Combining UV Irradiation and Alkaline Deacetylation to Obtain Vitamin D- and Chitosan-Enriched Fractions from Shiitake Mushrooms (Lentinula edodes)

Shiitake mushrooms (Lentinula edodes) contain compounds with hypocholesterolemic and immune-modulatory activities such as ergosterol and chitin that can be partially transformed into vitamin D and chitosan to yield extracts with enhanced bioactivities. A method was optimized to increase vitamin D2 levels by irradiating a non-soluble fraction obtained from shiitake mushrooms during 1 h with UV-light (254 nm) at 50 °C in methanol. After 1 h, almost all of the ergosterol was transformed into vitamin D2. The conversion ratio can be simply adjusted by reducing the irradiation time. A deacetylation treatment with 50% NaOH for 24 h at 95 °C was needed to generate chitosan and other water-soluble lower molecular weight derivatives from chitin. To obtain an extract enriched in both compounds, the defined methods can be combined, but the extracts should be firstly deacetylated and later irradiated. In this order, an extract is obtained containing vitamin D2 (4.65 mg/g) and chitosan (2.83%).


Introduction
Edible mushroom consumption has increased worldwide in the last decades mainly because, together with the Asiatic traditional large acceptance, Western consumers are now also integrating them in their usual menu. The extensive understanding that the scientific community is reaching about their composition and their potential benefits on human health is probably behind this trend (Chugh et al., 2022;El-Ramady et al., 2022;Roncero-Ramos & Delgado-Andrade, 2017).
Moreover, vitamin D 2 (ergocalciferol) can be found in shiitake and other species, although in highly variable contents in wild mushrooms and extremely low or undetectable levels in cultivated mushrooms (Mattila et al., 2002;Teichmann et al., 2007). However, these levels can be enhanced by UVlight action since ergosterol (only present in Fungi kingdom), the main sterol of hyphal membranes, is photoconverted into vitamin D 2 (Vallespir et al., 2019). This transformation can be stimulated not only by direct exposure of fruiting bodies but also by irradiating specific extracts or fractions Morales et al., 2017;Taofiq et al., 2017), and this procedure might be crucial to offer mushrooms or fungal products as an alternative vitamin D 2 source, considering the deficient levels noticed nowadays in Western population. Vitamin D plays a key role in many metabolic pathways, being involved not only in calcium homeostasis and bone system regulation, but also in immune-system modulation, neuronal and cardiovascular health, besides other biological functions (Janoušek et al., 2022). Its deficiency, among other disorders, is linked to infectious and autoimmune diseases, since it is strongly correlated with immune cells, being expressed in monocytes/macrophages, T, B, and dentritic cells together with 1α-hydroxylase, an enzyme that activates vitamin D (Ao et al., 2021). Furthermore, low levels of this molecule reduce vitamin D receptor activity, triggering an increase in the sterol regulatory element-binding protein 2 (SREBP2) activity and the 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) expression, with the consequent rise in cholesterol biosynthesis rate (Li et al., 2016).
Furthermore, chitins are present in fungal cell walls (approximately 2-10% in edible mushrooms), and these polysaccharides, considered as dietary fibres, can be subjected to chemical deacetylation (usually applying hot alkaline procedures) to obtain chitosan, another interesting polymer because of its hypocholesterolemic activities particularly when a high deacetylation degree is achieved. The positive charges of chitosan promote bile acids scavenging stimulating cholesterol transformation to restore their levels. Moreover, chitosan and other dietary fibres can also increase intestinal viscosity impairing cholesterol absorption (Silva et al., 2021). Regarding chitosan immune-modulating ability, it has shown potent immune-stimulatory effects by activating antigen-presenting cells and inducing cytokine secretion (Li et al., 2021) Although chitosan presence has been described in mushrooms, its content was significantly lower than chitin concentrations, particularly in shiitake mushrooms, being even undetectable in many cases (Chien et al., 2016;John Kasongo et al., 2020;Morales et al., 2019a, b and c). Apart from the mentioned physiological significance, chitosan has been applied as food packaging, coating or films, taking advantage of its antimicrobial potential and for preservation purposes (Amjadi et al., 2019;Lago et al., 2014;Viacava et al., 2022).
In a previous work, a dietary fibre extract containing shiitake β-glucans, chitins and ergosterol reduced the cholesterol levels in mice serum after 9 weeks of oral administration (Morales et al., 2019c). However, when a clinical study was conducted on hypercholesterolemic subjects, no significant reduction of total or LDLcholesterol levels was observed when compared to a placebo group after 8 weeks intaking. Moreover, no changes were noticed on the secretion of specific cytokines (IL1-β, IL-6 and TNF-α). The extract only modulated the gut microbiota promoting the growth of beneficial microorganisms (Morales et al., 2021). Therefore, in this work, a procedure to convert part of the ergosterol into vitamin D 2 and chitins into chitosan in shiitake fractions was explored using UV irradiation and alkaline deacetylation as an attempt to increase the effectivity of the extract as hypocholesterolemic and immunemodulatory functional ingredients. To our knowledge, this is the first time than a mushroom extract is simultaneously enriched in vitamin D 2 and chitosan.

Extraction of the NSF
The non-soluble fraction (NSF) fraction was prepared according to Morales et al. (2019a). Briefly, shiitake powder (SP) was placed into a Nylon mesh strainer and immerse into hot demineralized water (98 °C) during 1 h with a mushroom to solvent ratio of 60 g/L. After draining the liquid phase from the extraction chamber, the insoluble material that remained in the strainer bag (NSF) was freeze dried until further use (Fig. 1). The samples were obtained in triplicate.

UV Irradiation of NSF
This procedure was carried out according to Morales et al. (2017) with slight modifications. Powdered NSF (16.7 mg/ mL) was mixed with methanol, methanol-water solutions or placed dry in a cylindrical vessel, and exposed uncovered to the radiation (λ = 254 nm) under vigorous shaking at 4 cm of distance for 15, 30, 60 or 120 min using a VL-4.LC lamp from Vilber Lourmat (Eberhardzell, Germany). The methods were carried out in triplicate.

Hot Alkaline Deacetylation of NSF Chitins
Powdered NSF was treated, as described by Wu et al. (2004). Briefly, NSF (1:40 w/v) was suspended in 50% NaOH (w/v) and vigorously stirred and refluxed at 95 °C during 1, 2, 3, 4, 5, 8, 12 or 24 h. Afterwards, the suspension was centrifuged at 8000 rpm, 20 min, 20 °C and washed with water and filtered until pH 7 was reached. Then, the neutralized material was freeze dried until its use for chitin and chitosan isolation. The procedures were carried out in triplicate.

Combination of UV Irradiation and Hot Alkaline Deacetylation
A sequential combination of both treatments was also evaluated following two different possibilities, as indicated in Fig. 1. Thus, the NSF fraction that was subjected to the previously described UV irradiation protocol at 50 °C in methanol during 1 h, was dried using an IKA RV 10 rotaryvacuum evaporator (IKA-Werke, StauGermany) and, subsequently, UV-NSF followed the hot alkaline deacetylation protocol during 24 h, leading to UV-DA-NSF. Similarly, the order was inverted to obtain DA-UV-NSF, by previous freeze-drying of DA-NSF before UV treatment. The procedures were carried out in triplicate.

Determination of Ergosterol and Vitamin D 2
Ergosterol was extracted from the samples and quantified following the procedure described by Gil-Ramírez et al. (2013). The unsaponified fractions were injected into an Agilent 19091S-433 capillary column (30 m × 0.25 mm ID and 0.25-μm phase thickness). The column was connected to a 7890A System gas chromatograph (Agilent Technologies, Santa Clara, USA) including a G4513A autoinjector and a 5975C triple-axis mass spectrometer detector. All conditions (injector and detector parameters and column temperature program) were those described by Gil-Ramirez et al. (2013). Ergosterol (RT = 12.6 min) was used as standard, and hexadecane (10% v/v) was utilized as internal standard.
Vitamin D 2 was extracted and quantified, according to the protocol explained by Tejedor-Calvo et al. (2019). The unsaponified fractions (5 mg/mL) were dissolved in the mobile phase (95% methanol v/v) and injected (10 μL) into a C18 Spherisorb OD52 4 × 250-mm analytical column with a 5-μm particle size (Waters, Missisagua, Ontario, Canada) coupled to a HPLC system (ProStar 330, Varian, Madrid, Spain) with a photodiode array (PDA) detector (ProStar 363 module, Varian, Madrid). The flow rate was settled at 1 mL/ min and the detection wavelength at 265 nm. Ergocalciferol was used as standard (RT = 16 min) for quantification.

Determination of Chitin and Chitosan
Chitin and chitosan were isolated from the samples following the method described by Wu et al. (2004) with slight modifications. Briefly, the freeze-dried material obtained after deacetylation was mixed with 2% acetic acid (v/v) and vigorously stirred and refluxed at 95 °C (1:100 w/v) during 6 h. Then, the mixture was centrifuged (8000 rpm, 20 min and 20 °C), obtaining a chitin-enriched pellet and a chitosanenriched supernatant. The supernatant was adjusted to pH 10 using 10 M NaOH to precipitate chitosan. Both chitin and chitosan fractions were neutralized by washing with water and subsequently freeze-dried.
Chitin and chitosan quantification was determined, according to Smiderle et al. (2017). Samples were hydrolysed with 6 M HCl at 100 °C for 1 h and adjusted to pH 10 after cooling down. Later on, hydrolysed mixtures (250 μL) were treated, following Rementeria et al.'s (1991) procedure. Absorbance was measured at 530 nm and a glucosamine hydrochloride standard curve was used for quantification.

Statistical Analysis
Differences were evaluated at 95% confidence level (P ≤ 0.05) using a one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. Statistical analysis was performed using GraphPad Prism version 9.3 (GraphPad Software, San Diego, CA).

Enhancing the Vitamin D 2 Levels of the NSF Fraction
The NSF fraction contained significantly higher ergosterol values (4.83 mg/g) than the initial SP; its levels were 2.1fold increased since sterols were not soluble in hot water (SF) ( Table 1). Vitamin D 2 was not detected neither in NSF nor in the raw SP. The lack of this metabolite is common for cultivated mushrooms including L. edodes, as indicated by several authors (Lee et al., 2009;Morales et al., 2017;Won et al., 2018), since they are usually cultivated indoors in growing chambers with artificial dim lighting or in shaded rooms. Vitamin D 2 biosynthesis can be stimulated in wild shiitake and cultivated mushrooms by natural light irradiation, and their levels increase from 0.2 to 1.1 μg/g (Mattila et al., 2002), but these ranges are still negligible, compared to artificial UV irradiation. Some reports indicated increases from undetected to 29.9 μg/g in fresh shiitake mushrooms and from 2.8 to 106.4 μg/g when sliced shiitake was placed with their gills toward the light (Ko et al., 2008;Won et al., 2018). However, the most effective conversion was achieved when mushroom extracts and not their fruiting bodies were exposed, since in principle all ergosterol molecules could be transformed into ergocalciferol . Thus, taking into consideration these previous results obtained with shiitake fractions (extracted with supercritical fluids), different parameters were tested to optimize an irradiation method for NSFs.
Firstly, dry irradiation was tested by exposing the shiitake powder directly to the UV lamp and compared to irradiation of NSF while stirred in different water:methanol mixtures. The irradiation time was set at 1 h, considering previous works, and significant differences were obtained in both ergosterol and vitamin D 2 contents (Fig. 2). Results indicated that 100% methanol was required to reach the almost complete ergosterol conversion into vitamin D 2 (0.29 mg/g ergosterol; 4.78 mg/g vitamin D 2 ). Dry or methanol concentrations lower than 25% were ineffective, and 50-75% methanol irradiation led to significantly lower values than when NSF was suspended in absolute methanol. This effect was previously noticed when the oily extracts obtained with supercritical fluids (more soluble in methanol than NSF) were irradiated . However, although the fibrous NSF showed poor solubility in this organic solvent, it did not prevent methanol extraction of ergosterol molecules enhancing its transformation. With a higher methanol proportion, more ergosterol is solubilized, and once in the media, its conversion into ergocalciferol seemed to be facilitated.
Another factor influencing the ergosterol transformation level was irradiation temperature Won et al., 2018). Then, the NSF was warmed up during irradiation from 25 to 50 °C and indeed, the conversion rate was enhanced finding significant differences already after 30 min treatment. Both ergosterol (Fig. 3a) and vitamin D 2 (Fig. 3b) contents concomitantly changed and after 1 h, almost all of the ergosterol was converted into vitamin D 2 when exposed at 50 °C, while it required almost 2 h at 25 °C, confirming that mild heating speeded up the process (Won et al., 2018). This beneficial effect seemed to be related to an improved ergosterol solubilisation out of the NSF. Higher temperatures were not tested, since the synthesis or other intermediate products, photoisomers or side products such as lumisterol or tachysterol can be stimulated reducing vitamin D 2 yield Wittig et al., 2013;Won et al., 2018).
Thus, the optimal method to increase the levels of vitamin D 2 in the NSF fraction was by suspending the material Table 1 Ergosterol, vitamin D 2 , chitin, and chitosan levels of SP, SF, NSF, UV-NSF, DA-NSF, UV-DA-NSF and DA-UV-NSF fractions (described in Fig. 1) nd non-detected a-c Different letters denote significant differences (P < 0.05) between values of the same compound  (Ko et al., 2008;Won et al., 2018). Therefore, these parameters were selected for further experiments.

Enhancing the Chitosan Levels of the NSF Fraction
The NSF contained higher amounts of chitins than ergosterol doubling the levels found in the initial SP (Table 1), since these fibres are not soluble in hot water. Only traces of chitosan were detected in both SP and NSF, and these low concentrations were also noticed before for shiitake and other mushroom species (Chien et al., 2016;John Kasongo et al., 2020;Savin et al., 2020). Therefore, the NSF was subjected to a deacetylation method using a hot alkaline medium to release acetyl groups of chitins. Results indicated that a long treatment is needed to obtain an approximately 27% conversion being less effective than the ergosterol transformation (Fig. 4). Incubations during 5 and 8 h yielded, respectively, 1.63 and 1.86% chitosan, but after more than 12 h, the yields were only slightly increased up to 2.78% indicating that approximately 54% chitin (perhaps also part of the generated chitosan) was transformed into chito-oligosaccharides or the so-called low molecular weight derivative products (Morales et al., 2019b;Palanisamy et al., 2014). These chitin/chitosan derivatives showed lower polymerization degree, a lower molecular size and, therefore, they might not be recovered in the obtained chitins or chitosan fractions.
Besides chitosan, the partially deacetylated fraction obtained after 24 h (DA-NSF) still contained high ergosterol levels after such a strong treatment (Table 1) and, therefore, it was selected for further studies.

Enhancing Vitamin D 2 and Chitosan Levels of the NSF Fractions by Combining Treatments
UV irradiation and hot alkaline deacetylation methods were sequentially combined to generate extracts with both vitamin D 2 -and chitosan-enhanced levels. The sequence order was investigated to avoid target compound degradation and also to explore potential synergies. When first the NSF was treated with UV light and afterwards submitted to deacetylation (following sequence 1, Fig. 1), results indicated that UV irradiation did not significantly affect the initial chitin or chitosan content (Table 1); NSF and UV-NSF showed the same levels. It was previously indicated that UV light contains sufficient energy to cleavage covalent bonds, form free radicals and eventually initiate degradation reactions (Wasikiewicz et al., 2005); however, the intensity, wavelength and duration of the utilized method seemed to not be detrimental for the NSF chitinous material. Besides, the moderate temperature chosen (50 °C) did not compromise the polymers' stability that requires higher temperatures to be at risk (de Britto & Campana-Filho, 2007). Furthermore, hot alkaline treatment, necessary to transform chitin into chitosan generating UV-DA-NSF, used high temperature (95 °C), NaOH concentration (50%) and duration (24 h), but it did not significantly reduce ergosterol levels, as compared to UV-NSF (or DA-NSF to NSF). This lipid showed relatively good stability, as it can be extracted using high temperatures or saponified with strong alkalis Taofiq et al., 2019). However, vitamin D 2 content of UV-NSF was drastically reduced with the deacetylation method yielding a UV-DA-NSF with only 4.4% of the generated vitamin. Apparently, when powdered ergocalciferol was submitted to temperatures above 40 °C, this vitamin was degraded into products of higher polarity (Grady & Thakker, 1980). However, when it was dissolved in methanol and irradiated at 50 °C, no significant losses were noticed (Fig. 3b); on the contrary, the reduction of ergosterol levels and increasing of ergocalciferol levels were simultaneously occurring. Nevertheless, 95 °C heating seemed to be too high temperature for a thermolabile vitamin.
Considering the results, another combination was tested (sequence 2, Fig. 1) by enhancing first the deacetylation of NSF, since ergosterol was heat-resistant and chitosan or chitins by UV-light treatment did not affect the generated chitosan. The resulting DA-UV-NSF showed the same chitin and chitosan ratio than DA-NSF and the same ergosterol and ergocalciferol levels than UV-NSF, being then considered as the proper order to treat the NSF fraction to obtain an extract with improved contents of both vitamin D 2 and chitosan (4.65 mg/g and 2.83%, respectively). Unfortunately, no synergistic effects of the combination of both methods were noticed.

Conclusion
Vitamin D 2 -and chitosan-enriched fractions can be successfully obtained (individually or combined) from shiitake mushrooms by subjecting the extract that is not soluble in hot water to deacetylation with 50% NaOH for 24 h at 95 °C, followed by UV irradiation (at 254 nm) during 1 h, at 50 °C when the deacetylated extract is dissolved in 100% methanol. With this sequence, the ergosterol was almost completely transformed into vitamin D 2 , while chitin was not totally converted into chitosan, leading to the formation of soluble derivatives with smaller molecular size. To retain part of the ergosterol in the extract, less irradiation time should be adjusted. The reported results, as well as the increasing demand of novel edible products with high content of the studied compounds, encourage further investigation to scale up the production of this fraction and to evaluate its bioactive potential, particularly targeting immune-modulatory and hypocholesterolemic activities. Thus, future challenges must be linked to not only the study of efficient processing of high amounts of raw materials but also the evaluation of the biological activities in vitro and in vivo (animal and clinical trials) to assess the bioavailability and mechanism of action of the real impact on health of the bioactive molecules.
Author Contribution Diego Morales: investigation, methodology, formal analysis, writing of the original draft and figures; Adriana Jiménez Piris: investigation and methodology; Alejandro Ruiz-Rodríguez: supervision and data curation; Cristina Soler-Rivas: supervision, validation, data curation, funding acquisition, writing and reviewing.
Funding This research was supported by the national R + D program from the Spanish Ministry of Science and Innovation (project AGL2014-56211-R) and the regional program from the Community of Madrid, Spain (S2013/ABI-2728). D. M. received from the Spanish Ministry of Science and Innovation a Juan de la Cierva Formación scholarship (FJC2020-044585-I).
Data Availability Research data are not shared.

Conflict of Interest
The authors declare no competing interests.