Generation of flavor compounds by biotransformation of genetically modified hairy roots of Hypericum perforatum (L.) with basidiomycetes

Abstract Altogether, 14 basidiomycetes (12BAD, 95PCH, 9WCOC, 5PSA, 96BCI, 331SHIBD, 4MSC, 74HFA, 220MPS, 115PFLA, 111 ICO C, 16LED, 6TSU, and 61LYP) were grown on solid and in liquid media using hairy roots of genetically modified Hypericum perforatum (L.) as the only source of carbon and nitrogen. After the first screening by GC‐MS/MS‐O, two fungi (115PFLA and 61LYP) which resulted in the most pleasant complex natural flavor by biotransformation were selected for further analysis. Twenty‐four new volatile compounds were produced, from which 21 were identified (ethyl hexanoate, ethyl octanoate, benzaldehyde, 2‐undecanone, (E,E)‐2,4‐decadienal, 1‐octen‐3‐one, (E)‐2‐nonenal, ethyl nonanoate, 2‐heptenal, 1‐methoxy‐4‐methylbenzene, 3‐octanone, 1‐decen‐3‐one, (E)‐2‐octenal, 1‐octen‐3‐ol, β‐linalool, ±trans‐nerolidol, anisole, methyl benzoate, 2‐pentylfuran, 1,3‐dichloro‐2‐methoxybenzene, and 1‐dodecanol). Thereof, 15 compounds were perceived at the ODP, from which 13 were identified. Compound identification was performed by comparison of Kovats indices (KI) and mass spectra to those of authentic reference compounds on a polar VF‐WAXms column using headspace solid‐phase microextraction–gas chromatography–mass spectrometry (HS‐SPME‐GC‐MS).

As mentioned above, flavor compounds can be isolated from plant material by using a different organic solvent and methods, but unfortunately their concentration sometimes is lower than 0.01 g/L (Gounaris, 2010). As the industry needs are far bigger than that, synthetic manufacture of volatiles and flavor compounds is growing day by day, like vanilla, for example, 9/10 is coming by synthetic product (Dignum, Kerler, & Verpoorte, 2007). Production of flavor compounds via fermentation or biocatalysis means production via bio routes, during which the product is natural. Flavor compounds resulting from fermentation or biocatalysis are organic products.
Synthetic counterparts, compared to natural, have lower price but costumers are hesitating to use them (Janssens, Pooter, Schamp, & Vandamme, 1992). Biotechnological processes usually use conditions that are less harmful to the environment and the yields of flavor compounds are much higher than those found in, for example, fruit or vegetables (Vandamme & Soetaer, 2002). Considering all of this factors, biotechnological production of volatiles and flavor compounds by involving fungi, yeasts, and bacterial cultures results in "natural flavor compounds" (Gounaris, 2010). Using traditional starter cultures in biotechnological processes for fermentation of by-products is often not sufficient, for example, the spectrum of aroma compounds formed by yeasts of the genus Saccharomyces is limited (Carrau et al., 2008). Since the beginning of the 1950s, attempts have been made to utilize the great biochemical potential of higher fungi to produce various natural flavor compounds (Sugihara & Humfeld, 1954). Basidiomycetes are well known for their nutritional properties and their often pleasant taste for thousands of years (Lorenzen & Anke, 1998), and because of their unique system of extracellular enzymes, also known as secretome, they are able to produce different pharmacologically relevant secondary metabolites and flavor compounds de novo or by biotransformation (Bouws, Wattenberg, & Zorn, 2008;Fraatz & Zorn, 2011). As complex organisms, fungi were studied for so long by chemists, biochemists, genetics, ecologist, biologist, and so on (Tkacz & Lange, 2004). Using fungi for production of traditional products have a long tradition (Papagianni, 2004). Production of flavor compounds from microorganism was started at early of 1980s, and their importance as a big source for this production was academically proved in between 30 and 40 years before (Gatfield, 1999). As a mayor class of fungi, basidiomycetes are well known for their nutritional properties and their often pleasant taste for thousands of years (Lorenzen & Anke, 1998).
Generation of flavor compounds from fungi and their perspective in flavor production by biotechnological processes have been well described by several authors (Agrawal, 2004;Berger & Zorn, 2004;Vandamme, 2003). Pleurotus sapidus, for example, has been shown to efficiently transform the sesquiterpene hydrocarbon valencene to nootkatone, which is an important contributor to the flavor of grapefruit (Bouws et al., 2008).
Hypericum perforatum (L.) known also as Saint John's wort has been used for decades because of its pharmaceutical attributes (Nahrstedt & Butterweck, 2010). In order to avoid the influence of external factors (temperature, pollution, climate changes, etc.) on the quality and quantity of the active secondary metabolites of plant material, scientists are using the genetic transformation (Tusevski et al., 2013).
The first successful transformation of H. perforatum (L.) was done using Agrobacterium rhizogenic ATCC 15,834 (Di Guardo et al., 2003). The transformation protocol and establishment of hairy roots material used for this study was proved using PCR analysis (Tusevski et al., 2013). The application of this transformation verifies the production of some new phenolic compounds with important role as secondary metabolites with pharmaceutical activity of the hairy roots (Tusevski et al., 2013). By knowing the nature of the secondary metabolites, which are produced by the hairy roots of the genetically modified (GMO) H. perforatum (L.), we decided to use these hairy roots as unique and special material for bio-production of new flavor compounds with basidiomycetes. We decided to use roots part of genetically modified H. perforatum (L.) because after sniffing they had only a few flavor compounds compared with aerial part (shoots) which seemed to have many more, what means that we have used a raw material with few flavor compounds, to produce variety classes of flavor compounds using basidiomycetes.

| Substrates
Hairy roots of GMO H. perforatum (L.) were provided from the

| Submerged cultures
Substrate preparation. Hairy roots of GMO H. perforatum (L.) obtained from Skopje University (Tusevski et al., 2013) were used as a substrate for the submerged cultures. Hairy root material was ho-

| Surface screening
Roots of GMO H. perforatum (L.) were autoclaved at 121°C for 20 min separately in a small Erlenmeyer flask covered with aluminum foil. Solution of MEA 2% (2 g agar-agar/100 ml H 2 O) was autoclaved and mixed with the roots under sterile conditions. The suspension was poured into small Petri dishes. The plates were inoculated with the fungi mentioned above, and the growth was evaluated visually daily. When about 70% of the surface of the Petri dishes was covered by fungal mycelium, the surface cultures were sniffed by the trained panel; the screening process has been repeated three times.

| Headspace solid-phase microextraction (HS-SPME)
For HS-SPME, the DVB/CAR/PDMS and PDMS/DVB fibers were The fiber was moved to the SPME fiber conditioning station and heated at 250°C for 20 min.

| Gas Chromatography-olfactometry (GC-MS/ MS-O)
Gas chromatography apparatus used for all measurements was from

| Identification of flavor compounds
Identification of flavor compounds was performed by comparison of Kovats indices (KI), odor impressions, and mass spectra to those of authentic reference compounds and with the NIST 2011 MS library database.

| RE SULTS AND D ISCUSS I ON
The volatiles emitted from hairy roots from GMO H. perforatum (L.)

| Surface screening
In the surface screening, two of the 14 strains evaluated in the screening exhibited interesting flavor profiles (Table 2).

| Flavor analysis
Microorganisms are able to produce all classes of volatile compounds, but aldehydes, alcohols, and organic acid esters are often dominating. In general, biotransformation products were observed between 24 and 190 hr (Gounaris, 2010).
Some C-8 compounds are well known as fungal volatiles, originating from the enzymatic oxidation of linoleic acid. 1-Octen-3-one from 115PFLA, 2-octenal, and 1-octen-3-ol from 61 LYP were not found in the blank samples which indicates that they are biosynthesized as flavor compounds (Zhang et al., 2014).

F I G U R E 3
Chemical structures of the identified odor active compounds database together with their mass spectra (Tables 3 and 4). Two

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest to disclose.

E TH I C A L A PPROVA L
There was no human or animal testing in this study.