was ples from reforested plants which had higher concentra- higher when compared to reforested ones. According to tions of caffeic acid were considered bitterer by tasters tasters, beverage from native plants had lower bitter taste.
Astringent compounds, like gallic acid could help to Chlorogenic acid is the major phenolic compound decrease the perception of bitterness. However, as this responsible for astringency of green coffee ( compound was found at very low concentration, its influ- Chlorogenic acid, a caffeic acid ester, is known as ence on the erva-mate beverage astringent taste may not caffeoylquinic acid with an agliconic portion of caffeic acid.
As mentioned before, the astringent chlorogenic aciddecreases as a consequence of hydrolysis releasing the bit- ter caffeic acid in samples of monsooned coffee (). Besides, there are indications that small mol- Authors thank to Indu´stria Irma˜os Folle Ltda for finan- ecules of phenolic compounds can obtain astringency grad- cial support and sample donation, and Capes for a Master ually such as 5-CQA (5-caffeoylquinic) that has only one Degree Fellowship to N.M.S. Authors also thank professor caffeic acid residue and has been considered astringent Tatiana Emanuelli for valuable contributions on the prep- Please cite this article as: Nivia Maria Streit et al., Relation among taste-related compounds (phenolics and caffeine) and sensoryprofile of erva-mate (Ilex paraguariensis), Food Chemistry (2006), doi:10.1016/j.foodchem.2006.05.028.
N.M. Streit et al. / Food Chemistry xxx (2006) xxx–xxx Hoft, M., Verpoorte, R., & Beck, E. (1998). Growth and alkaloid patterns of roots of Tabernaemontana pachysiphon and Rauvolfia mombasi- Adzet, T., Camarasa, J., & Laguna, J. C. (1987). Hepatoprotective activity ana as influenced by environmental factors. Acta Botanica, 111, of polyphenolic compounds from Cynara scolymus against CCl toxicity in isolated rat hepatocytes. Journal of Natural Products, 50(4), Kilmartin, P. A., & Hsu, C. F. (2003). Characterization of polyphenols in green, oolong, and black teas, and in coffee, using cyclic voltammetry.
Anua´rio Brasileiro da Erva-Mate. (1999). Atividade importante no Sul do Paı´s (pp. 53–55). Santa Cruz do Sul:Grupo de Comunicac¸o˜es Matsubara, S. (2001). Polifeno´is em cha´s comercializados no Brasil.
Dissertac¸a˜o (Mestrado em Cieˆncia de Alimentos) (90pp). Faculdade de Aroulmoji, V., Aguie´-Be´ghin, V., Mathlouthi, M., & Douillard, R. (2004).
Engenharia de Alimentos, Universidade Estadual de Campinas, Effect of sucrose on the properties of caffeine adsorption layers at the air/solution interface. Journal of Colloid and Interface Science, 276, McManus, J. P., Davis, K. G., Lilley, T. H., & Haslam, E. (1981). The association of proteins with polyphenols. Journal of the Chemical Aroulmoji, V., Hutteau, F., Mathlouthi, M., & Rutledge, D. N. (2001).
Society. Chemical Communications, 309–311.
Hydration properties and the role of water in taste modalities of Meilgaard, M., Cirille, G. V., & Carr, B. T. (1991). Sensory evaluation sucrose, caffeine, and sucrose-caffeine mixtures. Journal of Agricultural techniques (2nd ed.). Boca Raton: CRC, pp. 135–185.
and Food Chemistry, 49, 4039–4045.
Naish, M., Clifford, M. N., & Birch, G. (1993). Sensory astringency of 5- ASTM (1968). Establishing conditions for laboratory sensory evaluation of O-caffeoylquinic acid, tannic acid and grape seed tannin by a time- foods and beverages. Manual on sensory testing methods. ASTM STP intensity procedure. Journal of the Science of Food and Agriculture, 61, Bate-Smith, E. (1973). Haemanalysis of tannins: the concept of relative Pfaffmann, C., Bartoshuk, L. M., & McBurney, D. H. (1971). Taste astringency. Phytochemistry, 11, 907–912.
Psychophysics. In L. M. Beidler (Ed.), Handbook of sensory Physiology.
Clarke, R. J., & Macrae, R. (1985). Coffee: chemistry (Vol. 1). Essex, IV. Chemical Senses. 2. Taste (pp. 75–101). Berlin: Springer-Verlag.
Purdon, M. P., & McCamey, D. A. (1987). Use of a 5-caffeoylquinic acid/ Clifford, M. N., & Ramirez-Martinez, J. R. (1990). Chlorogenic acid and caffeine ratio to monitor the coffee roasting process. Journal of Food purine alkaloids contents of mate (Ilex paraguariensis) leaf and beverage. Food Chemistry, 35, 13–21.
Robichaud, J. L., & Noble, A. C. (1990). Astringency and bitterness of Coelho, G. C. Variabilidade morfolo´gica e quı´mica da erva-mate. In selected phenolics in wine. Journal of the Science of Food and Congresso Sul-Americano da Erva-Mate, 2. e Reunia˜o Te´cnica da Erva-Mate, 3., 2000, Encantado. Anais. Porto Alegre: Edigraf.
SAAP (1997). Secretaria da Agricultura e Abastecimento do Parana´.
Erva-mate: prospec¸a˜o tecnolo´gica da cadeia produtiva. Curitiba: Duarte, F. (2000). Selec¸a˜o, treinamento de julgadores e metodologia para ana´lise sensorial de extrato de erva-mate. Dissertac¸a˜o (Mestrado em Scharbert, S., & Hofmann, T. (2005). Molecular definition of black tea Tecnologia Quı´mica) (84pp). Setor de Tecnologia, Universidade taste by means of quantitative studies, taste reconstitution, and omission experiments. Journal of Agricultural and Food Chemistry, 53, Dutcosky, S. D. (1996). Ana´lise sensorial de alimentos. Curitiba: Univer- sita´ria Champagnat, pp. 109–115.
´ cidos feno´licos como antioxidantes – Revisa˜o.
Filip, R., Lo´pez, P., Coussio, J., & Ferraro, G. (1998). Mate substitutes or Revista de Nutric¸a˜o, 15(1), 71–81, jan/abr, Campinas.
adulterants: study of xanthine content. Phytotherapy Research, 12, Variyar, P. S., Ahmad, R., Bhat, R., Niyas, Z., & Sharma, A. (2003).
Flavoring components of raw monsooned arabica coffee and their Filip, R., Lo´pez, P., Giberti, G., Coussio, J., & Ferraro, G. (2001).
changes during radiation processing. Journal of Agricultural and Food Phenolic compounds in seven South American Ilex species. Fitotera- Viani, R. (1988). Physiological active substances in coffee. In R. J. Clarke Finger, A., Kuhr, S., & Engelhardt, U. H. (1992). Chromatography of tea & R. Macrae (Eds.), Coffee: Physiology (pp. 1–25). London: Elsevier constituents. Journal of chromatography, 624, 293–315.
Hoft, M., Verpoorte, R., & Beck, E. (1996). Growth and alkaloid contents Vidal, S., Francis, L., Noble, A., Kwiatkowski, M., Cheynier, V., & in leaves of Tabernaemontana pachysiphon Stapf (Apocynaceae) as Waters, E. (2004). Taste and mouth-feel properties of different types of influenced by ligth intensity, water and nutrient supply. Oecologia, 107, tannin-like polyphenolic compounds and anthocyanins in wine.
Analytica Chimica Acta, 513, 57–65.
Please cite this article as: Nivia Maria Streit et al., Relation among taste-related compounds (phenolics and caffeine) and sensoryprofile of erva-mate (Ilex paraguariensis), Food Chemistry (2006), doi:10.1016/j.foodchem.2006.05.028.

Source: http://www.lamic.ufsm.br/papers/20070209_compounds.pdf