Citrus limonoids: analysis, bioactivity, and biomedical prospects

Limonoids are a prominent group of secondary metabolites in citrus fruit. The bitter character of some compounds in this group has historically compromised the quality of citrus fruit and juice. Detecting bitter limonoids in citrus, understanding their origins, and developing methods for their removal from citrus juices have provided the basis for citrus limonoid research. Evaluation of the biological activity of citrus limonoids has indicated the potential of these compounds to improve human health as anticancer, cholesterol-lowering, and antiviral agents. This review chronicles the evolution of citrus limonoid research from defining their participation in citrus bitterness to their potential utilization as important contributors to improving human health and well-being.     

Evaluation of the antioxidant capacity of limonin, nomilin, and limonin glucoside

The antioxidant capacity (AOC) of three representative citrus limonoids, limonin, nomilin, and limonin glucoside, was examined by the oxygen radical absorbance capacity (ORAC), Trolox equivalent antioxidant capacity (TEAC), beta-carotene-linoleic acid bleaching, and 1,1-diphenyl-2-picryl hydrazyl (DPPH) radical scavenging assays. Pure compounds and proper negative (cinnamic acid) and positive (2,6-di-tert-butyl-4-methylphenol (BHT) and ascorbic acid) controls were used to remove any ambiguity in interpreting results. In all cases, limonin and nomilin gave results equivalent to those of cinnamic acid, indicating that they do not possess any inherent AOC and should not be considered antioxidants. Similar results were observed for limonin glucoside, with the exception of an anomalous result obtained from the beta-carotene-linoleic acid bleaching assay. Limonin glucoside was deemed not to be an antioxidant on the basis of the three unequivocal assays.     

Calorimetric evidence of differentiated transport of limonin and nomilin through biomembranes

The effect exerted by two structurally similar limonoids possessing antifeedant and anticancer activity, limonin and nomilin, on the thermotropic behavior of model membranes constituted by dimyristoylphosphatidylcholine (DMPC) vesicles was studied by differential scanning calorimetry. Attention was directed to evaluate modifications in phytochemical-lipid interaction induced by compound structure and lipophilicity and to evidence their different membrane penetration. The two examined compounds, when dispersed in liposomes during their preparation, were found to exert a very different action on the L(beta)-L(alpha) gel-to-liquid crystal phase transition of DMPC multilamellar vesicles. Nomilin caused a detectable effect on the transition temperature (T(m)), shifting it toward lower values with a concomitant small decrease of the associated enthalpy (DeltaH) changes, while limonin was not able to modify the lipid vesicles thermotropic behavior. Modifications induced by nomilin were a function of phytochemical concentration, while the different behavior of limonin can be due to the different polarity induced by the presence of the single A ring in nomilin that possesses an acetyl group versus the A,A' ring system of limonin. Solid limonoids and aqueous dispersions of multilamellar (MLVs) or unilamellar vesicles (LUVs) (limonoids molar fraction 0. 045, 0.12, and 0.18) were left in touch for long incubation times at temperatures higher than T(m) to detect their spontaneous transfer through the medium. By following this procedure, no interaction was detected for limonin with lipid vesicles. The rate of transfer and interaction of nomilin was a function of the kind of vesicle species (faster for LUV, slower for MLV). The interaction, monitored by compound transfer from the solid phytochemical to the lipidic species after several periods of incubation, was on the same order as that detected by preparation carried out in organic solvent. The obtained results can be explained in terms of compound hydrophobicity, and a relation between compound structure and membrane interaction can be suggested. This allows the membrane interaction with nomilin, but the low water solubility of limonin hinders or totally blocks its transfer through the aqueous medium.     

Citrus limonoids and their semisynthetic derivatives as antifeedant agents against Spodoptera frugiperda larvae. A structure-activity relationship study

The antifeedant activity of Citrus-derived limonoids limonin (1), nomilin (2), and obacunone (3) and their semisynthetic derivatives 4-26 was evaluated against a commercially important pest, Spodoptera frugiperda. Simple chemical conversions were carried out on the natural limonoids obtained from seeds of Citrus limon. These conversions focused on functional groups considered to be important for the biological activity, namely the C-7 carbonyl and the furan ring. In particular, reduction at C-7 afforded the related alcohols, and from these their acetates, oximes, and methoximes were prepared. Hydrogenation of the furan ring was also performed on limonin and obacunone. The known antifeedant properties of the Citrus limonoids are confirmed. Comparison with previously reported data shows that insect species vary in their behavioral responses to these structural modifications. Highly significant antifeedant activity (P < 0.01) for two natural (1 and 3) and three semisynthetic limonoids (4, 8, and 10) was observed against S. frugiperda.     

Bioavailability of citrus limonoids in humans

This study utilizes liquid chromatography/mass spectrometry (LC-MS) to analyze the plasma of four groups of four healthy male and female subjects administered high doses of pure limonin glucoside (0.25-2.0 g in 200 mL of buffered water) for the presence of limonin to establish the absorption, metabolism, and bioavailability of citrus limonoids to humans. The plasma analysis revealed increasing amounts of limonin associated with increasing doses of limonin glucoside among the subject groups in mean maximum concentration amounts ranging from 1.74 to 5.27 nmol/L. A high degree of variability in the analyzed limonin concentration was observed within the subject groups. The mean time to maximum concentration was 6 h. A second compound with MS/MS characteristics identical to limonin was detected in amounts up to 5.13 nmol/L and is hypothesized to be a limonin epimer. Poststudy health evaluation established no ill effects among study subjects consuming high doses of limonin glucoside.     

Betaines in fruits of Citrus genus plants

Numerous compounds, many of them osmolytes, were quantified in natural juices and in frozen concentrate juices from fruits of plants of the Citrus genus. L-proline, N-methyl-L-proline (hygric acid), N,N-dimethyl-L-proline (stachydrine), 4-hydroxy-L-prolinebetaine (betonicine), 4-hydroxy-L-proline, γ-aminobutyric acid (Gaba), 3-carboxypropyltrimethylammonium (GabaBet), N-methylnicotinic acid (trigonelline), and choline in the fruit juices of yellow orange, blood orange, lemon, mandarin, bitter orange (Citrus aurantium), chinotto (Citrus myrtifolia), and grapefruit were analyzed by sensitive HPLC-ESI-tandem mass spectrometry procedure. It was found that the most represented osmolytes in the juices, that is, L-proline, stachydrine, and betonicine, can be quantified with minimal sample preparation and short analysis time (about 1 min) also by flow injection analysis (FIA) ESI-MS/MS with the same results as obtained by HPLC ESI-MS/MS. In all of the juices, discrete amounts of choline and trigonelline were present. Conversely, GabaBet was always below detection limits. Notably, N-methyl-L-proline and 4-hydroxy-L-prolinebetaine, which were discovered for the first time in the juice of bergamot (Citrus bergamia Risso et Poit), are also present in all of the citrus juices examined.     

Determination of 5-methyltetrahydrofolic acid and folic acid in citrus juices using stable isotope dilution-mass spectrometry

A stable isotope liquid chromatography-mass spectrometry (LC-MS) method was developed for the quantitative determination of 5-methyltetrahydrofolic acid (5-MTHFA) and folic acid in a variety of commercial citrus juices. Folates were extracted from juices, and the polyglutamyl side chain of 5-MTHFA was cleaved to the monoglutamate form using rat plasma conjugase. The folates were purified on a Bond-Elut column and analyzed by LC-MS with electrospray ionization. The analytes were quantified using the (13)C(5) analogues of 5-MTHFA and folic acid as internal standards. The relative standard error of the method was 3.35% based on replicate analyses (n = 4). This method was then applied to the determination of 5-MTHFA and folic acid in a variety of citrus juices obtained from local supermarkets. It was observed that although both "store" brands and "national" brands of fresh (nonfrozen) juices contained similar concentrations of 5-MTHFA, the "store" brands of fresh juices had on average >5-fold the amount of folic acid compared to the "national" brands. In addition, the "total" folate concentrations were generally below values listed on the food label.     

Estimating bergamot juice adulteration of lemon juice by high-performance liquid chromatography (HPLC) analysis of flavanone glycosides

The chemical composition of 30 samples of juices obtained from bergamot (Citrus bergamia Risso and Poit.) fruits is reported and compared to the genuineness parameters adopted by Association of the Industry of Juice and Nectars (AIJN) for lemon juice. It was found that the compositional differences between the two juices are distinguishable, although with difficulty. However, these differences are not strong enough to detect the fraudulent addition of bergamot juice to lemon juice. Instead, we found the high-performance liquid chromatography (HPLC) analysis of the flavanones naringin, neohesperidin, and neoeriocitrin, which are present in bergamot juice and practically absent in the lemon juice, is a convenient way to detect and quantify the fraudulent addition of bergamot juice. The method has been validated by calculating the detection and quantification limits according to Eurachem procedures. Employing neoeriocitrin (detection limit = 0.7 mg/L) and naringin (detection limit = 1 mg/L) as markers, it is possible to detect the addition of bergamot juice to lemon juice at the 1% level. When using neohesperidin as a marker (detection limit = 1 mg/L), the minimal percentage of detectable addition of bergamot juice was about 2%. Finally, it is reported that the pattern of flavonoid content of the bergamot juice is similar to those of chinotto (Citrus myrtifolia Raf) and bitter orange (Citrus aurantium L.) juices and that it is possible to distinguish the three kinds of juices by HPLC analysis.     

Purification of citrus peel juice and molasses

Citrus peel juice and molasses are extremely bitter and unpalatable byproducts of orange and grapefruit juice production. Major components of interest are soluble sugars, glucose, fructose, and sucrose, which account for 60-70% of the dry solids. Analyses indicate that the remaining components are suspended tissue fragments, proteins, organic acids, mineral ions, phenolic compounds, and polyols. A purification sequence that removed a majority of bitter limonoids and phenolic compounds by adsorption on nonionic, macroporous resins was tested. Residual phenolic compounds were removed by adsorption on activated carbon or anion-exchange resin, which also removed anions of organic and inorganic acids. Taste panel results suggested that debittered products could be acceptable for food uses.     

Antioxidant activity of citrus limonoids, flavonoids, and coumarins

A variety of in vitro models such as beta-carotene-linoleic acid, 1,1-diphenyl-2-picryl hydrazyl (DPPH), superoxide, and hamster low-density lipoprotein (LDL) were used to measure the antioxidant activity of 11 citrus bioactive compounds. The compounds tested included two limonoids, limonin (Lim) and limonin 17-beta-D-glucopyranoside (LG); eight flavonoids, apigenin (Api), scutellarein (Scu), kaempferol (Kae), rutin trihydrate (Rut), neohesperidin (Neh), neoeriocitrin (Nee), naringenin (Ngn), and naringin(Ng); and a coumarin (bergapten). The above compounds were tested at concentration of 10 microM in all four methods. It was found that Lim, LG, and Ber inhibited <7%, whereas Scu, Kae, and Rut inhibited 51.3%, 47.0%, and 44.4%, respectively, using the beta-carotene-linoleate model system. Lim, LG, Rut, Scu, Nee, and Kae showed 0.5% 0.25%, 32.2%, 18.3%, 17.2%, and 12.2%, respectively, free radical scavenging activity using the DPPH method. In the superoxide model, Lim, LG, and Ber inhibited the production of superoxide radicals by 2.5-10%, while the flavonoids such as Rut, Scu, Nee, and Neh inhibited superoxide formation by 64.1%, 52.1%, 48.3%, and 37.7%, respectively. However, LG did not inhibit LDL oxidation in the hamster LDL model. But, Lim and Ber offered some protection against LDL oxidation, increasing lag time to 345 min (3-fold) and 160 min (33% increase), respectively, while both Rut and Nee increased lag time to 2800 min (23-fold). Scu and Kae increased lag time to 2140 min (18-fold) and 1879 min (15.7-fold), respectively. In general, it seems that flavonoids, which contain a chromanol ring system, had stronger antioxidant activity as compared to limonoids and bergapten, which lack the hydroxy groups. The present study confirmed that several structural features were linked to the strong antioxidant activity of flavonoids. This is the first report on the antioxidant activity of limonin, limonin glucoside, and neoeriocitrin.     

Phloroglucinol from phlorin hydrolysis for testing quality of commercial orange juices and beverages

The European Association of Juices and Nectars Producers (AIJN) is evaluating the opportunity to introduce the content of phlorin (3,5-dihydroxyphenyl beta-D-glucopyranoside), a peel marker for oranges, as a parameter for testing the quality of orange juices. Because of the lack of a commercial standard of phlorin and its laborious isolation procedures, in this contribution is developed a simple and reliable method for measuring the phlorin level as the corresponding aglycon phloroglucinol, obtained after a total enzymatic hydrolysis of the sample. The method was applied to the quantification of phloroglucinol in several industrial and commercial blond and pigmented orange juices and beverages based on 12% orange juice. Under the same extraction procedure, the phloroglucinol content in the pigmented juices was higher than in the blond ones. No significant difference was obtained between not from concentrate juices and reconstituted from concentrate juices. The marker amount increases in the highly processed orange fruits and in the byproducts of citrus processing due to the contact of the juice with the albedo, which is the major source of phlorin. In orange-based beverages the phloroglucinol content revealed a large heterogeneity and a poor quality of the raw juices used.     

Using an ion-trap MS sensor to differentiate and identify individual components in grapefruit juice headspace volatiles

An ion-trap mass spectrometer chemical sensor has been utilized to differentiate between grapefruit juices that differ only in the concentration of a single component, and the sensor was able to identify that component. Grapefruit juice was fortified with 40 to 2000 ppm vanillin, a low-level naturally occurring compound in citrus juices. Principal components analysis and discriminant analysis of mass spectral data (m/z 50-200) provided clear separation of the grapefruit juice samples. Vanillin was observed in the juice headspace at the 40 ppm level, with identification possible at the 100 ppm level using either MS or MS/MS.     

Grapefruit juices impair the bioaccessibility of β-carotene from orange-fleshed sweet potato but not its intestinal uptake by Caco-2 cells

Among various factors influencing β-carotene (Bc) bioavailability, information on interactions between carotenoids or other micronutrients such as flavonoids during a meal that contains different plant-derived foods is quite limited. Because orange-fleshed sweet potato (OFSP) is an important Bc-rich staple food, a source of vitamin A in developing countries, this study focused on the effect of citrus fruit juice carotenoids and flavonoids on Bc bioaccessibility from OFSP. In vitro digestion coupled with the Caco-2 cell culture model was used to evaluate the bioaccessibility and cellular uptake of Bc from OFSP in the presence of pink grapefruit (pGF) or white grapefruit (wGF) juices. The addition of grapefruit juices significantly decreased the bioaccessibility, by up to 30%, but not the cellular uptake of Bc from boiled OFSP. Lycopene, but more probably naringin, present in grapefruit juices was suspected to be responsible for the inhibitory effect of the citrus juices on Bc bioaccessibility. This inhibition was apparently due in part to competition for incorporation between Bc and naringin into mixed micelles during in vitro digestion. In contrast, Bc uptake from dietary micelles was not impaired by naringin.     

Effect of extraction method on the concentrations of selected bioactive compounds in mandarin juice

A mandarin-type citrus fruit, ponkan (Citrus reticulata), was processed by in-line, chopper pulper, and hand-press extractions to investigate the effect of extraction method on the concentrations of bioactive compounds in processed juice. Concentrations of polymethoxylated flavones (tangeretin, nobiletin, and sinensetin) and beta-cryptoxanthin in juice, and inhibitory activities against arachidonate cyclooxygenase and lipoxygenases of the juice extract were analyzed. The juice processed by hand-press extraction contained the largest amounts of nobiletin (3.56 mg/100 mL), tangeretin (4.10 mg/100 mL), and sinensetin (0.13 mg/100 mL). Concentrations of beta-cryptoxanthin were 0.66, 0.59, 0.55, and 0.50 mg/100 mL in chopper pulper, in-line (5/64 in.), in-line (8/64 in.) and hand-press juices, respectively. Both extracts of in-line juices showed greater inhibitory activity toward platelet 12-lipoxygenase than the others. The inhibitory effect of hand-press juice extract on platelet cyclooxygenase activity was remarkable among juice extracts. All juice extracts effectively inhibited polymorphonuclear 5-lipoxygenase activity at nearly the same rate.     

Separation of flavanone-7-O-glycoside diastereomers and analysis in citrus juices by multidimensional liquid chromatography coupled with mass spectrometry

The major flavanone-7-O-glycoside constituents in citrus fruit juices (naringin, hesperidin, neohesperidin, narirutin, and eriocitrin) were separated as diastereomers by multidimensional liquid chromatography. The method consisted of coupling two HPLC columns: a reversed-phase (RP(18)) column was used for the separation of flavanone glycosides, which were, then, individually switched into a carboxymethylated beta-cyclodextrin (beta-CD)-based column and resolved as the corresponding stereoisomers. The method was used for the full analysis of flavanone glycosides in fresh hand-squeezed and commercial fruit juices by combining the quantitative estimation with the diastereomeric analysis. Quantitative data were in general consistent with previously reported data in this field. CC-LC isomer analysis was carried out by coupling the beta-CD column with a mass spectrometer operated with negative ion electrospray ionization (ESI-MS). The results showed that hesperidin was present in orange juices almost exclusively as the 2S isomer, whereas narirutin had mainly the 2R configuration. In grapefruit juices (2S)-naringin prevailed with the respect to the 2R isomer, whereas the opposite was true for narirutin. Lemon juices contained eriocitrin stereoisomers in equal amount (50% each), but hesperidin was almost exclusively found as the 2S isomer. Significant differences of the diastereomeric ratios were observed between freshly squeezed juices and juices from commercial sources.     

Capillary electrophoresis and high-performance liquid chromatography determination of polyglutamyl 5-methyltetrahydrofolate forms in citrus products

The 5-methyltetrahydrofolate (5mTHF) polyglutamates in citrus products were analyzed by capillary electrophoresis (CE) and high-performance liquid chromatography (HPLC). Folate species were purified from citrus products and concentrated from 2- to 100-fold using combined folate-affinity chromatography and C18 extraction. Seven polyglutamyl 5mTHFs were found in most not-from-concentrate (NFC) orange juices (OJ) in total amounts of approximately 1 nmol/mL, with varying distributions of individual polyglutamates. Folate amounts and distributions were also measured in orange fractions, single-strength OJ from concentrate, NFC grapefruit juice, and citrus peel molasses. Models containing ascorbic acid had folate thermal degradation rates one-seventh that of models without ascorbic acid. Pasteurization studies demonstrated that folate loss was <2% for commercial OJ pasteurization conditions (i.e., 93 degrees C for 5 s, 88 degrees C for 15 s, and 82 degrees C for 30 s). Both methods were precise, reproducible, and potentially faster than traditional analytical procedures requiring enzymatic deconjugation and microbial assays.     

Determination of vanillin in orange, grapefruit, tangerine, lemon, and lime juices using GC-olfactometry and GC-MS/MS

The presence of vanillin in orange, tangerine, lemon, lime, and grapefruit juices has been identified and confirmed using high-resolution GC retention index values, mass spectra, and aroma quality. The impact of vanillin on the flavor score for grapefruit juice is discussed and reported to be minimal. Vanillin concentrations are determined to be in the low parts-per-million range for the various citrus juices. The calculated concentrations in the orange, tangerine, lemon, lime, and grapefruit juices are 0.20, 0.35, 0.41, 0.35, and 0.60 ppm, respectively. Pasteurization produced an average 15% increase in the concentration of vanillin in grapefruit juices. Vanillin did not correlate well with the overall flavor score despite a rather intense signal using OSME gas chromatography-olfactometry software.     

Characterization of the anthocyanin fraction of sicilian blood orange juice by micro-HPLC-ESI/MS

Anthocyanins of red orange juices were analyzed by micro-HPLC coupled on-line with an MS detector equipped with an ESI source. The use of microcolumn HPLC greatly enhanced detection performance, allowing direct identification of the anthocyanins present in the orange juices. The use of a soft ionization technique allows detection of the molecular ions of the aglycons. Eight components were identified, five of them for the first time in red orange juice. Three additional anthocyanins were detected, of which only the aglycon was identified.