Research Article

Received: 15 June 2010 Revised: 1 September 2010 Accepted: 1 October 2010 Published online in Wiley Online Library: 27 October 2010

(wileyonlinelibrary.com) DOI 10.1002/jsfa.4205

Distribution of aroma volatile compounds in tangerine hybrids and proposed inheritance

Takayuki Miyazaki,a Anne Plotto,b Kevin Goodnerc and Fred G Gmitter Jra
Abstract
BACKGROUND: With the desirable combination of sugars and acids, volatile compounds contribute to the essential organoleptic attributes of citrus. This study evaluated the aroma volatiles of 20 tangerine hybrids of the University of Florida breeding program. Volatiles were sampled from hand-squeezed juice by headspace solid-phase microextraction (SPME), and analyzed by gas chromatographymass spectrometry. Principal component analysis (PCA) and cluster analysis (CA) were used to nd similarities among samples due to volatile composition with effect of genetic background. RESULTS: In total, 203 volatiles were detected in all samples. Volatiles in lower amounts were widely distributed among samples and were classied mainly as terpene hydrocarbons and oxygenated compounds, such as aldehydes, esters, alcohols and ketones. PCA, based on relative peak areas (content) clearly separated the samples higher in volatile content, mainly those with sweet orange genetic contributions in their background. CA, based on volatile presence/absence, grouped samples into ve clusters, each showing distinctive volatile proles. CONCLUSION: The genetic background of tangerine hybrids affected volatile composition and content of samples. In general, tangerines were characterized by fewer volatiles (in both quality and quantity) and more aldehydes, and hybrids with sweet orange in their background had more sesquiterpenes and esters, which would likely affect their aroma. Published 2010 by John Wiley & Sons, Ltd. Keywords: Citrus reticulata; citrus breeding; aroma volatiles; fruit quality; tangerine; mandarin

INTRODUCTION
Florida fresh tangerines (Citrus reticulata Blanco, sometimes referred to as mandarins in the USA) as well as oranges (C. sinensis L. Osb.) and grapefruit (C. paradisi Macf.) are among the largest agricultural commodities in the US citrus market. While grapefruit are very sensitive to endemic citrus canker disease,1 tangerines can be fairly tolerant and therefore provide an alternative to grapefruit production for the future fresh fruit industry in Florida. The fresh tangerine fruit is widely consumed because of its desirable qualities of aroma, avor and ease of peeling. In addition, the high nutritional content from vitamin C and carotenoids in tangerines benets human health. From 2007 to 2009, the average annual US production of tangerines/mandarins was 485 000 tons, increasing from 361 000 tons in 2006, and with fresh fruit production accounting for 7080%.2 In Florida, that production was ve and a half million 95 lb (43 kg) boxes (237 000 tons) in the peak year.2 Improvement in fruit quality is one of the primary goals of fresh tangerine breeding programs in Florida. Commercial tangerine fruits are currently graded by inspectors, with grading based only on fruit appearance factors such as color, size and damage.3 However, new quality standards based on avor attributes could be created to be used in the selection of high-quality fruits. Conventional citrus breeding, based on crossing and selection, requires a large amount of space, and considerable time and labor during the long juvenile period of citrus trees; such efforts

are rather costly to conduct. In the last decades, advancement of plant genetics and genomics has made citrus breeding more efcient. The application of molecular markers to citrus breeding may allow breeders to select superior recombinants improved for multiple traits by conventional breeding efforts, using marker-

Correspondence to: Anne Plotto, USDA/ARS, Citrus and Subtropical Products Laboratory, 600 Avenue S, NW, Winter Haven, FL 33881, USA. E-mail: plotto@ars.usda.gov

This article is a US Government work and is in the public domain in the USA. Mention of a trademark or proprietary product is for identication only and does not imply a guarantee or warranty of the product by the US Department of Agriculture. The US Department of Agriculture prohibits discrimination in all its programs and activities on the basis of race, color, national origin, gender, religion, age, disability, political beliefs, sexual orientation, and marital or family status. Part of this work was presented at the Florida State Horticultural Meeting, 1-2 June 2009. a Institute of Food and Agricultural Sciences, University of Florida, Citrus Research and Education Center, Lake Alfred, FL 33850, USA b USDA/ARS, Citrus and Subtropical Products Laboratory, Winter Haven, FL 33881, USA

449

c Sensus, LLC, Hamilton, OH 45011, USA

J Sci Food Agric 2011; 91: 449460

www.soci.org

Published 2010 by John Wiley & Sons, Ltd.

www.soci.org assisted selection (MAS).4 It can also effectively help in removing plants producing fruit with poor attributes from the program at an earlier stage. Molecular markers associated with citrus fruit quality can be a valuable tool for genetic improvement, leading to the faster release of new superior scions. Aroma, as well as taste, color and texture, is one of the most important quality attributes of citrus fruits. The aroma volatiles of the major processing orange and grapefruit cultivars have been well investigated, and over 300 aroma volatiles have been reported from gas chromatography (GC) and gas chromatographymass spectrometry (GC-MS) analyses of fresh orange juices.5 It is well known that citrus aroma components are a mixture of monoterpenes, sesquiterpenes, alcohols, aldehydes, acids, esters, ketones, etc. Although there have been several reports on aroma volatiles in tangerine essence and peel oil,6 10 very little information is available regarding those in fresh tangerine fruits.11 15 Information on volatile quality and quantity can be useful to evaluate tangerine fruit avor quality. In addition, peel and juice volatile composition have been analyzed for classication of different citrus fruits: yuzu (C. junos Sieb.) cultivars,16 lemon (C. limon Burm.) cultivars,17 grapefruit hybrids18 and tangerine/mandarin hybrids.13,14 Differentiation of tangerine hybrids based on their aroma proles may lead to better understanding of genetic control of aroma production. The main objectives of this study were to investigate aroma volatile compounds in a population of tangerine hybrids, and to analyze inter-varietal relationships from volatile proles by using multivariate statistics, principal component analysis (PCA) and cluster analysis (CA). It was hypothesized that if similarities were present they would be due to common genetic background.13 The study on aroma volatiles present among tangerine hybrids of diverse origins would provide fundamental information on fruit quality, maturity and development of early DNA-based MAS.

T Miyazaki et al.

Table 1. List of samples and corresponding selection names or parentage, hybrid numbers, harvest dates and sample codes. Several unique hybrids were evaluated from among the various crosscombinations listed Selection name 8-9 Murcott 8-9 Murcott 8-9 Murcott Robinson Fairchild Robinson Fairchild Robinson Fairchild Robinson Fairchild Fallglo Fallglo Fairchild Fallglo Fairchild Fallglo Fairchild 8-9 8-10 Unknown Unknown Unknown 8-9 Orlando 9-4 Blood4x Murcott Temple Sanguinelli Fortune Murcott Ortanique 8-8 Murcott 8-9 VAL4x Hybrid number Harvest date Sample code 1 2 3 1 2 3 4 1 2 3 1 1 1 1 1 1 1 16 Nov. 2007 16 Nov. 2007 16 Nov. 2007 16 Nov. 2007 16 Nov. 2007 14 Dec. 2007 14 Dec. 2007 16 Nov. 2007 14 Dec. 2007 14 Dec. 2007 14 Dec. 2007 14 Dec. 2007 14 Dec. 2007 11 Jan. 2008 11 Jan. 2008 11 Jan. 2008 11 Jan. 2008 11 Jan. 2008 14 Feb. 2008 14 Feb. 2008 14 Feb. 2008 14 Feb. 2008 25 Mar. 2008 25 Mar. 2008 25 Mar. 2008 a c d b e g h f i j k l m o p r n q s t u v w x y

Each sample is a juice composite from fruits harvested from one tree of individual hybrid or commercial cultivar.

EXPERIMENTAL
Plant materials All tangerine hybrids were obtained from the University of Florida Citrus Research and Education Center (UF-CREC) breeding program. The trees were grown under the same environmental conditions of soil, irrigation and illumination at the CREC groves. Fruit were harvested from November 2007 until March 2008 (Table 1). Twelve of the 56 tangerine hybrids evaluated in 2006200713 were selected for re-evaluation in the 20072008 season: samples a, c and d (three seedlings issued from an 89 Murcott cross); samples b and g (two seedlings of a Robinson Fairchild cross); sample i (Fallglo Fairchild); sample l (89); sample m (810); sample n (89 Orlando); samples p and r (two seedlings from unknown parentage); and sample y (89 Val4x). Maturity of fruit from the individual hybrids was determined based on the previous years results13 and conrmed by the breeder and his assistants. Eight new hybrids and ve named commercial cultivars (including one sweet orange) as references were also included for the present study. All hybrids have various genetic backgrounds from tangerines, oranges and grapefruit (Fig. 1).19 27 Each sample was a juice composite from approximately 5060 fruits harvested from one tree of each hybrid accession. A total of 25 samples were prepared at the USDA/ARS Citrus and Subtropical Products Laboratory. Sample preparation Fruit were soaked in warm water with 200 mL detergent (DECCO 241 Fruit and Vegetable Kleen, Monrovia, CA, USA) in a 16 L bucket, washed for about 30 s, and rinsed. Individual fruits were cut in half and juiced manually for 3 s with an electric juicer (model 3183; Oster, Rye, NY, USA), avoiding any scraping of the albedo or squeezing of the avedo to prevent potential peel components (peel oil) from entering the juice. Seeds were removed and aliquots (2.5 mL) of tangerine juice were placed in 20 mL glass vials (Gerstel, Inc., Baltimore, MD, USA) along with saturated sodium chloride solution (2.5 mL) to help drive volatiles into the headspace and inhibit any potential enzymatic activity. 3-Hexanone (1 ppm) was added as an internal standard. The vials were capped with magnetic crimp caps containing Teon-coated septa and stored at 20 C until analyzed. Headspace sampling and GC-MS analysis The extraction of aroma volatiles was performed using SPME with an MPS-2 autosampler (Gerstel). The vials were incubated at 40 C for 30 min in a water bath (Baxter Scientic Products, Cincinnati, OH, USA), and then a 2 cm SPME ber (50/30 m DVB/CAR/PDMS; Supelco, Bellefonte, PA, USA) was inserted into the headspace of the sample vial and exposed for 60 min. The analytes were thermally desorbed in the GC injector (splitless mode) port for 3 min at 250 C. The separation of volatile compounds was accomplished using an Agilent 6890 GC (Agilent Technologies, Santa Clara, CA, USA) equipped with DB-5 (60 m length, 0.25 mm i.d., 1.00 m lm thickness; J&W Scientic, Folsom, CA, USA) and DB-Wax (60 m length, 0.25 mm i.d., 0.50 m lm thickness; Agilent

450

wileyonlinelibrary.com/jsfa

Published 2010 by John Wiley & Sons, Ltd.

J Sci Food Agric 2011; 91: 449460

Evaluation of tangerine aroma volatiles

www.soci.org

Figure 1. Pedigree of tangerine hybrids (from references 1927).

Technologies) columns, coupled with a 5973N MS detector (Agilent Technologies). The column oven was programmed to increase at 4 C min1 from the initial 40 C to 230 C, then ramped at 100 C min1 to 260 C and held for 11.70 min for a total run time of 60 min. Helium was used as carrier gas at ow rate of 1.5 mL min1 . Inlet, ionizing source and transfer line were kept at 250, 230, and 280 C, respectively. Mass units were monitored from 40 to 250 m/z and ionized at 70 eV. Data were collected using the ChemStation G1701 AA data system (Hewlett-Packard, Palo Alto, CA, USA). Samples were run in triplicate on a DB-5 column, with a blank run between each hybrid to ensure ber cleanness between samples. A mixture of C-5 to C-15 n-alkanes was run at the end of each day to calculate retention indices (RIs).28 Samples were also analyzed (one run per sample) on a DB-Wax column to identify potential co-eluting compounds on the DB-5 column. Volatile compound identication Volatile compounds were identied by comparison of their mass spectra with library entries (NIST/EPA/NIH Mass Spectral Library, version 2.0d; National Institute of Standards and Technology, Gaithersburg, MA, USA), as well as comparing RIs with published RIs on both columns. Published RIs were from the NIST/EPA/NIH 29 and Adams.30 Mass Spectral Library, Kondoyan and Berdague Chemical authentic standards, when available, were run on both columns and their RIs and spectra conrmed compound identities. Statistical analyses Volatile compounds were semi-quantied by calculating each peak area relative to the peak area of the internal standard

(relative peak area). These variables were used to perform a PCA based on Pearsons correlations to account for large variation in peak scaling31 to differentiate individual samples based on their volatile composition using XLSTAT software (Addinsoft, Paris, France). With the data transformed into presence/absence (1/0) of volatiles, a cluster analysis was performed to nd correlations between volatile composition and sample origin.13 Clusters were formed using the unweighted pair-group average method with arithmetic mean (UPGMA) and the Kulczynski coefcient, to measure similarities between samples (XLSTAT, Addinsoft).

RESULTS AND DISCUSSION

A total of 146 compounds were identied by GC-MS in the 25 samples, with an additional 57 unknown compounds (Tables 2 and 3). Spectra of tentatively identied compounds, i.e., compounds for which identication with a chemical standard was not conrmed, are shown in Table 4. The average number of volatiles per sample was 77, ranging from 52 to 118, and more than 77 compounds were detected in 12 samples. The volatiles were widely distributed among samples; however, only a small number of volatiles (526 volatiles) contributed to more than 90% of the total content. Many volatile compounds in Tables 2 and 3 have already been reported in tangerine essence, peel oil and juice.6,7,11 15,32 37 All samples contained 15 volatiles in common: ethanol, hexanal, -pinene, -myrcene, octanal, -terpinene, p-cymene, d-limonene, terpinolene, dehydro-p-cymene, linalool, nonanal, decanal, -terpineol and d-carvone (Table 2). d-Limonene was the most abundant compound, representing 40.682.5%

451

J Sci Food Agric 2011; 91: 449460

Published 2010 by John Wiley & Sons, Ltd.

wileyonlinelibrary.com/jsfa

www.soci.org

T Miyazaki et al.

Table 2. List of most abundant aroma volatiles detected by GC-MS among samples and identied by linear retention index (LRI) on DB-5, DB-Wax column and conrmed with chemical standards. Volatiles are listed according to their frequency of appearance, 48100%, in samples 4871%a LRI DB-5 DB-Wax (E)-2-Pentenald Benzaldehyded Allo-ocimene -Selinene -Cubebened -Caryophyllened -Selinene ()- -Panasinsen -Cyclocitrald 2-Undecenald (Z )-p-Mentha-2,8-dien-1-ol -Terpineold Octyl acetated -Elemene Calamenened -Calacorened 1 Unknown monoterpenee 2 Unknown sesquiterpenesf 4 Unknownsg 755 985 1132 1545 1363 1499 1530 1577 1234 1366 1150 1162 1202 1409 1566 1593 1167 1604 1384 2228 1496 2013 2393 1854 2255 1933 1796 1491 1754 2631 3117 Acetaldehyde (E)-2-Decenal -Phellandrened -Terpinened Perillaldehyded Cadinene Propanald Terpinene-4-old p-Menth-1-en-9-al Copaened Caryophyllened 1-Octen-3-one 6-Methyl-5-hepten-2-one p-Menth-1-en-9-al isomer 1-Penten-3-oned (E)-2-Nonenald Geranyl acetoned Valencened -Iononed 2 Unknown monoterpenese 2 Unknownsg 7295%b LRI DB-5 DB-Wax 469 1262 1025 1073 1293 1557 511 1196 1231 1401 1458 988 994 1234 676 1163 1454 1539 1506 659 1868 1195 1253 2447 2309 747 1765 1839 1551 1801 1316 1057 1615 2680 2191 3187 Ethanold Hexanald -Pinened -Myrcened Octanald -Terpinened p-Cymened d-Limonened Terpinolened Dehydro-p-cymened Linaloold Nonanald Decanald -Terpineold d-Carvoned Acetone Ethyl acetated Pentanald (E)-2-Hexenald Heptanald (E)-2-Heptenald -Phellandrened (E)-2-Octenald 1,3,8-p-Menthatriene ()-4-Acetyl-1-methylcyclohexene Dihydrocarvoned 96100%c LRI DB-5 DB-Wax 489 805 956 999 1013 1035 1042 1049 1100 1105 1105 1109 1205 1208 1257 509 599 690 864 913 970 1052 1067 1127 1144 1212 929 1123 1061 1189 1283 1239 1272 1223 1282 1446 1600 1384 1538 2026 2225 775 870 1005 1233 1208 1315 1229 1433 1392 1670 1819

Volatiles detected in 1217 out of 25 samples. Volatiles detected in 1823 out of 25 samples. c Volatiles detected in 2425 samples. d Volatiles conrmed with chemical standards. e Unknown monoterpenes unidentied by GC-MS. f Unknown sesquiterpenes unidentied by GC-MS. g Unknown volatiles unidentied by GC-MS except monoterpenes and sesquiterpenes.
b

of the total aroma volatiles. Among 22 identied monoterpene hydrocarbons, four monoterpenes were also present in relatively high amount in most samples: -myrcene (0.54.9% of relative peak area), p-cymene (0.37.0%), -phellandrene (03.5%) and dehydro-p-cymene (0.59.5%). The other major compounds present in large amount in samples were valencene (024.9%), hexanal (0.114.1%) and linalool (0.59.5%). Conversely, 55 volatiles were present in only one or two samples (Table 3) and therefore might be more cultivar specic. Most volatiles were classied into six chemical classes: monoterpenes, sesquiterpenes, aldehydes, esters, alcohols and ketones (Table 5). The number of volatiles in these classes alone accounted for 76.8% of the total (203 volatiles). In addition, terpene-derived aroma volatiles were grouped into several classes: e.g., neral and geranial in aldehydes, linalyl acetate and citronellyl acetate in esters, linalool and citronellol in alcohols and - and -ionones in ketones. Most volatiles are originated from enzymatic synthesis in terpenoids and fatty acid metabolisms, these biosynthetic pathways playing an important role in aroma production in these

samples. Only a few volatiles were detected in other chemical classes: phenols, ethers, acids and epoxides (Table 5). Principal component analysis In this study, PCA was used as an exploratory technique to identify groups among samples based on volatile composition and relative content. A rst PCA was done using volatile quality (relative peak area) and based on Pearsons correlations to standardize the data. All 203 volatiles were used in this PCA as variables; the PCA reduced the number of variables to 24 principal components (data not shown). The rst and second principal components (F1 and F2) represented 39.80% of the total variance (Fig. 2(A)). The third component explained an additional 11.69% of the variance (Fig. 2(B)). Such a small percent variance in the rst three components is due to the fact that the variance is spread among all 24 components. The plot of scores in the PCA analysis illustrates that samples q (9-4 Blood4x), t (Temple) and u (Sanguinelli orange) were clearly different from the others due to their volatile prole. These samples were much richer in volatiles

452

wileyonlinelibrary.com/jsfa

Published 2010 by John Wiley & Sons, Ltd.

J Sci Food Agric 2011; 91: 449460

Evaluation of tangerine aroma volatiles

www.soci.org

Table 3. List of less abundant aroma volatiles detected by GC-MS among samples and identied by linear retention index (LRI) on DB-5, DB-Wax column and conrmed with chemical standards. Volatiles are listed according to their frequency of appearance, 147%, in samples 111%a LRI DB-5 DB-wax 2-Methyl-2-propanol 2-Butanone (E)-2-Butenal (Z )-3-Hexenald 1-Octen-3-ol p-Mentha-3,8-diene Octanoic acid Camphord Bornyl acetate (E,E)-2,4-Decadienald (Z )-Carvyl acetated -Elemened (S)-Perillyl acetate -Guaiened -Farnesene (E,E)-2,4-Hexadienald Ethyl tiglate 1-Methylbutyl butanoate Methyl octanoated Hexyl butanoated (Z )-Carveold Citronellold (E)-Carveold Nonanoic acid Nonyl acetated Methyl geranate exo-2-Hydroxycineole acetate -Terpinyl acetated Ylangene Decyl acetated -Farnesene 1 Unknown monoterpenee 11 Unknown sesquitepenesf 12 Unknownsg 530 583 639 799 987 1084 1152 1171 1294 1320 1332 1351 1442 1459 1518 923 944 1025 1120 1188 1229 1236 1243 1248 1300 1315 1343 1350 1396 1403 1460 Ethyl propanoated 894 1084 1436 1271 2076 Ethyl butanoated Methyl hexanoated Ethyl hexanoated Ethyl 3-hydroxyhexanoated Ethyl octanoated -Muurolene Caryophyllene oxided Nootkatoned Dimethyl sulde Methyl butanoated Ethyl 2-methylbutanoated Sabinened Hexyl acetate 3-Carened (E)-Ocimene -Cubebene 1-Penten-3-old 2-Pentanone 2,3-Pentanedioned 3-Methyl-butanold Ethyl 2-butenoate (2E)-2-Methyl-2-hexenal Hexanoic acidd 2,3-Octanedione (E,Z )-2,4-Heptadienal (E,E)-2,4-Heptadienald Thymol methyl ether Linalyl acetated -Iononed -Elemene 2,6-Bis(1,1-dimethylethyl)-phenol Nerolidold Dihydroactinidiolide 3 Unknown sesquitepenesf 10 Unknownsg 1223%b LRI DB-5 DB-wax 699 798 929 1001 1128 1188 1534 1658 1881 529 712 856 991 1018 1031 1056 1463 671 674 683 729 848 884 965 991 1007 1022 1231 1242 1439 1455 1523 1585 1592 974 1075 1208 1240 1967 1427 2160 3584 7433 1004 1054 1191 1266 1187 1241 1788 1150 1005 1164 1194 Butanald Nerald Geraniald Methyl acetate 3-Pentanone 4-Heptanone 4-Methyl-hexanal Styrened -Thujened 1,8-Cineoled Dodecanal -Ionone epoxide Undecanald Citronellyl acetated -Pinened Thymold Neryl acetated Geranyl acetated Selina-3,7(11)-diene Juniper camphor 2 Unknown sesquiterpenesf 7 Unknownsg 2447%c LRI DB-5 DB-wax 583 1241 1268 531 685 878 889 911 942 1054 1411 1511 1305 1341 1003 1289 1350 1370 1600 1741 861 2004 2005 784 1002 1162 1262 1063 1228 2076 3500 1782 1913 1149 4594 2126 2261 2137 5244

2212 2200 1397 1245 1373 2536 2186 2664 1689

2063 1973 1918

1517 1532 1766 1638 2735 1890 5439 3646

a c

Volatiles detected in 12 out of 25 samples. Volatiles detected in 35 out of 25 samples. Volatiles detected in 611 out of 25 samples. d Volatiles conrmed with chemical standards. e Unknown monoterpene unidentied by GC-MS. f Unknown sesquiterpenes unidentied by GC-MS. g Unknown volatiles unidentied by GC-MS except monoterpene and sesquiterpenes.
b

than the average sample (77 volatiles), since they contained 118, 109 and 111 volatiles, respectively. According to the rst twocomponent biplot data of this PCA analysis (Fig. 2(A)), many vectors representing sesquiterpenes were in the direction of the rst quadrant (upper right), along with some on the fourth quadrant (lower right), and most vectors representing esters were located on the fourth quadrant (lower right) (data not shown). Therefore, sample q (9-4 Blood4x) may be distinguished by its high number and amount of sesquiterpenes, and samples t (Temple) and u

(Sanguinelli) by their abundance in sesquiterpenes and esters, as can be seen in Table 6. Another type of PCA was performed using relative peak areas of total volatiles in each of the 11 chemical classes from Table 6 (Fig. 3). In this PCA, the rst two components explained 58.41% of the total variance. Ketones and alcohols were highly correlated with each other, as well as with aldehydes. Samples u (Sanguinelli) and t (Temple) had high scores on F1 (esters, other, ketones and alcohols), sample y (89 Val4x) had a high score on F2, and

453

J Sci Food Agric 2011; 91: 449460

Published 2010 by John Wiley & Sons, Ltd.

wileyonlinelibrary.com/jsfa

www.soci.org

T Miyazaki et al.

Table 4. Mass spectra (percent occurrence) data for compounds not conrmed with chemical standards in Tables 2 and 3 Compound Acetaldehyde Acetone 2-Methyl-2-propanol Dimethyl sulde Methyl acetate 2-Butanone (E)-2-Butenal 2-Pentanone 3-Pentanone Ethyl 2-butenoate 4-Heptanone (2E)-2-Methyl-2-hexenal 4-Methylhexanal Ethyl tiglate 1-Octen-3-one 1-Octen-3-ol 2,3-Octanedione 6-Methyl-5-hepten-2-one (E,Z )-2,4-Heptadienal Hexyl acetate 1-Methylbutyl butanoate p-Mentha-3,8-diene 1,3,8-p-Menthatriene Allo-ocimene ()-4-Acetyl-1-methylcyclohexene (Z )-p-Mentha-2,8-dien-1-ol Octanoic acid Thymol methyl ether p-Menth-1-en-9-al p-Menth-1-en-9-al isomer Nonanoic acid (E)-2-decenal Bornyl acetate Methyl geranate exo-2-Hydroxycineole acetate Ylangene -Elemene Dodecanal (S)-Perillyl acetate -Elemene -Farnesene -Cubebene -Ionone epoxide -Farnesene 2,6-Bis(1,1-dimethylethyl)-phenol -Selinene -Muurolene -Selinene Cadinene ()- -Panasinsen Dihydroactinidiolide Selina-3,7(11)-diene Juniper camphor m/z (%) 44(100), 43(21), 42(7), 41(3), 45(2) 43(100), 58(38), 42(8), 41(2), 57(1) 59 (100), 41(15), 43(12), 57(10), 42(7), 73(5), 60 (3), 55(2), 44(2), 58(2) 62(100), 47(79), 45(43), 61(32), 46(29), 49(6), 57(5), 59(4), 44(4), 63(4) 43(100), 74(28), 42(13), 59 (12), 44(8), 73(5), 41(3), 45(2), 75(2) 43(100), 72(38), 44(21), 57(12), 41(10), 42(10), 45(7), 70(5), 55(3), 69(2) 70(100), 41(66), 69(47), 42(14), 71(7), 50(3), 53(3), 52(2), 68(2), 43(2) 43(100), 86(29), 71(13), 41(11), 58(10), 55(10), 42(5), 44(4), 77(3), 84(2) 57(100), 86(28), 56(5), 42(3), 58(3), 77(2), 87(2), 53(2), 44(2), 41(2) 69(100), 99(41), 41(20), 68(7), 86(7), 70(6), 71(5), 43(5), 45(4), 42(3) 71(100), 43(77), 114(26), 41(18), 58(7), 42(6), 72(4), 55(4), 99(3), 44(3) 112(100), 55(61), 41(53), 97(42), 83(41), 43(30), 69(30), 79(18), 40(17), 53(15) 70(100), 41(60), 57 (47),55(47), 71(29), 43(21), 58(18), 67(16), 42(16), 85(12) 83(100), 55(86), 113(43), 100(39), 128(22), 43(18), 82(18), 85(17), 44(13), 53(13) 70(100), 55(94), 97(26), 43(19), 41(13), 83(13), 71(12), 42(7), 56(5), 99(4) 57(100), 72(19), 43(14), 68(14), 41(13), 67(13), 81(10), 85(8), 121(8), 79(7) 43(100), 99(52), 71(34), 41(19), 55(9), 42(7), 44(6), 142(4), 69)3), 87(3) 43(100), 108(62), 41(52), 69(49), 55(46), 111(27), 67(24), 58(21), 71(21), 93(20) 81(100), 41(28), 110(23), 53(21), 67(18), 79(17), 57(16), 43(16), 55(15), 68(12) 43(100), 56(58), 55(39), 41(37), 84(36), 69(31), 61(26), 42(22), 57(22), 44(17) 71(100), 43(44), 70(24), 41(18), 89(15), 93(14), 55(13), 115(12), 88(10), 42(9) 79(100), 107(48), 93(39), 91(34), 136(33), 121(29), 67(27), 77(26), 108(19), 105(17) 119(100), 134(94), 91(85), 105(31), 77(25), 92(20), 117(18), 79(16), 41(16), 57(13) 121(100), 105(42), 136(40), 91(33), 79(31), 77(18), 93(16), 106(9), 122(9), 40(8) 95(100), 138(99), 43(72), 67(69), 123(57), 79(42), 93(41), 77(29), 55(28), 68(27) 134(100), 109(65), 91(61), 119(43), 137(41), 79(39), 95(28), 93(27), 105(24), 69(24) 60(100), 73(83), 43(68), 41(65), 91(59), 55(58), 71(53), 117(53), 93(49), 68(45) 149(100), 164(27), 91(16), 119(11), 150(11), 117(8), 115(7), 134(7), 77(6), 105(5) 94(100), 79(58), 67(17), 95(14), 55(11), 77(10), 68(9), 93(9), 91(9), 81(8) 94(100), 79(57), 95(13), 93(11), 77(9), 41(8), 55(7), 53(6), 67(5), 145(5) 60(100), 73(98), 57(58), 55(43), 41(43), 115(37), 43(37), 129(29), 69(26), 98(19) 70(100), 55(85), 41(73), 43(72), 83(70), 57(57), 69(45), 81(34), 56(34), 98(30) 95(100), 121(55), 136(34), 43(30), 93(18), 81(18), 108(17), 83(15), 96(14), 154(13) 69(100), 41(45), 114(35), 123(30), 83(17), 82(15), 68(13), 70(10), 122(9), 151(9) 43(100), 108(97), 126(53), 109(38), 82(35), 71(35), 93(32), 111(32), 69(31), 83(26) 105(100), 119(98), 93(80), 120(73), 161(72), 91(53), 44(39), 92(37), 41(36), 121(35) 93(100), 81(91), 107(65), 67(61), 68(60), 79(55), 147(50), 121(48), 91(44), 105(44) 57(100), 82(77), 43(77), 41(74), 55(74), 68(59), 67(51), 69(50), 96(46), 81(43) 91(100), 119(70), 43(63), 92(60), 68(50), 134(48), 93(46), 67(39), 79(39), 105(36) 121(100), 93(70), 43(62), 107(55), 41(43), 69(43), 105(42), 161(41), 91(40), 119(33) 69(100), 43(69), 93(61), 41(57), 67(32), 91(32), 44(32), 133(30), 79(29), 55(25) 161(100), 105(37), 91(34), 119(24), 93(21), 120(20), 79(19), 41(19), 81(19), 162(18) 123(100), 43(33), 44(25), 177(15), 135(13), 41(13), 207(12), 55(10), 95(9), 124(9) 93(100), 44(65), 41(58), 69(52), 107(45), 79(42), 119(40), 55(40), 91(40), 43(40) 191(100), 206(15), 192(14), 57(14), 44(10), 207(8), 41(7), 163(5), 73(5), 74(5) 189(100), 133(61), 105(51), 91(44), 204(41), 107(30), 93(29), 147(29), 161(26), 79(24) 105(100), 161(59), 93(46), 91(44), 44(35), 119(34), 94(33), 204(32), 107(26), 81(22) 189(100), 93(64), 107(62), 133(58), 204(56), 91(54), 105(50), 161(45), 81(44), 79(42) 161(100), 119(56), 204(54), 134(53), 105(48), 91(32), 162(21), 81(19), 189(19), 93(17) 161(100), 122(79), 107(51), 105(24), 91(23), 93(20), 204(19), 81(18), 79(17), 119(14) 111(100), 44(80), 43(50), 109(46), 157(45), 137(43), 67(35), 41(28), 180(25), 142(23) 161(100), 122(55), 107(54), 204(54), 105(41), 91(40), 93(38), 44(28), 121(28), 119(27) 204(100), 189(99), 81(84), 161(79), 43(78), 71(60), 95(57), 109(56), 93(55), 67(54)

454
wileyonlinelibrary.com/jsfa
Published 2010 by John Wiley & Sons, Ltd. J Sci Food Agric 2011; 91: 449460

Evaluation of tangerine aroma volatiles

www.soci.org

Table 5. Tangerine aroma volatiles among 11 chemical classes Monoterpenes -Thujene -Pinene Sabinene -Myrcene -Pinene -Phellandrene 3-Carene -Terpinene p-Cymene d-Limonene -Phellandrene (E)-Ocimene -Terpinene p-Mentha-3,8-diene Terpinolene Dehydro-p-cymene 1,3,8-p-Menthatriene Allo-ocimene 4 Unknown monoterpenes Sesquiterpenes -Elemene -Cubebene Ylangene Copaene -Elemene -Elemene Caryophyllene -Farnesene -Guaiene -Cubebene -Caryophyllene -Farnesene -Selinene -Muurolene Valencene -Selinene Cadinene Calamenene ()- -Panasinsen -Calacorene Selina-3,7(11)-diene 18 Unknown sesquiterpenes Aldehydes Acetaldehyde Propanal Butanal (E)-2-Butenal Pentanal (E)-2-Pentenal (Z )-3-Hexenal Hexanal (E)-2-Hexenal 2-Methyl-2-hexenal 4-Methyl-hexanal Heptanal (E,E)-2,4-Hexadienal (E)-2-Heptenal Benzaldehyde (E,Z )-2,4-Heptadienal Octanal (E,E)-(2,4)-Heptadienal (E)-2-Octenal Nonanal (E)-2-Nonenal Decanal p-Menth-1-en-9-al p-Menth-1-en-9-al Isomer -Cyclocitral Neral (E)-2-Decenal Geranial Perillaldehyde Undecanal (E,E)-2,4-Decadienal 2-Undecenal Dodecanal Acids Hexanoic acid Octanoic acid Esters Methyl acetate Ethyl acetate Ethyl propanoate Methyl butanoate Ethyl butanoate Ethyl 2-butenoate Ethyl 2-methylbutanoate Methyl hexanoate Ethyl tiglate Ethyl hexanoate Hexyl ethanoate 1-Methyl butyl butanoate Methyl octanoate Ethyl 3-hydroxyhexanoate Hexyl butanoate Ethyl octanoate Octyl acetate Linalyl acetate Bornyl acetate Nonyl acetate Methyl geranate (Z )-Carvyl acetate Citronellyl acetate exo-2-Hydroxycineole acetate Neryl acetate -Terpinyl acetate Geranyl acetate Decyl acetate (S)-Perillyl acetate Alcohols Ethanol 2-Methyl-2-propanol 1-Penten-3-ol 3-Methylbutanol 1-Octen-3-ol Linalool cis-p-Mentha-2,8-dien-1-ol -Terpineol Terpinene-4-ol -Terpineol (Z )-Carveol Citronellol (E)-Carveol Nerolidol Juniper camphor

Ketones Acetone 2-Butanone

Phenols Thymol 2,6-Bis(1,1dimethylethyl)phenol

Ethers 1,8-Cineole Thymol methyl ether

Epoxides -Ionone epoxide Caryophyllene oxide

Others Dimethyl sulde Styrene ()-4-Acetyl-1-methylcyclohexene 35 Unknowns

1-Penten-3-one 2-Pentanone 2,3-Pentanedione 3-Pentanone 4-Heptanone 1-Octen-3-one 2,3-Octanedione 6-Methyl-5-hepten-2-one Camphor Dihydrocarvone d-Carvone -Ionone Geranyl acetone -Ionone Dihydroactinidiolide Nootkatone

Nonanoic acid

455
wileyonlinelibrary.com/jsfa

J Sci Food Agric 2011; 91: 449460

Published 2010 by John Wiley & Sons, Ltd.

www.soci.org

T Miyazaki et al.

Figure 2. PCA using volatile relative peak areas (peak area/peak area of IS) among all 25 samples. The percent variance explained by each of three factors is shown in parentheses: (A) Factors 1 and 2; (B) Factors 1 and 3. Letters refer to sample codes (Table 1).

samples s (Murcott) and v (Fortune Murcott) had relatively high negative scores on F2. In general, this graph shows that the samples on the right side, q (9-4 Blood4x), s (Murcott), t (Temple), u (Sanguinelli), v (Fortune Murcott), w (Ortanique) and y (89 Val4x) showed high total content of volatiles except for ethers and phenols. Indeed, sample t (Temple) produced 3.1, 6.9 and 5.3 times more sesquiterpenes, esters and ketones than the averages of all samples, respectively (Table 6). Sample u also contained high amounts of sesquiterpenes, esters and alcohols. On the left side of Fig. 3, the samples with more mandarin/tangerine genetic components had fewer volatiles overall. Cluster analysis based on qualitative volatile composition The main goal of CA is to assign individual samples into groups based on their volatile quality or quantity. In the present study, CA is performed using volatile presence/absence; a dendrogram provides information on volatile qualitative differences among samples as well as relationships of their volatile composition based on genetic background (Fig. 4). Cluster 1 (C1) C1 grouped samples a (89 Murcott) and b (Robinson Fairchild) of which each contained fewer volatiles than the

average of all samples. These samples were characterized by generally having fewer aldehydes (except sample b), esters and ketones (Table 7). The aliphatic aldehydes, pentanal, (E )-2-hexenal, heptanal and (E )-2-heptenal were not detected in sample a (89 Murcott), but were in all other samples. Sample b (Robinson Fairchild) was the only sample without any esters; esters usually impart a fruity note to orange juice.38 On the other hand, these samples were the richest in monoterpenes among the ve clusters both in quality (Table 7) and, for sample a, quantity (Table 6). Above all, the highest amounts of -myrcene, d-limonene and -terpinene contribute to the distinctive volatile prole of sample a (89 Murcott) (data not shown). Since samples a (89 Murcott) and b (Robinson Fairchild) are grouped in the same cluster as determined by their volatile prole, it is hypothesized that one or more of their common ancestors, Clementine, Duncan grapefruit and Dancy tangerine, determines higher production of monoterpenes to the detriment of other classes of volatiles. Cluster 2 (C2) C2 grouped samples c (89 Murcott) and l (89), and had fewer volatiles than the average of the samples, especially fewer sesquiterpenes and aldehydes, like cluster 1 (Table 7). In

456

wileyonlinelibrary.com/jsfa

Published 2010 by John Wiley & Sons, Ltd.

J Sci Food Agric 2011; 91: 449460

Evaluation of tangerine aroma volatiles

www.soci.org

Table 6. Amount (sum of relative peak areas (peak area/IS peak area) in each chemical group) of aroma volatiles arranged by 11 chemical classes in 25 samples Sample name Sample code Monoterpenes Sesquiterpenes Aldehydes Esters Alcohols Ketones Phenols Ethers Acids Epoxides Others Total 8-9 M R FC 8-9 M 8-9 M R FC FG R FC R FC FG FC FG FC FG FC 8-9 8-10 8-9 O Unknown Unknown 9-4 Blood4x Unknown M T SANG FM ORT 8-8 M 8-9 Val4x a b c d e f g h i j k l m n o p q r s t u v w x y 27.58 16.05 9.04 13.74 10.36 3.69 6.22 10.61 17.01 7.76 4.88 10.03 5.88 8.55 5.64 1.75 7.62 1.8 11.78 19.44 19.05 17.63 8.35 4.95 6.12 0.09 0.18 0.08 0.05 0.03 0.46 0.66 0.05 1.73 0.04 0.67 0.11 2.45 0.32 0.13 0 3.99 0 0.03 3.3 5.56 0.03 2.13 0.03 4.31 0.61 0.42 0.41 0.5 0.83 0.78 0.8 0.96 0.46 0.59 0.71 0.32 0.72 0.43 0.98 0.93 1.14 1.11 1.57 2.03 0.99 1.07 0.65 0.73 0.58 0.01 0 0.01 0.03 0.06 0.03 0.03 0.05 0.24 0 0.01 0.05 0.02 0.23 0.01 0.03 0.02 0.03 0.07 1.32 1.11 0.35 0.44 0.04 0.59 0.8 0.87 0.48 0.37 0.36 0.27 0.47 0.74 0.73 0.66 0.46 0.63 0.76 0.42 0.31 0.1 0.49 0.11 0.65 3.43 1.35 0.72 0.27 0.18 0.47 0.02 0.2 0.12 0.18 0.25 0.13 0.16 0.36 0.19 0.19 0.14 0.04 0.07 0.09 0.22 0.14 0.35 0.18 0.48 0.48 0.63 0.35 0.37 0.1 0.24 0.002 0 0.015 0 0 0 0 0 0 0 0 0.064 0.043 0 0 0 0 0 0 0 0 0 0 0.03 0.022 0 0 0.198 0.13 0.114 0 0.08 0.17 0 0 0.105 0.003 0.024 0.157 0.068 0 0 0 0 0 0 0 0 0 0.044 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.004 0 0.003 0 0 0 0 0.003 0 0.003 0.051 0 0 0 0 0 0.005 0.004 0.003 0 0.005 0.006 0 0 0 0 0.003 0 0.005 0.003 0 0 0 0 0.002 0 0.01 0.03 0.02 0.09 0.11 0.06 0.06 0.13 0.11 0.05 0.05 0.15 0.1 0.03 0.06 0.03 0.34 0.04 0.12 0.71 0.35 0.11 0.15 0.05 0.33 29.13 17.76 10.39 15.09 12.12 5.43 8.49 13.08 20.46 9.3 7.03 11.4 10.05 10.22 7.43 2.98 13.94 3.27 14.69 30.7 29.03 20.26 12.36 6.1 12.76

M, Murcott; R, Robinson; FC, Fairchild; FG, Fallglo; F, Fortune; O, Orlando; T, Temple; SANG, Sanguinelli; ORT, Ortanique.

looking at the relationship between volatile composition and parentage, sample c (89 Murcott) as well as sample a (89 Murcott) in C1 was more similar to sample l (89) than s (Murcott) in terms of qualitative volatile composition, indicating that the seed parent might be a dominant parent for its aroma volatile production. However, it is not a rule since sample d (89 Murcott) is in the same cluster (C3) as its parent Murcott (sample s). It has been observed that some volatiles in parents are not synthesized in progenies and, conversely, some new volatiles are synthesized in progenies.13,14 In the current study, 1-octen-3-one was found in both parents (sample l, 89; sample s, Murcott), whereas it was absent in the progenies, samples c, a and d (all are 89 Murcott) (data not shown). This volatile is the most intense odor-active aliphatic ketone, with a mushroom-like odor, that was reported in orange juice.5 The content of 1-octen-3-one in sample s (Murcott) was the highest and more than twice the average of all samples. Thus its production was likely suppressed in the progenies. Moreover, carotenoid-derived volatiles were not found in sample a, c or d (all are 89 Murcott). Both parents (sample l, 89) contained 6-methyl-5-hepten-2-one and geranyl acetone, originating from lycopene and -carotene, respectively.39 -Cyclocitral, -ionone and dihydroactinidiolide oxidative degradation products from -carotene39 were only present in the pollen parent (sample s, Murcott), as well as neral and geranial (data not shown). -Ionone is well known as an aroma active compound, and its odor is usually described as oral in orange juice.40,41 This suggests that the carotenoid degradation to aroma volatiles is affected in this specic cross.

Some sesquiterpenes ( -cubebene, -caryophyllene, -muurolene, calamenene) were detected in the progenies but not in their parents (sample l, 89; sample s, Murcott). The capacity to produce these compounds might be inherited from the parents in an additive genetic fashion. Cluster 3 (C3) C3 grouped 12 samples equally divided into ones that contain more or fewer volatiles than the average of all samples. The common characteristics of the samples in this cluster are that they contain many aldehydes and few sesquiterpenes and esters in comparison with the other clusters that either have samples with many or few compounds in all three chemical groups (Table 7). Most samples share a common genetic background in this cluster, mostly by having Murcott or Fairchild as a pollen parent. They were samples d (89 Murcott), v (Fortune Murcott) and x (8-8 Murcott), together with their pollen parent Murcott (sample s). Samples j and k are siblings of the same parents: Fallglo and Fairchild. Samples e, g and h are siblings originated from the same cross between Robinson and Fairchild. Cluster 4 (C4) C4 contained samples with a higher than average number of volatiles, except sample n (89 Orlando). Samples in this cluster contain a distinctly higher number of volatiles in the sesquiterpene and ester categories (Table 7). Samples t (Temple) and u (Sanguinelli), the outliers in the PCA analysis, are grouped in this cluster owing to their distinctive volatile composition,

457

J Sci Food Agric 2011; 91: 449460

Published 2010 by John Wiley & Sons, Ltd.

wileyonlinelibrary.com/jsfa

www.soci.org

T Miyazaki et al.

Figure 3. PCA using the sum of relative peak areas of compounds classied into chemical categories among all 25 samples. Letters refer to sample codes (Table 1). Lines represent vectors for each of the variables (sum of compound relative peak areas).

Figure 4. CA using volatile presence and absence among samples. Letters indicate sample codes (Table 1).

as previously mentioned. Sample n (89 Orlando) is also characterized by its richness in esters, and samples i (Fallglo Fairchild), w (Ortanique) and y (89 Val4x) are characterized by their high content of sesquiterpenes and esters. With regard to samples u (Sanguinelli), w (Ortanique) and y (89 Val4x), the sesquiterpene valencene was the second most abundant volatile, although not important for aroma, accounting for 12.5%, 24.8% and 11.9% of relative peak area, respectively (data not shown). Nootkatone, a putative derivative from valencene, was also detected in the three samples and in sample t (Temple). This compound is a major aroma impact compound that contributes to characteristic aroma and avor of grapefruit and pummelo.42,43 These samples are grouped in the same cluster, probably because they have some orange genetic background that results in the production of esters. Kerbiriou et al.13 had found a similar clustering in 20062007. Temple and Ortanique are believed to be tangors (presumed tangerine-sweet orange hybrids) that originated in Jamaica.19 Sanguinelli is a blood orange and Val4x is a tetraploid Valencia orange. Sample i (Fallglo Fairchild) does

not have a direct orange parent; however, it does have orange in its background with Temple as a grandparent.

Cluster 5 (C5) C5 grouped samples that are outliers based on their volatile composition. Samples p and r are themselves in a sub-cluster (Fig. 4). The parentage of samples p and r is unknown, though they may have originated from the same parents owing to their very similar volatile composition. The marked characteristics of these samples are signicantly lower amounts of total volatiles and an absence of sesquiterpenes. In addition, aldehydes are numerous and in large amount in samples p and r, accounting for 31.0% and 33.9% of relative peak area, respectively (Tables 6 and 7). 2-Methyl-2-hexenal, (E ,Z )-2,4-heptadienal and (E ,E )-2,4heptadienal were detected only in the three samples in this cluster (data not shown). Most aldehydes present in these samples have already been found in many citrus fruits,5,13,17,43 45 and they are well known to impart green and fatty notes. Therefore, the

458

wileyonlinelibrary.com/jsfa

Published 2010 by John Wiley & Sons, Ltd.

J Sci Food Agric 2011; 91: 449460

Evaluation of tangerine aroma volatiles

www.soci.org

Table 7. Samples in ve clusters formed by cluster analysis based on presence and absence of volatiles, and their number of volatiles in 11 chemical classes Cluster no. Sample code Sample name Monoterpene Sesquiterpene Aldehyde Ester Alcohol Ketone Phenol Ether Acid Epoxide Other Total C1 C2 C3 a b c l f m d e k g o h j x s v n i y t u w q p r 8-9 M R FC 8-9 M 8-9 FG 8-10 8-9 M R FC FG FC R FC Unknown R FC FG FC 8-8 M M FM 8-9 O FG FC 8-9 Val4x T SANG ORT 9-4 Blood4x Unknown Unknown 18 19 17 16 13 18 13 14 14 14 14 15 16 14 15 15 14 16 13 14 17 14 14 10 8 10 11 7 8 8 17 8 5 11 14 12 8 8 6 4 5 13 20 19 19 23 20 33 0 0 9 17 14 14 16 21 16 17 21 23 24 26 22 20 26 24 16 19 20 24 18 19 28 26 25 2 0 2 2 2 3 2 2 1 5 2 3 1 2 5 9 10 11 9 17 20 10 4 2 2 5 9 6 6 3 6 4 5 5 6 5 6 5 5 7 7 4 5 6 7 8 6 7 5 4 4 5 6 8 8 7 4 8 8 9 9 8 9 11 12 10 6 8 11 12 11 12 9 9 9 1 0 2 2 0 1 0 0 0 0 0 0 0 2 0 0 0 0 1 0 0 0 0 0 0 0 0 2 1 0 1 1 1 1 1 1 1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 1 0 0 2 0 0 0 1 0 0 0 0 0 0 1 0 0 0 2 1 0 1 1 1 1 0 0 0 1 1 1 1 0 1 1 3 5 6 8 7 15 10 9 9 12 12 14 8 6 12 9 5 11 10 15 13 13 22 8 7 52 66 62 65 58 89 58 61 72 85 80 82 70 68 82 80 69 90 93 109 111 95 118 61 56

C4

C5

M, Murcott; R, Robinson; FC, Fairchild; FG, Fallglo; F, Fortune; O, Orlando; T, Temple; SANG, Sanguinelli; ORT, Ortanique.

high level of aldehyde content might negatively inuence the overall aroma of samples p and r, as well as q (9-4 Blood4x). As shown in Fig. 2 as well as Tables 6 and 7, sample q (9-4 Blood4x) is clearly characterized by a high number and quantity of sesquiterpenes. This sample contained 33 compounds of the 39 different sesquiterpenes detected in the 25 samples. Sample q (9-4 Blood4x) and sample y (89 Val4x) in C4 have twothirds chromosome complement from Blood4x (tetraploid Ruby blood orange) or Val4x, respectively. While C4 grouped samples that have some orange in their background and thus abundant esters, this sample is differentiated from them due to its lower amount of esters. Interestingly, although sample q (9-4 Blood4x) contained the highest number of volatiles (118 volatiles) (Table 7), their amount was only slightly higher than the average of the 25 samples and similar to those of the siblings of 9-4, 89 and 810 (all are hybrids from the cross Clementine Minneola) (Table 6). Sample y (89 Val4x) in C4 contained a higher number of volatiles (93) than its parent l (89) (65) (Table 7), whereas the total content is similar to that of the parent (Table 6). From these results, it may be deduced that samples q (9-4 Blood4x) and y (89 Val4x) can synthesize aroma volatiles from the pollen parents, though the total volatile content is regulated by the seed parents. So far, inheritance of genes involved in volatile synthesis and their expression in sexual citrus progenies are still unknown. Gancel et al.46 showed that leaf volatile composition of a somatic hybrid (2n = 4x = 36) between Willow leaf mandarin (C. deliciosa Ten.) with sweet orange was similar to that of the

mandarin parent. Since the parents 9-4 and 89 have background from mandarin, their genetic components might play an important role in the control of aroma production.

CONCLUSION
The qualitative and quantitative differences of volatile composition were observed in a population of tangerine hybrids: among 203 aroma volatiles detected by GC-MS, 92 were identied and conrmed with chemical standards, and 54 were tentatively identied using mass spectra and RIs on two columns. Most hybrids with mandarin/tangerine genetic contribution were distinguished from the other samples by their lower number and/or quantity of volatile compounds. Different siblings of the same parentage (89 Murcott, Robinson Fairchild, Fallglo Fairchild, or 89 and 810) showed different volatile composition, implying complex genetic controls for aroma volatile production. Nevertheless, hybrids with sweet orange in their background appeared to produce more sesquiterpenes and esters, except 9-4 Blood4x, which produced fewer esters, similar to the other tangerines. These volatile differences among hybrids provide fundamental information for improved genetic understanding and future improvement in tangerine aroma and avor.

REFERENCES
1 Gottwald TR, Graham JH and Schubert TS, Citrus canker: the pathogen and its impact. Plant Health Progress (2002). [Online]. Available:

459

J Sci Food Agric 2011; 91: 449460

Published 2010 by John Wiley & Sons, Ltd.

wileyonlinelibrary.com/jsfa

www.soci.org
http://www.plantmanagementnetwork.org/pub/php/review/ citruscanker [8 October 2010]. US Department of Agriculture, Citrus Fruits 2009 Summary. [Online]. Available: http://usda.mannlib.cornell.edu/usda/current/ CitrFrui/CitrFrui-09-24-2009.pdf [14 April 2010]. US Department of Agriculture, United States Standards for Grades of Florida Tangerines. [Online]. Available: http://www.ams.usda.gov/ AMSv1.0/getle?dDocName=STELPRDC5050394 [1 September 2009]. Gmitter FG, Grosser JW, Castle WS and Moore GA, A comprehensive citrus genetics improvement programme, in Citrus Genetics, Breeding and Biotechnology, ed. by Khan IH. CAB International, Wallingford, UK, pp. 918 (2007). rez-Cacho PR and Rouseff RL, Fresh squeezed orange juice odor: a Pe review. Crit Rev Food Sci Nutr 48:681695 (2008). Coleman RL and Shaw PE, Analysis of tangerine essence oil and aroma oil. J Agric Food Chem 20:12901292 (1972). Moshonas MG and Shaw PE, Quantitative and qualitative analysis of tangerine peel oil. J Agric Food Chem 22:282284 (1974). Buettner A, Mestres M, Fischer A, Guash J and Schieberle P, Evaluation of the most odour-active compounds in the peel oil of Clementines (Citrus reticulata Blanco cv. Clementine). Eur Food Res Technol 216:1114 (2003). Chisholm MG, Jell JA and Cass DM Jr, Characterization of the major odorants found in the peel oil of Citrus reticulata Blanco cv. Clementine using gas chromatography olfactometry. Flavour Fragr J 18:275281 (2003). Sawamura M, Minh Tu NT, Onishi Y, Ogawa E and Choi HS, Characteristic odor components of Citrus reticulata Blanco (Ponkan) cold-pressed oil. Biosci Biotechnol Biochem 68:16901697 (2004). Elmaci Y and Altug T, Flavor characterization of three mandarin cultivars (Satsuma, Bodrum, Clementine) by using GC/MS and avor prole analysis techniques. J Food Qual 28:163170 (2005). rez-Lopez Pe AJ and Carbonell-Barrachina AA, Volatile odour components and sensory quality of fresh and processed mandarin juices. J Sci Food Agric 86:24042411 (2006). Kerbiriou P, Plotto A, Goodner K, Baldwin E and Gmitter FG Jr, Distribution of aroma volatiles in a population of tangerine hybrids. Proc Fla State Hortic Soc 120:267275 (2007). Barboni T, Luro F, Chiaramonti N, Desjobert JM, Muselli A and Costa J, Volatile composition of hybrids Citrus juices by headspace solidphase micro extraction/gas chromatography/mass spectrometry. Food Chem 116:382390 (2009). Tietel Z, Bar E, Lewinsohn E, Feldmesser E, Fallik E and Porat R, Effects of wax coatings and postharvest storage on sensory quality and aroma volatile composition of Mor mandarins. J Sci Food Agric 90:9951007 (2010). Lan-Phi NT, Shimamura T, Ukeda H and Sawamura M, Chemical and aroma proles of Yuzu (Citrus junos) peel oils of different cultivars. Food Chem 115:10421047 (2009). Allegrone G, Belliardo F and Cabella P, Comparison of volatile concentrations in hand-squeezed juices of four different lemon varieties. J Agric Food Chem 54:18441848 (2006). Shaw PE, Goodner KL, Moshonas MG and Hearn J, Comparison of grapefruit hybrid fruit with parent fruit based on composition of volatile components. Sci Hortic 91:7180 (2001). Hodgson RW, Horticultural varieties of citrus, in The Citrus Industry, Vol. 1, ed. by Reuther W, Webber HJ and Batchelor LD. University of California Press, Berkeley, CA, pp. 431591 (1967). Wutscher HK, Maxwell NP and Shull AV, Bower, a new mandarin variety for Texas. J Rio Grande Valley Hortic Soc 27:4043 (1973). Futch SH and Jackson LK, Murcott (Honey Tangerine). Fact sheet HS174. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Futch SH and Jackson LK, Orlando Tangelo. Fact sheet HS175. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Jackson LK and Futch SH, Dancy Tangerine. Fact sheet HS169. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Jackson LK and Futch SH, Fallglo Tangerine. Fact sheet HS173. Horticultural Sciences Department, Florida Cooperative Extension

T Miyazaki et al.
Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Jackson LK and Futch SH, Minneola Tangelo. Fact sheet HS171. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Jackson LK and Futch SH, Robinson Tangerine. Fact sheet HS178. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Jackson LK and Futch SH, Temple Tangor. Fact sheet HS181. Horticultural Sciences Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida (2003). Goodner KL, Rouseff RL and Hofsommer HJ, Practical retention index models of OV-101, DB-1, DB-5, and DB-Wax for avor and fragrance compounds. LWT Food Sci Technol 41:951958 (2008). Kondjoyan N and Berdagu e JL, A Compilation of Relative Retention Indices for the Analysis of Aromatic Compounds (1st edn). Laboratoire Flaveur, INRA de THEIX, Saint Genes Champanelle, France (1996). Adams RP, Identication of Essential Oil Components by Gas Chromatography/Mass Spectrometry (4th edn). Allured Publishing, Carol Stream, IL (2007). Johnson RA and Wichern DW, Applied Multivariate Statistical Analysis (3rd edn). Prentice Hall, Englewood Cliffs, NJ (1992). Moshonas MG and Shaw PE, Analysis of volatile constituents from tangerine essence. J Agric Food Chem 20:7071 (1972). Moshonas MG and Shaw PE, Quantitation of volatile constituents in mandarin juices and its use for comparison with orange juices by multivariate analysis. J Agric Food Chem 45:39683972 (1997). Minh Tu NT, Thanh LX, Une A, Ukeda H and Sawamura M, Volatile constituents of Vietnamese pummelo, orange, tangerine and lime peel oils. Flavour Fragr J 17:169174 (2002). Njoroge SM, Koaze H, Mwaniki M, Minh Tu NT and Sawamura M, Essential oils of Kenyan citrus fruits: volatile components of two varieties of mandarins (Citrus reticulata) and a tangelo (C. paradisi C. tangerina). Flavour Fragr J 20:7479 (2005). rez-Lopez Pe AJ, Saura D, Lorente J and Carbonell-Barrachina AA, Limonene, linalool, -terpineol and terpinen-4-ol as quality control parameters in mandarin juice processing. Eur Food Res Technol 222:281285 (2006). Dharmawan J, Kasapis S, Curran P and Johnson JR, Characterization of volatile compounds in selected citrus fruits from Asia. Part I: Freshly-squeezed juice. Flavour Fragr J 22:228232 (2007). Plotto A, Margar a CA, Goodner KL and Baldwin EA, Odour and avour thresholds for key aroma components in an orange juice matrix: esters and miscellaneous compounds. Flavour Fragr J 23:398406 (2008). Lewinsohn E, Sitrit Y, Bar E, Azulay Y, Ibdah M, Meir A, et al, Not just colors: carotenoid degradation as a link between pigmentation and aroma in tomato and watermelon fruit. Trends Food Sci Technol 16:407415 (2005). Hinterholzer A and Schieberle P, Identication of the most odour active volatiles in fresh, hand-extracted juice of Valencia late oranges by odour dilution techniques. Flavour Fragr J 13:4955 (1998). Mahattanatawee K, Rouseff R, Valim MF and Naim M, Identication and aroma impact of norisoprenoids in orange juice. J Agric Food Chem 53:393397 (2005). Macleod WD Jr and Buigues NM, Sesquiterpenes. I. Nootkatone, a new grapefruit avor constituent. J Food Sci 29:565568 (1964). Sawamura M and Kuriyama T, Quantitative determination of volatile constituents in the pummelo (Citrus grandis Osbeck forma Tosabuntan). J Agric Food Chem 36:567569 (1988). Shaw PE, Fruits II in Volatile Compounds in Foods and Beverages, ed. by Maarse H. Marcel Dekker, New York, pp. 305327 (1991). Shaw PE, Moshonas MG, Hearn CJ and Goodner KL, Volatile constituents in fresh and processed juices from grapefruit and new grapefruit hybrids. J Agric Food Chem 48:24252429 (2000). Gancel AL, Ollitrault P, Froelicher Y, Tomi F, Jacquemond C, Luro F, et al, Citrus somatic allotetraploid hybrids exhibit a differential reduction of leaf sesquiterpenoid biosynthesis compared with their parents. Flavour Fragr J 20:626632 (2005).

2 3

25

26

27

5 6 7 8

28 29 30 31 32 33 34 35

10 11 12 13 14

36

37 38

15

16 17 18 19 20 21

39

40

41 42 43 44 45 46

22

23

24

460

wileyonlinelibrary.com/jsfa

Published 2010 by John Wiley & Sons, Ltd.

J Sci Food Agric 2011; 91: 449460