Biotransformation of [U-13C]linoleic acid suggests two independent ketonic- and aldehydic cycles within C8-oxylipin biosynthesis in Cyclocybe aegerita (V. Brig.) Vizzini

Although the typical aroma contributing compounds in fungi of the phylum Basidiomycota are known for decades, their biosynthetic pathways are still unclear. Amongst these volatiles, C8-compounds are probably the most important ones as they function, in addition to their specific perception of fungal odour, as oxylipins. Previous studies focused on C8-oxylipin production either in fruiting bodies or mycelia. However, comparisons of the C8-oxylipin biosynthesis at different developmental stages are scarce, and the biosynthesis in basidiospores was completely neglected. In this study, we addressed this gap and were able to show that the biosynthesis of C8-oxylipins differs strongly between different developmental stages. The comparison of mycelium, primordia, young fruiting bodies, mature fruiting bodies, post sporulation fruiting bodies and basidiospores revealed that the occurance of the two main C8-oxylipins octan-3-one and oct-1-en-3-ol distinguished in different stages. Whereas oct-1-en-3-ol levels peaked in the mycelium and decreased with ongoing maturation, octan-3-one levels increased during maturation. Furthermore, oct-2-en-1-ol, octan-1-ol, oct-2-enal, octan-3-ol, oct-1-en-3-one and octanal contributed to the C8-oxylipins but with drastically lower levels. Biotransformations with [U-13C]linoleic acid revealed that early developmental stages produced various [U-13C]oxylipins, whereas maturated developmental stages like post sporulation fruiting bodies and basidiospores produced predominantly [U-13C]octan-3-one. Based on the distribution of certain C8-oxylipins and biotransformations with putative precursors at different developmental stages, two distinct biosynthetic cycles were deduced with oct-2-enal (aldehydic-cycle) and oct-1-en-3-one (ketonic-cycle) as precursors.

Furthermore, the comparison of the volatile composition at different developmental stages or parts focused mainly on oct-1-en-3-ol (6) or, in case several C8-oxylipins have been analysed, was limited to fruiting bodies. However, when different developmental stages of fruiting bodies from fungi of the phylum Basidiomycota were observed, maturation was accompanied with vigorous variations in the occurence of mainly compounds 5-8 while compounds 1-4 seem to undergo minor variations (Cho et al. 2006, Li et al., 2016Matsui et al., 2003;Mau et al., 1997). Unfortunately, all studies lack a comparison of the C8-oxylipin composition at different stages of the fungal life cycle. So far, differences in the occurrence of C8-oxylipins were only shown for fruiting bodies, where different maturity stages of the basidiocarp were investigated. An exception is the recently published work on the volatilome of Cyclocybe aegerita (V. Brig.) Vizzini, where different stages during fructification have been analysed. Nevertheless, C8-oxylipin biosynthesis in basidiospores was completely neglected, and only oct-1-en-3-ol (6) and octan-3-one (8) have been detected (Orban et al., 2020). In this study, this gap is addressed, and the C8-oxylipin composition as well as the C8-oxylipin-biosynthesis via biotransformations with [U-13 C]labeled linoleic acid to C8-oxylipins in mycelium, primordia, immature fruiting bodies, mature fruiting bodies, basidiospores and fully mature fruiting bodies after sporulation within the mushroom C. aegerita, used as a model fungus (Frings et al., 2020), is investigated. Additionally, the fatty acid composition between different morphological stages is compared.

Fungal cultivation and harvesting
For mycelium propagation, 1.5% malt-extract-agar was used. High petri-dishes with 1.5% malt-extract were inoculated with an C. aegerita AAE-3 (Herzog et al., 2016) myceliumovergrown circular agar slice (diameter 0.5 cm) and incubated at 24 °C in darkness until it reached the edge of the petri-dishes. The obtained mycelium was gently removed from the agar-plate and used subsequently for volatile measurements and biotransformations without storage or any other treatment. For fruiting, the overgrown petri-dishes were further cultivated at 24 °C with a 12 h dark/night shift at 95% relative air humidity (Rumed-Rubarth Apparate GmbH, Laatzen, Germany) until the desired developmental stage was obtained. Primordia, immature fruiting bodies with closed caps and post sporulation fruiting bodies were harvested with a scalpel, sliced and immediately used for further experiments. Basidiospores of C. aegerita were harvested by cultivation of substrate blocks with immature fruiting bodies (provided by druid-Austernpilze, Ottrau, Germany) at 24 °C in a 12 h dark/night shift at 85% relative air humidity in a climate chamber until sporulation. Dropped basidiospores were gently isolated from the fruiting bodies without damaging them and used for further measurements. Images of the different developmental stages were taken with a Samsung 32 MP camera. Microscopic images were taken using an Olympus microscope BX43 (Olympus Europa SE & CO. KG, Hamburg, Germany) with 20 × and 100 × magnification equipped with a SC50 (5.0 MP) camera.

C8-oxylipin identification and quantification
C8-oxylipins were identified by using authentic standards. Furthermore, the identities were confirmed by their characteristic fragmentation pattern based on standards and the National Institute of Standards and Technology (NIST) Chemistry WebBook. All oxylipins were measured in the headspace of commercial vials. To avoid contaminations and minimise known discriminations of headspace compositions by internal standards or other supplemented compounds, no internal standard was added.

Fatty acid analysis
Fatty acid extraction was carried out via homogenation of the samples with N 2(liq) , followed by the addition of 3 mL n-hexane. The mixture was incubated for 20 min at room temperature. Cell debris were removed by centrifugation for 10 min at 4,063 g. Fatty acid containing organic phase was removed from the aqueous phase and concentrated to ~ 1 mL via a rotary evaporator. Next, 4 mL 0.5 M methanolic NaOH-solution was added to the concentrated extract, followed by incubation for 10 min at 80 °C. For methylation, 3.5 mL 20% methanolic boron trifluoride was added to the extract and incubated for another 5 min at 80 °C. After the mixture was cooled down, 5 mL saturated NaCl-solution was added. The organic phase was separated, dried over anhydrous Na 2 SO 4 and analysed via GC-MS using a VFWax column (30 m × 0.25 mm × 0.25 µm film thickness, Santa Clara, USA) operating in splitless mode under the following parameters: carrier gas, He with a constant flow of 1.2 mL · min −1 . Oven temperature was at 40 °C (3 min), 3 °C · min −1 to 240 °C and hold for 12 min. The mass spectrometer operated in electron impact mode with an electron energy of 70 eV and scanned in a range of m/z 33-300. Fatty acids were identified by their characteristic fragmentation pattern based on authentic standards and the National Institute of Standards and Technology (NIST) Chemistry WebBook.

Data processing
The peak areas of the identified C8-oxylipins were used to determine the relative amount of the respective substance. Each relative peak area of a detected C8-oxylipin at a certain developmental stage was added up and considered 100%, meaning that the peak area of a certain C8-oxylipin correlates to the added up peak areas. This approach enables an appropriate overview of the C8-oxylipin composition. Additionally, means of the peak area of a certain C8-oxylipin were expressed as AU/g of fresh weight tissue to display the occurence of each C8-oxylipin at different developmental stages to distinguish between the relative composition of C8-oxylipins and the actual levels. The processed data were used to generate the heatmaps using OriginPro (OriginLab Corporation, Northampton, MA, USA).

Endogenous C8-oxylipin composition and biotransformation of [U-13 C]linoleic acid
The composition of endogenous C8-oxylipins at different developmental stages were analysed by means of SPME-GC/MS peak areas. To verify which developmental stage produced the highest amount of a certain C8-oxylipin, the obtained peak areas were expressed as arbitrary units per gram of fresh weight tissue (AU/g ft) ( Fig. 1). Measuring the volatilome at different developmental stages only allows detection of accumulated C8-oxylipins prior to harvesting. Since we were also interested in the overall enzymatic activity towards linoleic acid conversion at a certain developmental stage, additional biotransformation experiments with [U-13 C]linoleic acid were carried out to provide informations on enzymatically produced [U-13 C]C8-oxylipins and conceivable induction of the endogenous biosynthesis. In total, eight C8-oxylipins were identified, which eluted as a mixture of endogenous and [U-13 C]-labeled compounds (Fig. 2).
In the mycelium and fruiting body samples, the fatty acid composition was alike (Table 1). However, the ratio of SFA and MUFA increased by ~ 15% with ongoing maturation. Simultaneously, the linoleic acid ratio decreased with  maturation. In basidiospores, the fatty acid composition was drastically different compared to the other analysed compartments. SFA, especially palmitic acid rose to ~ 35% while linoleic acid accounted for only ~ 41%. Furthermore, oleic acid soared to ~ 19% in basidiospores (Table 1).

C8-oxylipin composition
Unlike prior studies in which oct-1-en-3-ol (6) was the predominant volatile in basidiomycetous mycelium or fruiting body homogenates (Akakabe et al., 2005, Cho et al. 2006, Cruz et al., 1997, Li et al., 2016, Matsui et al., 2003, Tressl et al., 1982, the most abundant C8-oxylipins in C. aegerita throughout all developmental stages was octan-3-one (8). This discrepancy is mainly caused due to the used extraction and homogenation method, which is often overlooked. This becomes clear in studies of Combet et al., (2009) and Rapior et al., (1998), who compared either homogenised, sliced and whole samples of fruiting bodies of A. bisporus or different extraction methods of fruiting bodies of C. aegerita, respectively. Disruption of fungal tissue could lead to the release of membrane bound fatty acids, interfering with endogenous formation and, thus, complicate the understanding of C8-oxylipin biosynthesis. Moreover, cellular compartments could also be harmed and, thus, might trigger the formation and release of C8-oxylipins (Combet et al., 2009). While homogenised samples and liquid extraction led to higher production of oct-1-en-3-ol (6), headspace measurements of whole or sliced samples showed higher octan-3-one (8) levels (Combet et al., 2009). Previous studies focused on disruption and liquid extraction methods and consequently found oct-1-en-3-ol (6) as the main C8-oxylipin in different fungal species. Nonetheless, the variety of different C8-oxylipins (up to eight) detected in these studies is widely in accordance with our findings (Akakabe et al., 2005, Cho et al. 2006, Cruz et al., 1997, Li et al., 2016, Matsui et al., 2003, Tressl et al., 1982. Although octan-3-one is the predominant oxylipin in this study and not oct-1-en-3-ol (6), both were detected throughout all developmental stages emphasising their biological importance. Electroantennographic experiments showed that the fungivorous beetle Bolitophagus reticulatus is able to differentiate between the most common C8-oxylipins to assess host quality (Holighaus et al., 2014). Here, oct-1-en-3-ol (6) acts as repellent and octan-3-one (8) as attractant. Contradictory, the three wood-living generalist beetles Malthodes fuscus, Anaspis marginicollis, Anaspis rufilabris and the moth Epinotia tedella were attracted to oct-1-en-3-ol (6), whereas the generalist predator on fungus-insects Lordithon lunulatus distinguished between oct-1-en-3-ol (6) and octan-3-one (8) and was attracted by a mixture of both C8-oxylipins (Fäldt et al. 1998). Besides interspecies communication, C8-oxylipins might also operate as intra-species signals as shown in Penicillium paneum where oct-1-en-3-ol (6) functions as a self-inhibitor signal in spore germination (Schulz-Bohm et al., 2017). These studies reveal that C8-oxylipins have different functions and, consecuently, their occurence depends on the developmental stage of the fungus. To investigate the enzymatic potential of each developmental stage towards the production of different C8-oxylipins, [U-13 C]linoleic acid has been added to different fungal tissues. The resulting [U-13 C]C8-oxylipins co-eluted with the endogenous C8-oxylipins and, therefore, the detected mass spectra were a mixture of nonlabeled α-cleavage fragmentations, McLafferty ions and their corresponding fragmentations with a 13 C-mass shift ( Fig. 2A-H).
In primordia, only three C8-oxylipins (octan-3-ol (5), oct-1-en-3-ol (6) and octan-3-one (8)) occured endogenously, although primordia were able to produce all eight C8-oxylipins when [U-13 C]linoleic acid was added (Fig. 1A,  B). Interestingly, with the addition of [U-13 C]linoleic acid to primordia, all C8-oxylipins were detected as non-labeled and [U-13 C]compounds that indicates the induction of endogenous C8-oxylipin biosynthesis. In contrast, all eight C8-oxylipins emerged endogenously in immature fruiting bodies. Therefore, an upregulation of aldehydic-and ketonic-cycle related genes in the transition from primordia to immature fruiting bodies seems conceivable. By comparing the AU/g ft of each C8-oxylipin in these two developmental stages, it is clearly noticeable that immature fruiting bodies produced more C8-oxylipins of both cycles compared to primordia, which is consistent with a conceivable upregulation of the aldehydic-and ketonic-cycle between these two stages (Fig. 3A, B). Furthermore with ongoing maturity, a downregulation of the aldehydic and an upregulation of the ketonic-cycle seems reasonable (Fig. 1). This is elucidated by the comparison of the C8-oxylipin composition and biosynthesis in immature fruiting bodies, mature fruiting bodies, post sporulation fruiting bodies and basidiospores, where the total number of different C8-oxylipins declined from eight in immature fruiting bodies, to five in mature and post sporulation fruiting bodies, and three in basidiospores (Fig. 1). The vanishing C8-oxylipins with maturation are all related to the aldehydic-cycle, while the ketoniccycle related C8-oxylipins remain in high amounts. This strengthens the hypothesis of two distinct biosynthetic pathways. Similar observations were made by Li et al., (2016), who showed that egg-shaped, bell-shaped and mature fruiting bodies of Tricholoma matsutake produced predominantly octan-3-one (8) and oct-1-en-3-ol (6). With ongoing maturation of T. matsutake increasing levels of octan-3-one and decreasing oct-1-en-3-ol (6) levels have been detected. On the other hand, octan-1-ol (2), octanal (4) and oct-2en-1-ol (1) were either not detected or in low levels that decreased with ongoing maturity. This is in accordance with our hypothesis that a ketonic cycle (e.g. octan-3-one (8)) is getting more prominent during maturation, whereas the aldehydic-cycle (e.g. octan-1-ol (2)) is decreasing concurrently. Furthermore in T. matsutake, oct-2-enal was present in higher proportions in immature fruiting stages (egg-and bell-shaped fruiting bodies) followed by a strong decrease with maturation, which underlines the downregulation of the aldehydic-cycle in mature developed stages and is fairly in accordance with our hypothesis. Mau et al., (1997) Fig. 4 Hypothetical C8-oxylipin pathway in Basidiomycota. The oxygenation of linoleic acid by lipoxygenases/dioxygenases (LOX, DOX) is considered as the initial step. Subsequentally, the hydroperoxy-fatty acids (HPOD) are cleaved by putative hydroperoxide lyases (HPL) through an unknown mechanism which could lead to two distinct biocatalytic cycles with oct-1-en-3-one (7) and oct-2-enal (3) as precursors. Further modifications via ene-reductases (ER), double bond reductases (DBR), aldo-keto reductases (AR) and alcohol dehydrogenases (ADH) could lead to the variety of known C8-oxylipins compared the octan-3-ol (5), oct-1-en-3-ol (6), octan-1-ol (2) and oct-2-en-1-ol (1) levels in liquid extracts from aging fruiting bodies of V. volvacea. Amounts of ketonic-cycle related C8-oxylipins, particularly oct-1-en-3-ol (6) increased with maturity. This indicates a higher activity of the ketoniccycle in more developed fruiting bodies. In contrast, aldehydic-cycle related C8-oxylipins were detected overall in lower amounts, which did not undergo any significant change in their composition. Moreover, Cho et al. (2006) analysed different maturity grades of fruiting bodies from T. matsutake that showed an increase in the overall level of ketonic-and aldehydic-cycle related C8-oxylipins. Yet, higher relative peak areas of the ketonic-cycle related C8-oxylipins were identified in all developmental stages conceivably induced by a higher biocatalytic activity of the ketonic-cycle. Nevertheless, it has to be mentioned that liquid extracts from T. matsutake and not the headspace was analysed. Studies by Kleofas et al., (2015) and Tressl et al., (1982) demonstrated that liquid extracts of fruiting bodies of Calocybe gambosa and Agaricus campestris contained strikingly more ketoniccycle related C8-oxylipins compared to the aldehydic-cycle, which supports our hypothesis of an increased respectively high activity of the ketonic-cycle in mature developmental stages.
It is noteworthy that the total amount and number of C8-oxylipins decline with maturation ( Fig. 1). Although fatty acids are involved in many biological functions, the differences in the fatty acid composition, especially the steady decline of linoleic acid contribution at more mature developmental stages and the decrease of C8-oxylipins is noticeable (Table 1). Furthermore, we were able to demonstrate that the fatty acid composition of basidiospores consists of significantly higher ratios of SFA and MUFA, which could be explained by their differing role in morphological and physiological stages ( Table 1).

Conclusion
Although further research in C8-oxylipin biosynthesis is required, we were able to show that two morphology dependent and distinct biocatalytic cycles seem reasonable by combining metabolomic analysis. The aldehydic-cycle appears to be more active in early developmental stages, whereas the ketonic-cycle is active throughout all developmental stages, peaking at the stage of sporulation. The existance of distinct pathways was underlined with biotransformation experiments using the putative precursor of the aldehydic-cycle oct-2-enal (3) and the precursor of the ketonic-cycle oct-1-en-3-one (7). With the addition of each precursor, a drastic increase of the corresponding aldehydicor ketonic-cycle related C8-oxylipins was detected, which supports our hypothesis. Nonetheless, genes encoding for putative LOX, DOX as well as the modifying enzymes of shortened oxylipins have to be the focus of further studies. included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.