OUP user menu

Indole derivatives produced by the fungus Colletotrichum acutatum causing lime anthracnose and postbloom fruit drop of citrus

Kuang-Ren Chung , Turksen Shilts , Ümran Ertürk , L.W. Timmer , Peter P. Ueng
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00605-0 23-30 First published online: 1 September 2003


Postbloom fruit drop (PFD) of citrus and Key lime anthracnose (KLA) are caused by Colletotrichum acutatum. Both fungal isolates can infect flower petals, induce young fruit abscission and result in severe yield loss on many citrus cultivars. Previous studies revealed that infection of citrus flowers by C. acutatum caused higher levels of indole-3-acetic acid (IAA), which could be synthesized from the host plant and/or the fungal pathogen. The ability for IAA production by C. acutatum isolates was investigated. Similar to many microorganisms, the production of indole compounds in the medium by C. acutatum was dependent solely on the presence of tryptophan (Trp). In total, 14 PFD and KLA fungal isolates were tested, and revealed that they all were capable of utilizing Trp as a precursor to synthesize IAA and other indole derivatives. High-performance liquid chromatography analysis and chromogenic stains after a fluorescence thin-layer chromatography separation unambiguously identified IAA, tryptophol (TOL), indole-acetaldehyde, indole-acetamide (IAM), indole-pyruvic acid, and indole-lactic acid (ILA) from cultures supplemented with Trp. The data suggest that C. acutatum may synthesize IAA using various pathways. Interestingly, increasing Trp concentrations drastically increased the levels of TOL and ILA, but not IAA and IAM. The ability of C. acutatum to produce IAA and related indole compounds may in part contribute to the increased IAA levels in citrus flowers after infection.

  • Indole-3-acetic acid
  • Plant pathogen
  • Sweet orange
  • Tryptophan

1 Introduction

In addition to higher plants, numerous bacteria and fungi also have the ability to synthesize plant growth regulators such as indole-3-acetic acid (IAA) and other indole-related compounds [13]. IAA has been demonstrated as a pathogenicity factor and also a cause of gall formation caused by Pseudomonas syringae pv. savastanoi and Agrobacterium tumefaciens, and increased root induction caused by A. rhizogenes [4]. IAA has also been implicated to be responsible for smut gall formation in the Ustilago maydis–maize association [5], and for the formation of witches' bloom and leaf curl caused by Taphrina spp. [6]. Despite the likelihood that many symptoms of plant diseases caused by fungi are due to the disturbance of endogenous IAA, the genetic determinant of IAA in the fungal–host interactions has not been exclusively investigated at molecular and genetic levels. The only gene so far isolated from fungi with a role in IAA biosynthesis is iad1, encoding a putative indole-3-acetaldehyde dehydrogenase, isolated from U. maydis [7].

Postbloom fruit drop (PFD) of sweet orange and Key lime anthracnose (KLA) are caused by the fungus Colletotrichum acutatum [8]. PFD fungal isolates infect flower petals of most citrus cultivars, causing blossom blight, but exhibit low pathogenicity on leaves and fruit of Mexican lime [9]. In contrast, KLA fungal isolates infect all parts of Mexican lime, causing typical anthracnose symptoms. KLA fungal isolates also infect the flower petals of sweet orange, resulting in similar blossom blight, and can cause all symptoms of PFD on sweet orange.

PFD is a common disease in the humid citrus-growing areas of the Americas, and causes devastating yield loss, due to induction of young fruit drop, in some tropical, high-rainfall areas in Mexico, Belize, Costa Rica, and in the Caribbean [10]. The occurrence of PFD in Florida and Brazil is more endemic, and is highly dependent on the rainfall during the blooming periods. Distinguishing characteristics of PFD include induction of premature fruit drop and formation of persistent calyces. The persistent calyces, commonly called ‘buttons’, develop in the affected flower clusters and remain attached on peduncles of the trees all year long, providing a characteristic symptom for diagnosis. The healthy flowers and young fruit that are adjacent to affected flowers are also inclined to fall before setting fruit, and form persistent calyces. The surrounding leaves become twisted with swollen veins. Those observations suggested that phytohormones might be involved in the development of these symptoms [8].

Little is known on how C. acutatum causes fruit drop of citrus. Previous studies using gene expression profiles revealed that the genes specific for auxin regulation were highly up-regulated after fungal invasion [11,12]. Further analyses of the hormone contents also indicated that IAA increased significantly in citrus flowers infected with C. acutatum. As compared to a water control, the amount of IAA in the infected petals increased up to 140-fold [12]. Furthermore, preliminary studies in the screenhouse indicated that applying clofibrate (a putative auxin inhibitor) after C. acutatum infection resulted in increased fruit retention (K.R. Chung, unpublished data). These results suggest that the imbalance of IAA in C. acutatum-infected flowers may be involved in young fruit drop. Higher levels of IAA in the infected tissues can result from increased biosynthesis by the host and/or by the pathogen.

To determine the role of IAA in fruit drop associated with C. acutatum, the ability of a pathogen to produce IAA or related indole compounds was first evaluated in vitro. In this study we report the identification and quantification of IAA and other indole derivatives from the cultures of 14 different C. acutatum isolates using high-performance liquid chromatography (HPLC) as well as color reactions followed by fluorescence thin-layer chromatography (TLC) and staining with chromogenic reagents that specifically react with indole compounds. The result indicates that higher levels of IAA in symptomatic and asymptomatic tissues after fungal infection may be partially due to production by C. acutatum. Understanding the ability of C. acutatum to produce IAA and indole derivatives in vitro provides an opportunity to further investigate their roles in the young fruit drop.

2 Materials and methods

2.1 Fungal strains and cultural conditions

Isolates of C. acutatum J.H. Simmonds used in this study were recovered from infected sweet orange (Citrus sinensis Osbeck) petals or Key lime (C. aurantifolia Christm.) leaves (Table 1) as previously described [13]. For indole production, fungal isolates were grown in 50 ml of Czapek solution (pH 6.5) containing 2 g NaNO3, 1 g KH2PO4, 0.5 g MgSO4·7H2O, 0.5 g KCl, 0.01 g FeSO4·7H2O, and 30 g sucrose per liter, for 4 days. Fungal mycelium was harvested using low-speed centrifugation (3000×g) for 10 min, washed twice with sterilized water, and ground in 50 ml water with a sterile blender. 1 ml of the resulting suspension was added to the medium with or without dl-tryptophan. All liquid cultures in 250-ml flasks were incubated at 28°C on a rotary shaker.

View this table:
Table 1

Colletotrichum isolates used in this study and their production of indole compounds and IAA

IsolatesHost plantOriginsIndole (µM)IAA (µM)
KLA207Key limeFlorida, USA101.5±22.03.7
HomesteadKey limeFlorida, USA213.8±14.113.6
MGG-1Key limeBrazil189.7±18.17.3
M3Key limeMexico146.9±32.513.7
Orosi-LKey limeCosta Rica242.3±9.924.2
Orosi-FKey limeCosta Rica245.7±36.121.7
ADS-LKey limeBelize146.0±72.810.5
ADS-FKey limeBelize157.4±36.38.6
KLA-2Key limeDominican Republic140.0±27.314.4
IM-2CSNavel orangeFlorida, USA124.6±38.014.3
NavelNavel orangeFlorida, USA238.0±37.648.1
SM3Valencia orangeFlorida, USA139.0±31.612.3
SRL-FTPValencia orangeFlorida, USA156.8±17.918.1
CS-1SNavel orangeFlorida, USA214.6±9.9nd
  • Fungal isolates were grown in the medium containing 1 mM Trp for 7 days. The supernatant (1 ml) was mixed with Salkowski reagent and quantified at wavelength 540 nm with spectrophotometer. The data are the mean of three different experiments except for isolate Orosi-F, ADS-L, and IM-2CS.

  • The fungal culture (25 ml) was extracted with ethyl acetate and separated in a C18 reversed column using methanol/acetate acid as a mobile phase. The data represent only a single set of samples for each isolate. nd, not determined.

2.2 Detection of IAA and related indole compounds using chromogenic reagents

Salkowski reagent consisting of 27.6 mM FeCl3 and 6.6 M H2SO4[14,15] was used to identify indole derivatives. 1 ml of each fungal culture (25 ml) was mixed with equal volumes of Salkowski reagent, and incubated in the dark for 30 min. The optical density was measured using a spectrophotometer at wavelength 540 nm. Regression line and correlation coefficient (r2>0.998) were determined using authentic IAA (Fluka, Milwaukee, WI, USA) ranging from 2.9×10−5 to 2.6×10−4 M (5–45 µg ml−1) as a standard. Other chromogenic reagents used to stain and verify the IAA and other indole compounds on the TLC plates included Ehrlich reagent (2 g p-dimethylamino benzaldehyde in 20 ml HCl and 80 ml ethanol), 2,4-DNPH reagent (0.5% 2,4-dinitrophenyl hydrazine in 0.2 N HCl), nitrite-nitric reagent (HNO3:acetone:5% NaNO2=5:45:1, v/v), ninhydrin-acetic acid reagent (0.2% ninhydrin in acetone:glacial acetic acid=9:1, v/v) [16], Renz and Loew reagent (1 g dimethyl aminocinnamylaldehyde in 100 ml concentrated HCl) [17], and Prochazka reagent (35% formaldehyde:25% HCl:ethanol=1:1:2, v/v) [18].

2.3 Purification of IAA and related indole compounds produced by C. acutatum

Each of 25-ml cultural solutions at the end of the incubation period was filtered through a filter paper. The pH of cultural filtrate was adjusted to 2.5 using 5 N HCl, and extracted with equal volumes of ethyl acetate. The organic solvent was evaporated, and indole compounds were dissolved in 1 ml of 25% methanol and 1% acetic acid (pH 4.5). The recovery rate for IAA from cultural solution was over 90%.

2.4 Detection and quantification of IAA and related indole compounds using HPLC and TLC

For HPLC analysis, indole compounds were analyzed in a 600E Model HPLC (Waters-Millipore, Milford, MA, USA) equipped with a Nucleosil 120-5 C18 reverse column (5 µm, 250×4 mm) (Richard Scientific, Novato, CA, USA) and using 72% solvent A (1% acetic acid) and 28% solvent B (100% methanol) as a mobile phase. The flow rate was set to 0.8 ml min−1, and the injection volume was 30 µl. The presence of tryptophan (Trp), IAA and indole derivatives was detected using a Model 490E of UV monitor (Waters-Millipore) at 280-nm wavelength, and confirmed by retention time and co-migration (spiking with an authentic standard). The compounds were quantified by reference to the peak area obtained for authentic standards for Trp, IAA, tryptophol (TOL), indole-pyruvic acid (IPA), indole-acetaldehyde (IAAld), indole-acetamide (IAM), indoleacetonitrile (IAN), tryptamine (TNH2), and indole-lactic acid (ILA).

Identification of indoles was also performed using chromogenic reagents after separation on TLC with silica gel 60 F254 fluorescent plates (20×20 in length, EM Science, Gibbstown, NJ, USA). In total, 15 µl of each sample was spotted 5 cm from the edge of TLC plates, and separated in a solvent containing n-hexane:ethyl acetate:isopropanol:acetic acid (40:20:5:1, v/v) for 1 h. After the plates were dry, the indole compounds could be visualized as bands using a short-wavelength (254 nm) hand-held UV lamp, and were recorded using a digital camera. The identities of indole compounds were verified as compared to the authentic standards based on their same mobility (Rf value). Compound identifies were also verified by spraying chromatograms evenly in an upright position with chromogenic reagents followed by heating to 90°C. All indole standards were purchased from Sigma (St. Louis, MO, USA) or Aldrich/Fluke.

3 Results and discussion

3.1 Production of indole compounds by C. acutatum

In this study a colorimetric technique using Salkowski reagent was first employed to assess the ability of C. acutatum isolates to produce IAA in culture. Salkowski reagent recognizes and reacts specifically with indole compounds to form a pink color in solution, which can be further quantified using a spectrophotometer [15,18]. Reaction of Trp alone with Salkowski reagent resulted in a faint yellow color, but the value of OD540 was below the detection limit, and thus was considered not different from zero. To test the importance of Trp as a precursor for indole production, C. acutatum Orosi-L isolate was grown in the medium supplemented with various concentrations of Trp. As shown in Fig. 1, indole compounds were not produced in the absence of Trp. However, in the presence of Trp, indole compounds accumulated to high levels, and the amounts of indole produced by C. acutatum increased with increasing concentrations of Trp.

Figure 1

Production of total indole compounds by the fungus C. acutatum. Fungal isolate Orosi-L was grown in the medium with or without various amounts of Trp for 7 days. Cultural filtrate was mixed with Salkowski reagent, and measured by spectrophotometer at 540 nm. The data shown are the means of three different experiments.

To survey the prevalence of indole production among C. acutatum isolates, 14 KLA and PFD fungal isolates from various countries were tested for indole production using the colorimetric technique (Table 1). The results indicated that all KLA and PFD fungal isolates tested were able to accumulate significant levels of indole compounds as compared to that of the no-Trp control.

3.2 Identification and verification of indole compounds using HPLC

HPLC analysis was conducted to more precisely identify and quantify IAA and other indole derivatives. Several commercial standards were readily separated and detected as a distinct peak by HPLC using a C18 reverse column (Fig. 2A). Under the conditions analyzed, the retention time of IAAld, Trp, TNH2, IAM, ILA, IAA, TOL, IAN, and IPA was 3.6, 6.5, 7.0, 8.8, 14.5, 15.0, 16.1, 22.1, and 31.2 min, respectively.

Figure 2

HPLC chromatograms of the authentic indole compounds (panel A) and the ethyl acetate extract of fungal supernatant of C. acutatum (panel B). The authentic indole compounds (100 µM each, except IPA at 2 mM) were separated using a C18 reverse HPLC column. The retention time of IAAld, Trp, TNH2, IAM, ILA, IAA, TOL, IAN, and IPA as indicated was 3.6, 6.5, 7.0, 8.8, 14.5, 15.0, 16.1, 22.1, and 31.2 min, respectively. The fungal culture grown in the medium amended with Trp was also analyzed by HPLC. Peaks #4, 8, 9, 11, and 12 co-eluted with IAAld, IAM, ILA, IAA, and TOL, respectively. The identities of peaks #1–3, 6, 7, 10, and 13 that occurred only in the presence of Trp remain to be determined.

Extracts from fungal culture filtrates were analyzed by HPLC. Typically, 12–15 peaks, varying between isolates and cultural conditions, were identified using a C18 reverse HPLC column (Fig. 2B). As compared to the authentic standards, peaks #4, 8, 9, 11, and 12 co-eluted with IAAld, IAM, ILA, IAA, and TOL, respectively (Fig. 2B). The identity of each indole compound was confirmed by co-elution with the corresponding reference standard. No corresponding peaks were identified in the extract derived from cultures with no fungus or no Trp (data not shown). However, supplementing the C. acutatum culture with 1 mM Trp resulted in the production of IAM (peak #8) and ILA (peak #9) (Fig. 2B). Small amounts of IAA (peak #11), TOL (peak #12), and IAAld (peak #4) were also identified. The presence of IPA in the C. acutatum culture was identified only in few instances using HPLC (data not shown), likely due to its instability and low concentration. No peaks corresponding to IAN or TNH2 were identified from the fungal cultures. Overall, metabolism of Trp by C. acutatum isolates apparently resulted in several indole derivatives including IAA, IAAld, IAM, ILA, TOL, and IPA. The identities of peaks #1–3, 6, 7, 10, and 13 (Fig. 2B) remain to be determined. Identification of IPA, IAAld, TOL, and IAM suggested that C. acutatum can synthesize IAA through different biosynthetic pathways. The presence of both IPA/IAAld and IAM pathways for IAA biosynthesis has also been reported in Colletotrichum gloeosporioides f. sp. aeschynomene [19].

The levels of IAA or other indoles accumulated in response to amounts of Trp were also quantified by HPLC. Unlike the total indole compounds, no correlation was observed between the levels of IAA and the amounts of Trp added (Fig. 3). The IAA levels increased slightly while the Trp concentration was increased from 1 to 5 mM. However, the production of IAA was inhibited at higher Trp concentrations (≥10 mM). In contrast, increasing Trp drastically elevated the levels of TOL and ILA. TOL has been proposed to serve as a reservoir for IAA and to maintain the levels of IAAld in equilibrium in plants [20]. Both TOL and ILA are also common byproducts of IAA biosynthesis via the IPA/IAAld pathway [21,22]. The production of IAM in the medium showed no direct correlation with the amounts of Trp, and was also inhibited at high dosages of Trp (Fig. 3).

Figure 3

Biosynthesis of IAA, TOL, ILA, and IAM by C. acutatum grown in the medium supplemented with various amounts of Trp. Fungal isolate KLA 207 was grown for 7 days and the filtrate was extracted with ethyl acetate. The resulting indole compounds were analyzed by HPLC using a C18 reverse column. The concentrations of indole compounds were calculated by reference to the peak area obtained for authentic standards using a linear regression curve.

IAA produced by various isolates was also quantified by HPLC, indicating that production of IAA among different isolates was in close agreement with the total indole compounds as assessed using the Salkowski reagent (Table 1). The results clearly confirm that both KLA and PFD fungal isolates were capable of utilizing Trp as a precursor to produce the phytohormone IAA.

3.3 Identification of indole compounds by TLC and chromogenic reagents

Although TLC has relatively lower sensitivity for identification of indole compounds as compared to that of HPLC, combinations of TLC and chromogenic reagents can provide another means for verification of indole derivatives. Many chromogenic reagents with different specificities and sensitivities can recognize and react with indole derivatives to give a distinct color, which can be easily visualized and used to differentiate indole derivatives. Some indoles were confirmed based on their characteristic color changes as plates dried. For example, Ehrlich reagent reacted with ILA to give purple color, which gradually turned to gray color. In this study we have taken the advantage of chromogenic stains to further confirm the results obtained from HPLC analysis.

By using fluorescent TLC plates, several commercial indoles were readily separated and could be visualized as distinct bands under short-wavelength UV light (Fig. 4A). IAN and IAAld standards had a similar mobility and were separated in different lanes. Trp and TNH2 were not separated under the solvent conditions used, whereas separation of ILA resulted in a smeared, faint band (Fig. 4A, lane 4). IPA with Rf 0.48 was easily resolved into bands corresponding to IAA, IAAld, and other unknown compounds (Fig. 4A, lane 3), indicative of instability or impurities in commercial IPA.

Figure 4

Identification of indole compounds produced by C. acutatum using a fluorescence TLC plate (panel A) and chromogenic reagents (panels B–D). Ethyl acetate-extracted samples from fungal culture supplemented with 10 mM Trp (lanes 5 and 6; two identical samples were loaded on the plates), or without Trp (lane 7), or a medium plus Trp but with no fungal isolate as a control (lane 8), were loaded onto silica gel fluorescent plates and developed in a solution containing n-hexane, ethyl acetate, isopropanol, and acetic acid. After development, the plate was visualized directly using a short-wavelength UV light (panel A), or stained with Salkowski (panel B), Ehrlich (panel C), and 2,4-DNPH (panel D) reagents. The authentic indole compounds with 5 mM each (lanes 1–4) were loaded onto the plates for reference. Their mobility (Rf) and related positions are indicated at the left. The bands were compounds derived from fungal cultures using Trp as a precursor. The upward arrow at the right is the direction of chromatographic development.

Analysis of fungal cultures identified at least 11 different bands (Fig. 4A and Table 2). No such bands were identified in the absence of fungal inoculum (Fig. 4A, lane 7) or in the absence of Trp (Fig. 4A, lane 8). Several bands (#1, and #7–9) were either too faint or smeared to be distinctly visualized under UV light. Based on their Rf values, bands #2, 4, 6–8, 11 were likely IAA, TOL, IPA, IAM, ILA, and Trp, respectively. Band #1 had mobility similar to IAAld and IAN, and later was identified as IAAld using chromogenic reagent 2,4-DNPH. To confirm their identities, TLC plates were sprayed with seven different chromogenic reagents. IAA could be detected by forming characteristic colors with most of the tested chromogenic reagents (Fig. 4BD and Table 2). Salkowski and Ehrlich reagents reacted with most of the indole compounds except IPA and Trp. The bands #2 and #7 reacted with Salkowski reagent, resulting in crimson and purple colors (Fig. 4B), whereas reacting with Ehrlich reagent resulted in purple and violet colors as with standard IAA and IAM, respectively (Fig. 4C). Thus, band #2 was apparently IAA and band #7 was IAM.

View this table:
Table 2

Identification of indole compounds extracted from a culture of C. acutatum isolate KLA207 using chromogenic reagents after TLC separation

BandRfUVSalkowskiEhrlich2,4-DNPHNitrite-nitricNinhydrin-acetateProchazkaRenz and LoewProbable compound
10.79+DrabBluish purplePale brownCrimsonPale yellowPale brownPale yellowIAAld
20.71+CrimsonPurplePale crimsonPinkPale pinkPinkish yellowBluish pinkIAA
30.68+BrownPale grayPale brown?
40.58+YellowPinkish purple, turning greenish yellowPale brownPale yellowYellowish brownYellow with bluish borderTOL
50.53+YellowPale pink?
60.48+OrangePale grayIPA
70.25+PurpleViolet, turning grayPale purpleGrayish bluePale purpleIAM
80.23+Pale yellowPale purple, turning grayBeigePale yellowDrabILA
90.16VioletVioletPale purple−PalePurple?
100.06+VioletVioletViolet bluePale grayPurple?
110+BrownYellowPinkish orangePale yellowPurpleTrp
  • The bands correspond to those in Fig. 4.

  • Distinct colors are indicated in bold. −, indicates no reaction.

Salkowski reagent reacted strongly with commercial TOL (5 mM), but reacted weakly with band #4 (co-migrated with TOL), giving a faint greenish yellow color, indicating that Salkowski had low sensitivity to TOL. The presence of TOL in the cultural extract was evident when the plate was stained with Ehrlich reagent, giving a pinkish purple color (Fig. 4C, band #4). The identities of two bands that reacted with Ehrlich reagent, resulting in a yellow color, were unknown. IAN reacted with Salkowski (Fig. 4B) and nitrite-nitric (data not shown) reagents, giving a characteristic blue color. No corresponding blue color was observed in the cultural extract of C. acutatum. HPLC also failed to identify any peak corresponding to IAN, thus ruling out its presence.

Detection of IPA and IAAld using HPLC was inconsistent, likely due to their low quantities and instabilities. The identities of IAAld and IPA in the cultural extracts were confirmed when the TLC plate was stained with 2,4-DNPH reagent (Fig. 4D). IPA was visualized as a strong band under UV fluorescence, but was undetectable using Salkowski or Ehrlich reagent. However, 2,4-DNPH reagent reacted strongly and specifically with IPA, giving a distinct orange color. Band #6 identified in the cultural extract had similar mobility and orange color as IPA, therefore verifying its identity as IPA. The 2,4-DNPH reagent also reacted weakly with IAAld that was barely visible under UV fluorescence, and gave a faint beige color (Fig. 4D, lane 2). Band #1 in the cultural extract that reacted with 2,4-DNPH, giving a similar beige color, was apparently IAAld. IAAld was barely detectable using HPLC (Fig. 2B) and could also be identified by chromogenic reagents (Table 2). Band #8 exhibited similar color change from purple to gray as the standard ILA after staining with Ehrlich reagent, thus confirming its identity (Fig. 4C and Table 2). ILA was also identified by HPLC. The presence of IAA, TOL, IAM, and IAAld in the Trp-supplemented medium was also confirmed using other chromogenic reagents (Table 2). Nitrite-nitric reagent reacted specifically with IAAld and IAA isolated from cultural extract, giving a very distinct crimson and pink color, respectively (Table 2), but reacted weakly with other indole derivatives. Ninhydrin-acetate reagent reacted strongly only with Trp, but barely or not at all with indole compounds. Both Prochazka and Renz and Loew reagents reacted with many indole compounds, also confirming the identities of indole derivatives produced by C. acutatum. Bands #3, 5, 9, and 10 observed under UV fluorescence had no corresponding reference compounds, and their identities remain to be determined.

In conclusion, we have successfully identified several indole derivatives, including IAA, IAAld, IPA, TOL, ILA, and IAM, from the Trp-supplemented medium using HPLC analysis and chromogenic stains. The data clearly suggested that C. acutatum isolates were capable of utilizing multiple pathways to synthesize IAA. The significant role of IAA production by C. acutatum may contribute to the elevated IAA levels in citrus flowers after infection. The identification protocols developed in this study will be very useful for screening IAA non-producing mutants. Our goals are to identify the role of phytohormones, especially IAA, in the development of young fruit drop of citrus caused by C. acutatum. Further experiments by analyzing IAA non-producing mutants will enable us to establish the relationship between the ability of IAA production in culture and the induction of young fruit drop in planta.


We gratefully thank J.K. Burns and J.W. Grosser for their helpful comments. We also thank H.N. Nigg and R. Yuan for assistance with HPLC. This research was supported by the Florida Citrus Production Research Advisory Council (FCPRAC) Grant number 012-04P, and approved for publication as Journal Series No. R-09428.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
View Abstract