Furanocoumarins (FCs) in the human diet irreversibly inhibit human cytochrome P450 3A4 (CYP 3A4) and are responsible for the “grapefruit/drug” interaction phenomenon. Previously, we reported that FCs in grapefruit juice (GFJ) bind to autoclaved Aspergillus niger, and this binding reduced the GFJ inhibition of CYP 3A4. However, A. niger is not an edible fungus, and thus, potentially similar binding by edible fungi was also characterized. In this study, autoclaved Morchella esculenta, an edible ascomycete, removed much of the FC content in GFJ, resulting in decreased inhibition of CYP 3A4 activity by the GFJ. Three other edible fungi, Monascus purpureus, Pleurotus sapidus, and Agaricus bisporus, were evaluated for their binding with two of the major FCs in GFJ, 6′,7′-dihydroxybergamottin (DHB) and bergamottin (BM). These autoclaved edible fungi removed these FCs from GFJ, similar to M. esculenta, indicating that binding is a general, passive interaction between FCs and fungal hyphae. The removal of FCs was independent of pH in GFJ. Dried fungal material of M. esculenta was also effective in removing FCs from GFJ and occurred with GFJ samples prepared from both fresh grapefruit and GFJ concentrate.
Introduction
Coumarins and furanocoumarins (FCs) are a class of phenolic compounds produced in certain varieties of citrus, including grapefruit, pummelo, limes, lemons, and others
(1, 2). Grapefruits are particularly rich in the FCs, where two of the major FCs are 6′,7′-dihydroxybergamottin (DHB) and bergamottin (BM)
(1, 3). Significantly, most of the FCs in grapefruit juice (GFJ) are powerful inhibitors of human cytochrome P450 (CYP) enzymes 3A4, 2B6, 3A5, 2D6, and 2C9, responsible for the metabolism of many widely prescribed medications
(4-7). This inhibition can possibly influence the bioavailability of these affected medications, which potentially creates a risk of adverse events
(8). Moreover, this “grapefruit/drug” interaction has adversely affected the citrus industry for years, even though grapefruit possesses antioxidant activity and numerous beneficial health phytochemicals and putatively acts as a protector against cancer and cardiovascular diseases
(9, 10).
To reduce these interactions, there have been recent attempts to remove FCs from GFJ using chemical, physical, and microbiological methods
(11-14). The method developed by Paine et al.
(12) used a series of chemical extractions and reconstitutions of compounds in GFJ to produce a FC-free GFJ, while Uesawa and Mohri
(13, 14) used ultraviolet (UV) radiation and heat to degrade the FCs in GFJ. Meanwhile, the potential use of autoclaved fungus to adsorb and remove FCs in GFJ has also been explored
(11).
Many edible fungi are rich in vitamins, minerals, and folates and are regarded as nutritive foods, nutraceuticals, and antitumor agents
(15-18). The edible fungi are in the two major fungal groups, ascomycetes and basidiomycetes, differentiated by the formation of their sexual spores, where a diverse array of fruiting bodies are produced
(19). The facts that adsorption of FCs has recently been described in the ascomycete
Aspergillus niger (11) and that ascomycetes can be differentiated from basidiomycetes by features such as morphology and chemical composition
(20) raised a question whether basidiomycetes can similarly adsorb the FCs in GFJ. The objectives of this study were to examine whether edible fungi, including two ascomycetes,
Morchella esculenta and
Monascus purpureus, and two basidiomycetes,
Pleurotus sapidus and
Agaricus bisporus, can remove FCs in GFJ and to evaluate whether the removal can be affected by different factors, i.e., pH of GFJ, sources of GFJ (fresh versus concentrate), and fungal materials (wet versus dry).
Materials and Methods
Fungal Organisms and Cultures A culture of M. esculenta Dill. ex Pers. (FP-140146) was obtained from the United States Department of Agriculture (USDA) Forest Service, Center for Forest Mycology Research in Madison, WI. M. purpureus Tiegh. was obtained from the Agricultural Research Service (ARS) Culture Collection (NRRL 1596) at the National Center for Agricultural Utilization Research. P. sapidus (Fr.) P. Krumm. nom. cons. was obtained from the American Type Culture Collection (24987). A culture of A. bisporus L. was prepared from an isolation made from the trama tissue of the pileus from a fresh basidiocarp.
The cultures of M. esculenta and M. purpureus were placed on potato dextrose agar (PDA) (BD/Difco Sparks, MD), and these cultures were maintained to provide initial inoculum by placing fungal hyphae (0.5 g) into 100 mL flasks of YM broth (10.0 g of glucose, 5.0 g of peptone, 3.0 g of yeast extract, and 3.0 g of malt extract in 1 L water) and placing them on an orbital shaker at 100 rpm and 23 °C for 1 week. Mycelial pellets from the cultures were broken apart by placing them into a sterile 1 L Waring stainless-steel blender cup and macerating the tissue for 2 min, and aliquots of the resulting suspension were equally divided into six 2 L flasks containing 1 L of YM broth. The flasks were shaken on a large orbital shaker at 140 rpm and 20 °C until there was enough mycelial tissue (≈20 g). P. sapidus and A. bisporus were maintained on YM agar containing thiamine (100 μL L−1) and inoculated on YM broth as described above, with the exception that YM broth contained thiamine.
The fully grown fungal tissues in the flasks were harvested and autoclaved for 20 min at 121 °C to kill the fungus. After cooling, the contents (broth + dead mycelium) of the flasks were vacuum-filtered and the remaining mycelial mass was immediately used as an adsorbent. The wet mass of M. esculenta was further vacuum-dried at 40 °C for 24 h and ground into fine powder.
Interaction of GFJ with Fungal Materials The GFJ samples were prepared from either fresh grapefruit or GFJ concentrate. The GFJ from fresh grapefruit was prepared by manually squeezing white Marsh grapefruit (Citrus paradisi). The GFJ samples from GFJ concentrate were prepared by mixing 1.5 L water with GFJ concentrate (355 mL), which was purchased at a local store. The pH of GFJ samples was further adjusted to 5.0 to minimize degradation of GFJ FCs.
The macerated M. esculenta samples (1.5, 3.0, and 9.0 g) were added to 50 mL of GFJ in 250 mL flasks and mixed for 4 h at 250 rpm and 25 °C to determine optimal amounts of biomass for the removal of GFJ FCs. For additional experiments using M. esculenta, M. purpureus, P. sapidus, and A. bisporus, 2.0, 4.0, and 6.0 g of the autoclaved wet fungal hyphae were mixed with GFJ at the same conditions as described above. For experiments using dry M. esculenta, 0.26, 0.52, and 0.78 g corresponding to their wet weights (2, 4, and 6 g) were added to GFJ samples and mixed as described above. Dry weights of M. esculenta were determined by weighing 10 g of samples before and after vacuum drying at 40 °C for 24 h. After 4 h, the untreated and fungal-treated GFJ samples were vacuum-filtered through a Whatman no. 1 filter, and the resulting filtrates and fungal material on the filter were collected.
Analysis of FCs in GFJ Both filtrates and fungal materials were extracted with 100 mL of ethyl acetate 3 times. The extracts were dried using a rotary evaporator; the resulting pellets were dissolved in 10 mL of acetone, and 40 μL aliquots were subjected to high-performance liquid chromatography (HPLC) analysis as previously described
(21). The FCs obtained from control (no fungi added) and treated (various amounts of edible fungal biomass added) GFJ samples were identified by comparing their elution times, UV absorbance at 320 nm, and mass spectrometry (MS) data to authentic FCs and analyzed as previously described
(21). Concentrations (means ± standard deviations) of DHB and BM in GFJ samples prepared from fresh grapefruit were 5.32 ± 0.46 and 3.34 ± 0.37 ppm, respectively, while those of DHB and BM in GFJ samples prepared from GFJ concentrate were 4.11 ± 0.59 and 1.88 ± 0.53 ppm, respectively.
CYP 3A4 P450 Enzyme Inhibition Assay Inhibition assay of CYP 3A4 enzyme activity was carried out according to the instructions of the manufacturer (Invitrogen, Carlsbad, CA). The 2.5 μL of orange juice, untreated GFJ, or Morchella-treated GFJ was used for the inhibition assays. For treated GFJ, 6.0 g (wet weight) of autoclaved M. esculenta was mixed with 50 mL of GFJ for 4 h, the GFJ was filtered through a Whatman no. 1 filter, and the resulting GFJ filtrates were used for the inhibition assays. The control activity (100% activity) using water was determined as 1.086 nmol min−1 (nmol of CYP 3A4)−1.
Statistical Procedures All experiments in this study were conducted with three replicates. Student t test was used to compare the differences in changes between controls and treatments if applicable. Two-tailed p values were calculated to report significant differences in the mean values.
Discussion
The edible
M. esculenta efficiently adsorbed most of the nonpolar FCs in GFJ, where only a trace amount of BM (
7) remained in the GFJ treated with 9 g of the autoclaved fungal material (Figure
1). The removal of FCs in GFJ by the fungus was comparable to that of FCs by
A. niger, as previously reported
(11). Polar FCs, FCU-1, FCU-2, and DHB (
1−
3, Figure
1), still remained in GFJ treated with 9 g of
M. esculenta, which was similar to the findings of the same study observed with
A. niger (11). The similarity in removal efficiency of FCs by
M. esculenta and
A. niger suggests that these fungi contain a comparable amount of components responsible for the binding within tissue. In addition, 40% of the CYP 3A4 inhibition remained in the fungal-treated GFJ (Figure
2), suggesting that the residual activity was likely due to incomplete removal of FCs, i.e. FCU-1, FCU-2, and DHB (
1−
3, Figure
1). Nonetheless, together with the results shown in Figures
1 and
2, our data demonstrate that
M. esculenta can remove FCs in GFJ, leading to a decreased inhibition of CYP 3A4 enzymatic activity.
Because basidiomycetes can be differentiated from ascomycetes by chemical composition
(20), the removal of FCs by
M. esculenta led us to examine a possible removal of FCs by other edible fungi. Interestingly, removal of DHB and BM by
M. purpureus,
P. sapidus, and
A. bisporus also occurred as shown by
M. esculenta (Table
1 and Figure
3). The results show that the binding may occur with any fungus, regardless of ascomycetes or basidiomycetes, suggesting that the binding is a general interaction, where FCs passively bind to fungal hyphae. It is likely that the contents of the components responsible for the binding may not differ greatly among these fungi, as evidenced by the removal of DHB and BM from GFJ by these fungi (Table
1 and Figure
3).
The DHB and BM are distributed in the “cloud” of GFJ, defined as the portion of suspended particles retained in suspension after centrifugation at 360
g for 10 min
(22). Even though precise localization of FCs in the cloud has not been characterized, given that lipophilic compounds, such as flavonoids and alkanes, are found in endoplasmic reticulum (ER)-derived vesicles in plant cells
(23) and that the cloud is rich in lipids derived from juice sacs
(24), it can be postulated that the lipophilic FCs are localized in lipid vesicles protected from an aqueous environment in the GFJ cloud. Moreover, with hypothesized hydrophobic interaction between lipids and FCs
(11, 25) and little effect of GFJ pH on the binding phenomenon (data not shown), the binding of FCs to fungal biomass further suggests that lipophilic components in fungi sequester FCs, so that nonpolar FCs, particularly BM and FC dimers, associate with hydrophobic sectors in fungal material. In addition, the removal by the wet and dry fungal biomass was observed from GFJ samples prepared from commercial GFJ concentrates (data not shown), suggesting that chemical stability of FCs in GFJ concentrates has been maintained through industrial GFJ processing. Furthermore, the efficient removal of DHB and BM by wet and dry biomass (Figure
3) indicates that the drying process does not affect the abundance and binding or physical property of components responsible for the removal. With regard to this, other dried biomaterials efficiently sequestered FCs in GFJ (our unpublished results).
In summary, FCs in GFJ were removed by an edible fungus, M. esculenta, and the removal of FCs from GFJ reduced an inhibition of CYP 3A4. The removal of DHB and BM was also observed from other edible fungi, M. purpureus, P. sapidus, and A. bisporus, suggesting that the binding of FCs to fungal hyphae is a passive interaction and these fungi contain components responsible for the binding. The removal of FCs in GFJ was also observed with dry M. esculenta, making it possible to use the dry material for the removal of FCs in GFJ and further characterize constituents binding to the FCs.