The majority of chiral natural compounds are present in only one specific enantiomeric form, as they are products of enzymatic reactions. Chiral center(s) present in a molecule of a natural optically active compound can be a base for a number of syntheses of optically active compounds. For example, d-glucose is a substrate in the syntheses of, among others, l-ascorbic acid (vitamin C) [12] and optically active 1-deoxynojirimycin [13]. The current article discusses the synthesis of flavoring lactones from lipids as well as selective degradation of the sterol side chain to C-19 or C-22 product(s). The pharmaceutical industry has expressed interest in selective sterol side chain degradation, because this process results in formation of steroid intermediates which can be used as building blocks for further synthesis of bioactive steroids of medicinal use.
2.1.1. Production of Flavor-Active Lactones
Use of biocatalysts in the production of flavoring compounds structurally identical with those present in natural sources is preferred, because of, among others, formal (regulatory) reasons. According to the U.S. and European regulations (e.g., CFR 1990 and EEC 1334/2008), compounds isolated from natural resources, or obtained in microbial or enzymatic processes involving precursors isolated from nature, are classified as “natural”. Flavoring compounds found in nature, but obtained by chemical routes, are referred to as “nature-identical”. Consumer preferences reflecting the trends towards “healthy lifestyle” decide that the vast majority of fragrance food additives are compounds classified as “natural”. Lactones are a family of flavors and fragrances that are affected by such a trend. They are widely distributed in food, beverages and cosmetic preparations. Among these compounds one can find decalactone (4-decanolide), which provides a suitable characteristic aroma of peaches, apricots or strawberries. The price of γ-decalactone was 20 thousand US$/kg before the development of biotechnological methods of its synthesis. The possibility of producing this lactone from natural oils reduced the price of this product down to 1.2 thousand US$/kg (1986) and after the optimization of the fermentation, there was a further reduction of this price to 500 US$/kg (1998) and ca 300 US$/kg in 2004 [4,14].
A general metabolic pathway for the production of lactones from hydroxy fatty acids consists of the β-oxidation cycle and intramolecular esterification. Depending on the position of the hydroxyl group of the carboxylic acid, lactonisation can give γ-, δ- or ɛ-lactones. It is worth mentioning that lactones obtained by biocatalysis have a very high optical purity—their stereochemistry is determined by chirality of the asymmetric centers of their hydroxy fatty acid precursors.
Studies on the development of the efficient synthesis of γ-decalactone by biocatalysis were started by selecting a strain capable of conversion of ricinoleic acid (or its methyl ester) to 4-hydroxydecanoic acid [15], which can be lactonised spontaneously under acidic conditions (pH 2) and/or after heating. Ricinoleic acid is one of only a few natural hydroxy fatty acids available in large quantities. The hydrolysis product of castor oil contains about 90% of this acid. The transformation of ricinoleic acid to 4-hydroxydecanoic acid includes four β-oxidation cycles; at each cycle a two-carbon-shorter metabolite—SCoA and an acetyl-CoA are produced (Figure 2).
Many processes using the bioconversion of ricinoleic acid to produce γ-decalactone have been patented (see in [16]). The oxidation of ricinoleic acid is conducted by microorganisms belonging to Yarrowia, Saccharomyces, Candida, Rhodotorula, Sporobolomyces, Monilia, Pichia, Aspergillus, Mucor and Cladosporium genera. The processes with the highest product concentrations use Candida sorbophila[17] and Yarrowia lipolytica strains [18,19]. The conversion of ricinoleic acid by Candida sorbophila can produce about 50 g/L of γ-decalactone [17]. The maximum production of γ-decalactone by Yarrowia lipolytica, in the optimum conditions of agitation and aeration, was more than 11 g/L in less than 70 h [19].
There are two main reasons for commonly much lower (up to 4–5 g/L) yields of γ-decalactone. The first one is the degradation of newly synthesized lactone by the producing yeast. The second one is only a partial use of ricinoleic acid or intermediate at the C10 level, which is simultaneously the precursor for other γ-lactones (3-hydroxy-γ-decalactone and the products of its dehydration—dec-2-en-4-olide and dec-3-en-4-olide) (Figure 2).
The β-oxidation loop in the mitochondrial catabolism of fatty acids is theoretically repeated until the complete degradation of the substrate. In yeast, peroxisomal β-oxidation does not proceed via channelization, and at each stage, the metabolite can be hydrolyzed and accumulated depending on the pool of free CoA and acetyl-CoA, the pool of necessary cofactors or the saturation of the next enzyme of the cascade [21]. Thus, the amount of produced lactone is a result of competition between different reactions. The research on the improvements in the productivity of γ-decalactone focuses on the role of the genes that encode the isozymes of acyl-CoA oxidase (Aox1 to 6, encoded by POX1 to 6[22]) catalysing the first step of β-oxidation. Aox is generally considered as the fundamental rate-limiting enzyme. The best-known ones are the Aox isoenzymes from Yarrowia lipolytica. The specific role of each acyl-CoA oxidase was presented in several reviews [14,21]. It was confirmed that short-chain-specific Aox3 is involved in the oxidation after the C10 level and the disruption of the POX3 gene decreases lactone degradation [23,24]. Aox4 and Aox5 are non-chain-length-specific acyl-CoA oxidases and their activity is weak, albeit directed towards the wide range of substrates, whereas Aox1 is inactive [25]. The long-chain-specific Aox2 was significant for conversion of ricinoleic acid and hence for the production of γ-decalactone. Deleting all the POX3–5 genes resulted in an increased accumulation and an inhibition of γ-decalactone degradation [22,26]. The designed mutant produced 10 times more lactone than the wild type, and its growth was only slightly altered in comparison to the native strain. Recently, a recombinant of the diploid strain Y. lipolytica, with expression of POX2 gene and disruption of POX3 genes on two chromosomes (but without disruption of POX4 and POX5 genes) was constructed, and this mutant could be grown in the continuous fermentation of methyl ricinoleate. Compared with the wild type, the production of γ-decalactone was increased 4-fold, and there was no re-consumption of the product. It could be concluded that Aox2’s positive effect had a greater influence than the Aox3’s negative action to the γ-decalactone production [27].
Another problem is the modification of β-oxidation flux, which allows a shift in the equilibrium between production of γ-decalactone and production of 3-hydroxy-γ-decalactone. It can however be achieved by decreasing the Aox2 and Aox3 activity. For a mutant with disrupted POX2 and POX3 genes the production of hydroxylactone was minimized [14,21,24]. It was confirmed that accumulation of 3-hydroxy-γ-decalactone occurs when the amount of oxygen is lowered [20,21]. Low aeration conditions (e.g., during cell growth) resulted in low 3-hydroxy-acyl-CoA dehydrogenase activity, because its cofactor regeneration (NAD+) is not sufficient (Figure 2). This cofactor is regenerated through a shuttle mechanism, which probably depends on mitochondrial respiration. Under the hypoxiation regime, a lack of oxygen stopped the fluxes of the oxidation cascade at earlier stages and causes the accumulation of γ-decalactone. 3-Hydroxy-γ-decalactone is the precursor of two decenolides with flavoring properties (Table 1). Although both decenolides are characterized by interesting sensory properties, they are not commercially produced because of the lack of simple methods for their separation [14].
Apart from β-oxidation activity, another factor influencing the yield is the toxicity of the γ-decalactone and its C10-precursor against the producing strains. The mechanism involved in this toxicity is linked with the fact that the side-chain of the lactone interacts with cellular membranes, increasing their fluidity and decreasing their integrity [28]Sporidiobolus strains are particularly sensitive to the presence of γ-decalactone [29]. A number of strategies were developed to reduce the cytotoxity of γ-decalactone (e.g., adsorption on porous hydrophobic sorbents, inclusion in β-cyclodextrins, utilization of immobilization, addition of natural gum, surfactant and natural inert oils—mainly hydrogenated coconut oil or a mixture of tripalmitin, tristearin, triolein). For more details we refer the reader to the recent reviews and research articles [14,21,27,30].
Besides γ-decalactone, many other lactones can be produced by biotechnological processes (Table 1). Their price depends on the availability of the corresponding hydroxy fatty acids. Moreover, microorganisms can also perform the step for the introduction of the hydroxyl group, if adequate substrates (hydroxyacids) are not available.
(11S)-Hydroxypalmitic acid (jalapinolic acid) and (3S,11S)-dihydroxymyristic acid (ipurolic acid) can be extracted from a jalap resin or sweet potato for the purpose of their use in producing natural (S)-δ-deca- and (S)-δ-octalactone, respectively [31]. (R)-δ-Decalactone can be obtained by the β-oxidative degradation of (13R)-coriolic acid, which is the major fatty acid of the seed oil of Coriaria nepalensis, whilst (S)-δ-decalactone can be produced from (13S)-coriolic acid of Monnina emarginata seed oil [32] (Figure 3).
Industrially useful lactones can be readily produced in large amounts from inexpensive fatty acids. γ-Dodecalactone is obtained from oleic acid in a process involving two biocatalysts [33,34]. A Gram-positive strain of bacteria catalyzed the conversion of oleic acid to (10R)-hydroxystearic acid, which was subsequently oxidized by bakers’ yeast to (4R)-hydroxydodecanoic acid (Figure 4). After cyclization, (R)-γ-dodecalactone was obtained in over 22% yield with respect to the initial substrate, oleic acid [33].
Many studies have focused on 10-hydroxystearic acid production from oleic acid. Recently, it has been reported that whole cells of recombinant Escherichia coli containing oleate hydratase from Stenotrophomonas maltophilia produced 49 g/L of 10-hydroxystearic acid in 4 h, with a conversion yield of 98% and a volumetric productivity of 12.3 g/L/h—the highest thus far among cells, which can make a significant contribution in the industrial synthesis of this hydroxyacid [35].
Linoleic acid and α-linolenic acid can be converted using microorganisms belonging to the genus Pediococcus or Bifidobacterium into 13-hydroxy-9-octadecenoic and 13-hydroxy-9,15-octadecadienoic acids, which are good substrates for the production of δ-decalactone and δ-jasminlactone by microorganisms belonging to the genus Kluyveromyces or Zygosaccharomyces, respectively [36] (Figure 5).
Using different vegetable oils containing hydroxylated or non-hydroxylated fatty acids, it is possible to produce 6-pentyl-2-pyrone—an unsaturated δ-lactone with a strong odor of coconut [37,38] (Figure 6).
The biosynthesis of this lactone from linoleic acid in the Trichoderma species is initiated by a lipoxygenation at C-13 of the fatty acid with the formation of 13-hydroperoxide. The lipoxygenase reaction is followed by β-oxidation and isomerization to form 5-hydroxy-deca-2,4-dienoic acid. Low activity of the NADPH-dependent 2,4-dienyl-CoA reductase is essential for achieving good performance and yields of 6-pentyl-2-pyrone.
2.1.2. Microbiological Degradation of the Side Chain of Sterols
Steroids secreted in animals by the sexual organs and adrenal cortex (e.g., sex hormone—testosterone) play a critical regulatory role in development of reproductive structures, they also influence many aspects of metabolism and immune function (hydrocortisone—glucocorticoids), help maintain blood volume and control renal excretion of electrolytes (mineralocorticoids). Steroid compounds have a wide therapeutic activity spectrum, namely anti-inflammatory, anti-allergy, anti-fungal, anti-viral, anti-tumor, neuroprotective and immunosuppressive. They are used as agents for prevention and therapy of some disorders, for example neurodegenerative diseases, hormone-dependent breast and prostate cancer, colon cancer; metabolic, cardiovascular and central nervous system disorders (Figure 7). Steroids, frequently found among the most marketed medical products, constitute the second largest group of pharmaceuticals, next to antibiotics [39].
Availability of raw materials is one of the fundamental issues related to the production of steroid drugs. The majority of hormones and steroidal drugs are synthesized by semi-synthesis using natural steroids as starting material. Sapogenins and sterols produced in large scale by plants are the raw material in the production of steroidal drugs. Sapogenins (which include diosgenin, hecogenin, solasodine) and stigmasterol (a sterol with the double bond in the side chain) are chemically transformed to C21 products (16-dehydropregnenolone, progesterone), which are convenient intermediates in the synthesis of steroidal pharmaceuticals [40] (Figure 8).
The limited and unstable supply of diosgenin (for many years the major raw material in steroid production), together with the increase in the production of steroidal pharmaceuticals, were impulses to carry out research on transformations of (phyto)sterols with a saturated side chain (β-sitosterol, campesterol, cholesterol) to give valuable steroidal pharmaceutical intermediates (Figure 9). Phytosterols are the major raw materials resulting from the processing of vegetable oils, from sugarcane, and from the paper industry. A deodorizate waste product of vegetable oil processing can contain over 95%, while tall-oil effluent of a paper pulp industry—over 80% of phytosterols—mainly β-sitosterol, stigmasterol, campesterol [41].
In sterol molecules there is a double bond at C-5 and a hydroxyl group at the 3β-position, which can be easily converted to a 4-en-3-oxo moiety, present in the majority of steroids used in clinical settings [40] (Figure 7); unfortunately, attempts failed to develop effective methods for chemical degradation of the 17β-aliphatic side chain of sterols [42]. It was rational to assume that, due to the selectivity of enzymes, it would be possible to carry out selective degradation of the aliphatic side chain of sterols by microbial transformation. A number of strains of Gram-positive (Rhodococcus, Mycobacterium, Streptomyces, Brevibacterium) and Gram-negative (Pseudomonas, Comamonas, Burkholderia, Chromobacterium) bacteria can use sterols as the sole carbon source [43] (Figure 10). CO2 and H2O are the final products of microbial degradation of sterols, but among the metabolites there were identified synthetically useful C19 steroids: androsta-4-en-3,17-dione (AD), androsta-1,4-dien-3,17-dione (ADD), 9α-hydroxy-androsta-1,4-dien-3,17-dione (9α-hydroxy-ADD), or the C22 products: 20-carboxy-pregna-4-en-3-one, 20-carboxy-pregna-4,17(20)-dien-3-one. Also 3aα-H-4α(3′-priopionic acid)-7aβ-methylhexahydro-1,5-indanedione (HIP)—the product of partial degradation of steroidal skeleton—is the starting compound for the synthesis of medically important steroids, such as 19-norsteroids [44] (Figure 10).
During microbiological transformation, cleavage of the aliphatic side chain of sterols occurs as a result of β-oxidation of an acid containing the carboxyl group at C-26; the said acid is formed in two subsequent redox reactions: hydroxylation at C-26 and oxidation of the introduced hydroxyl group to the carboxyl function [42]. AD is obtained as the result of oxidative degradation of the side chain of cholesterol after subsequent cleavage of propionyl-CoA, acetyl-CoA and propionyl-CoA, respectively. In the process of degradation of the C-24 branched sterols (β-sitosterol, campesterol) to AD, the C-28-carboxylation additionally occurs and there are released two molecules of propionyl-CoA, then acetyl-CoA, and finally propionyl-CoA; initial stages of degradation of the side chain of C-24 branched sterols and cholesterol are analogous (Figure 11). After formation of enoyl-CoA, before the first propionyl-CoA molecule is released, carboxylation of the allylic position (C-28) occurs. Subsequently, the product of hydration of the double bond releases propionyl-CoA in the course of a retro-aldol reaction. After CoA activation of the simultaneously formed β-keto-C26-oic product, its side-chain degradation occurs to AD, as in cholesterol [45] (Figure 11).
Many research groups experimented on selecting a strain able to carry out selective degradation of the sterol side chain to product(s) with a conserved steroid nucleus. Hundreds of microorganisms were tested, some among them efficiently metabolized sterols, but none of the studied strains converted a sterol to either a C19 or C22 steroid as a major product. Studies on microbiological degradation of sterols established that during transformations by wild-type strains, besides oxidative degradation of the side chain of the sterol, also oxidation of the steroid nucleus occurs. The key role in the process of steroid nucleus degradation is played by oxidative reactions in the rings A and B: 1,2-dehydrogenation and 9α-hydroxylation. 9α-Hydroxy-androsta-1,4-dien-3,17-dione (9α-hydroxy-ADD) is a structurally unstable chemical, spontaneously reacting by opening of ring B and aromatization of ring A (Figure 10). Inhibition of activity of 3-ketosteroid-9α-hydroxylase (Ksh) or/and 3-ketosteroid-1-dehydrogenase (KstD) is significant in the design of microbiological transformations of sterols leading to industrially valuable intermediates (C19 or C22 steroids).
During transformation of cholesterol by Mycobacterium phlei, in the presence of α,α-dipyridyl, 8-hydroxyquinoline or Ni2+ cations—9α-hydroxylase inhibitors, a mixture of AD and ADD was obtained with a good yield [11]. The accumulation of AD and ADD suggests that the side-chain degradation occurs prior to sterol ring degradation. Due to environmental concerns and production costs, optimal methods of selective degradation of the sterol side chain are such transformations in which the biocatalysts are mutants of wild-type strains with either 9α-hydroxylase or 1,2-dehydrogenase deficit (or both). After isolation of the first mutant of Mycobacterium sp. which performed degradation of sterols to AD, a number of mutants were described and patented [11,46]. These mutants were obtained by chemical or physicochemical mutagenesis of the native strains performing effective degradation of sterols. After identification in 2000 and 2002 of the KstD and Ksh genes responsible for the catabolism of sterols in Rhodococcus erythropolis SQ1 [47–49], rapid progress has been noted in the studies on identification of the genes involved in microbial degradation of sterols [50–52]. In 2007, a wealth of information on sterol catabolism was provided by the discovery of the existence of a gene cluster encoding sterol catabolism in Rhodococcus and Mycobacterium species [51]. Further, a steroid-coenzyme A ligase (FadD19) was identified and characterized as having an essential in vivo role in the degradation of the side chains of C-24 branched-chain sterols in Rhodococcus rhodochrous DSM43269 [52]. Currently, conventional mutagenesis based on random mutations induced by UV/chemical treatments is more and more frequently replaced by targeted disruption of genes encoding either Ksh or KstD to generate highly selective strains [50,53]. Genetic manipulation is potentially able to overcome a problem, significant for industrial applications, connected with the fact that most of the mutant strains in practical use produce AD and ADD simultaneously [46]. A mutant of Mycobacterium neoaurum NwIB-01 after inactivation of the KstD-encoding gene produced mainly AD (AD:ADD 11:1). Another mutant of M. neoaurum NwIB-01, in which the KstD was augmented, proved to be a good ADD-producing strain (AD:ADD 1:50); during transformation by M. neoaurum NwIB-01 a mixture of AD and ADD was formed with a 1:2.5 ratio [54]. An important pharmaceutical androgen—testosterone—can also be the product of microbiological transformation of sterols. Testosterone is obtained from the AD(D) either by a four-step chemical synthesis, or using microbial 17β-reduction by yeasts or fungi [55]. Microbial conversion of sterols to testosterone by using the strains capable of both sterol side chain cleavage and 17β-reduction of AD allows formation of testosterone in one biotechnological step. The strain of Mycobacterium sp. VKM Ac-1815D, capable of sterol side chain degradation and expressing 17β-hydroxysteroid dehydrogenase activity, converted β-sitosterol (5 g/L) to testosterone with 50%–55% molar yield [55].
Ergosterol—a sterol with the 5,7-diene system and a double bond in the side chain—is produced mainly by fungi and phytoplankton. Some strains able to catalyze transformation of β-sitosterol to C19 products were also able to oxidize ergosterol to AD [56,57]. During conversion of ergosterol, besides side chain degradation and transformation of the 3β-hydroxy-5-ene to the 3-oxo-4-ene moiety, hydrogenation of the double bond at C-7 also occurred. Conversion of ergosterol led to much lower yields of AD when compared to the sterol substrates mentioned earlier, therefore reports on microbiological degradation of ergosterol are rare [57]. Products of transformations of ergosterol derivatives (the C3-OH hydrogen atom substituted by a group which is not eliminated under the conditions of the transformation) by wild-type strains can be C22 and/or C19 products with a preserved 5,7-diene system [57,58] (Figure 12).
Transformation of 3β-methoxy-methoxy-ergosterol by the strain Mycobacterium sp. VKM Ac-1815D led to a mixture of 3β-methoxy-methoxy-androsta-5,7-dien-17-one (the major metabolite) and (20S)-3β-methoxy-methoxy-20-hydroxymethyl-pregna-5,7-diene [57]. Modification of the ring A structure of ergosterol has not exerted significant influence on the process of side chain degradation by Mycobacterium sp. VKM Ac-1815D; the major product was obtained with 60% yield, at the 5 g/L substrate concentration. 3β-Hydroxy-androsta-5,7-diene-17-one is a key intermediate for the synthesis of novel vitamin D derivatives. Naturally occurring forms of vitamin D (i.e., 1α,25-dihydroxylated ergo- or cholecalciferol derivatives) can produce hypercalcemia or phosphatemia when administered in pharmacologically relevant doses. For this reason, natural and synthetic analogs of vitamins D2 and D3, called deltanoids, have been actively investigated, and modifications include the rings as well as the side chain of the vitamin [59,60].
Since the 1980s, products of microbiological degradation of sterols have found practical applications [5]; currently, almost half of the starting materials for steroid drug production are obtained by the bioconversion process of sterols [40]. Efforts are continuing to increase the efficiency of the process of microbiological degradation of sterols to valuable steroids. The efficiency of the enzymatic transformation of sterols is limited mainly by the low water solubility of (phyto)sterols and the fact that the final products are toxic to microorganisms. The preferred method of enhancing the bioavailability of phytosterols by improving their solubility is surfactant-facilitated emulsification [61,62]. During the transformation process, sterol uptake occurs by direct contact between microbial cells and the phytosterol particles [61,63,64]. Therefore some microorganisms can use poorly water-soluble hydrocarbons as carbon sources, especially if these microorganisms have the following features: lipophilic cell walls, active transporters and membrane-associated enzymes, as well as the capacity to secrete biosurfactants or bioemulsifiers increasing the availability of hydrophobic compounds [65–67]. The cell walls of actinobacteria contain ca. 60% (dry mass) of the long-chained mycolate acids. These can be modified to form a thick and rigid envelope, which enables the microorganisms to decompose a wide range of hydrophobic compounds [67,68] and to enhance the uptake and bioavailability of phytosterols.
Hydrophobic phytosterols show in water a tendency to form agglomerates, which also limits their bioaccessibility. Addition of synthetic surfactants facilitated emulsification, inhibiting the aggregation of substrates, although the presence of artificial surfactants is usually toxic to bacteria and leads to necrosis and cell lysis [69]. It is known that in natural environments bacteria secrete bioemulsifiers or biosurfactants to make the hydrophobic substrates more available, for example mycolate-containing Mycobacteriun and Rhodococcus, which use phytosterols as carbon sources, synthesize biosurfactants [65]. Pursuing this property might become an alternative to the addition of artificial surfactants.
The mechanism of the cellular uptake of sterols, especially its transmembrane phase, is not yet fully understood, but its understanding would improve rational genetic modifications. Electron microscopy results led to a “flexible multi-component mesophase” model [63] providing a sharp concentration gradient in the mesophase between particles of sterols and the membrane. The model suggests that there should exist channels of proteins, stretching from the cytoplasm to the surface of the cell, capable of sterol binding or transformation, thus enhancing the transport. Further studies [51,70] confirmed the presence of a locus containing 10 genes, conserved in various mycolate-containing actinobacteria, which—if deactivated—resulted in inhibition of the sterol uptake.
The second major factor which limits conversion of phytosterols to steroid products is their toxicity to microbial cells. For example, AD and ADD are considered to inhibit cell growth and inhibit the enzyme activity in the sterol degradation pathway [71]. The tolerance of microorganisms towards toxicity of steroid products is one of the important features which can improve the strains to be able to transform phytosterols, although the mechanisms which allow the strains to stay resistant are still unknown. The improvement of the product yield is carried out by extracting the steroid products from the reaction media (in situ product recovery) and by creating mutants which are tolerant to the steroid products [62,71]. Amberlite XAD-7 resin, dimethyl siloxane and cyclodextrins are used to recover AD and ADD from the reaction media [72], also organic-aqueous two-phase systems are useful for the immediate recovery of the steroid product during the biotransformations [62].
The effective means of solving the problem of toxic products is to develop mutants with improved tolerance to toxicity of the steroids. This can be achieved by an increase in the efflux capability of the cell, or by directed evolutionary mutagenesis towards steroid-tolerant mutants. In the literature there are reports concerning improving the resistance and product yield by adding high amounts of androstanes to the culture after nitrosoguanidine mutagenesis, and further isolation of steroid-tolerant mutants [71].
In view of the fact that currently only the location of genes directly responsible for sterol catabolism is known in Rhodococcus and Mycobacterium, but usually not the function of the encoded proteins [51], the literature suggests that the immediate tasks are characterization of the mechanisms of phytosterol catabolism, phytosterol uptake, tolerance to toxic products, and global regulations involved in the sterol metabolism [46]. This might not be easy using known microorganisms of good performance, because the metabolic processes are complicated and interlaced. Therefore, instead of modifying microorganisms already known to transform phytosterols, the route suggested by Wang et al. is a complete reconstruction of the transformation pathway leading from phytosterols to the desired steroids in a heterologous host organism with suitable physiological trains. This can be achieved in two ways: First, by selecting a robust host which already is superior to the known phytosterol-transforming microorganisms with respect to phytosterol uptake and resistance to the product toxicity. This route would require thorough determination and understanding of the phytosterol metabolism, including degradation of the C-17 side chains. An alternative would be to follow the studies on yeast [73] where it is possible to achieve rerouting of the native biosynthesis of ergosterol to analogous brassicasterol and campesterol. Such reconstitution of the phytosterol transformation system in yeast has the advantage of being based on an already proven route and well-known microorganism and fermentation process, but one of the disadvantages is that yeast is vulnerable to the toxic effects of the steroid products; therefore, intensive research is necessary to utilize industrially this promising route.
A biocatalyst for industrial use should be easily cultured and non-pathogenic. This, unfortunately, is not the case with Mycobacteria which exhibit the overall best performance in transformations. There are however pathogens or opportunistic pathogens, not easily cultured, and slow in transformation (taking even a week). Future investigations might be therefore directed towards the less explored genera Arthrobacter, Bacillus, Brevibacterium, Corynebacterium, Norcardia, Rhodococcus, and others [46]. Experiments are ongoing on the isolation of new, useful biocatalysts from the natural environment, whose catalytic properties differ from currently known cases. An Actinomycete Gordonia neofelifaecis, isolated recently from the faeces of Neofelis nebulosa, carries out the conversion of cholesterol giving ADD as practically the sole product (ca. 90%) [53].
Microbial sterol side chain degradation is also a powerful tool for generation of novel biologically active steroid compounds. Some products of this process and their derivatives are highly cytotoxic or possess immune system stimulating properties, effective against various infections (Figure 13). (20S)-20-Hydroxymethylpregna-1,4-dien-3-one showed strong cytotoxicity against HeLa cell lines; cytotoxic activity is also exhibited by its 15β-hydroxy derivative [74]. 3β-Hydroxyandrosta-5,7-diene-17-one—a product of microbial degradation of 3β-methoxy-methoxy-ergosterol, is a precursor of androsta-5,7-diene-3β,17β-diol [75]. Another steroidal 5,7-diene—3β-hydroxyandrosta-5,7-dien-17β-carboxylic acid—was tested in the context of inhibition of the proliferation of normal keratinocytes and normal and malignant melanocytes. It was also effective as a condition-dependent regulator of fibroblast proliferation, finally it stimulated leukemia cell differentiation [76].