10-15-2015, 12:50 PM
Please note:
All figures and schemes referenced throughout the following text are shown in the PDF attached to this thread. Additionally, all references, which are indicated by numbers in [brackets] (sorry, I could not figure out how to make them superscripts), are given at the end of the aforementioned PDF. I genuinely hope you enjoy the read; I guarantee that you will learn something new before finishing.
Estrogens and progestogens. Estrogens are a class of steroid hormones that are characterized by an estrane skeleton made up of 18 carbons. Although they are found naturally in both sexes, estrogens are significantly more prevalent in females than males; they are the major female sex hormones and are produced primarily in the ovaries. The three major estrogens that naturally occur in women are estrone, estradiol, and estriol (Figure 2a). Estriol is released during pregnancy and is the least potent of the three. Estrone is the least prevalent of the three, being released only during menopause. Because it is released throughout the reproductive years of women, estradiol is the most prevalent of the three. Furthermore, it is also the most potent, being around 80 times more potent than estriol.[4]
Progestogens, which are named after their pro-gestational functions, are a class of steroid hormones characterized by a pregnane skeleton made up of 21 carbon atoms. Because they precede other steroids in the first step of the steroidogenic pathway, certain progestogens are the precursors to all other steroids—thus, all steroid-producing tissues must be capable of producing progestogens. The major naturally occurring progestogen is progesterone (Figure 2b). Synthetic progestogens are called progestins.[5]
Bioconversion of phytosterols to androstenedione. The first step of the synthesis of EE is converting phytosterols to androstenedione (AD), an important precursor to many steroid-based drugs. Peréz et al. tested three different soybean oil samples containing various proportions of three different phytosterols: stigmasterol, β-sitosterol, and campesterol (Figure 4). Increased concentration of β-sitosterol appears to correlate with the higher yield of AD (Figure 5). The final optimized reaction is shown in Scheme 1.[9] After the discovery of Mycobacterium, this method became widely used in industry because of its low cost and ease of transformation into AD.[10]
Mycobacterium MB3683 was chosen because it gave the highest yield of AD and the lowest yield of ADD. These cells were grown in NB medium for 48 hours, at 30°C, and with shaking at 200 rpm. The cultures were grown to 10 % (v/v), and placed in either 50 mL NB or MS media containing the phytosterols. The media were prepared with a phytosterol concentration of 1 mg mL-1. After the 5 days of incubation at the same temperature and shaking, the cultures were autoclaved and the product concentrations were tested.[9]
All figures and schemes referenced throughout the following text are shown in the PDF attached to this thread. Additionally, all references, which are indicated by numbers in [brackets] (sorry, I could not figure out how to make them superscripts), are given at the end of the aforementioned PDF. I genuinely hope you enjoy the read; I guarantee that you will learn something new before finishing.
Introduction
History of hormonal contraception. The term ‘hormonal contraception’ refers to birth control methods in which steroid hormones are the active pharmaceutical ingredients (API). There are two types of hormonal contraception: progestogen-only, in which one steroid hormone (specifically, one belonging to the progestogen class, hence the name) is used, and combined, in which two hormones are used. In 1940, the chemist Russell Marker helped develop the first progestogen-only drug by extracting the phytosterol diosgenin from barbasco, a wild Mexican yam, and converting it into progesterone. However, because progesterone is destroyed by the digestive system when taken orally, it was only available in injection form; thus, a chemical analog that could be more easily and conveniently delivered was sought. In 1951, Carl Djerassi’s lab chemically synthesized norethisterone, the first highly orally active progestin, which was eight times more potent than the progesterone synthesized by the body. Subsequently, an isomer of norethisterone was used as an API in Enovid, the first combined oral contraceptive pill (COCP), which was approved by the FDA in 1960.[1]
Steroid pharmaceutical industry. Steroids represent one of the largest sectors of the worldwide pharmaceutical industry. To date, there are over 300 approved steroid-based drugs to treat a large variety of health complications, ranging from certain cancers to central nervous system and metabolic disorders. In 2007, the global market of the steroid pharmaceutical industry was around $10 billion, and it continues to consistently increase each year. Currently, only the antibiotic sector is larger.[2]
Steroids and steroid derivatives. The primary characteristic structural feature of steroids is an arrangement of four fused rings—three cyclohexanes (rings A, B, and C) and one cyclopentane ring (ring D)—that are comprised of 17 carbon atoms bonded together. This carbon skeleton is called gonane (Figure 1a), and it forms the core of all steroid molecules. However, because the oxidation states of the rings can differ and various functional groups can be attached to the four-ring core, steroids are a diverse group of compounds with many possible derivatives.[3]
Two types of steroid derivatives are sterols and steroid hormones. Sterols, or “steroid alcohols,” are a steroid subgroup characterized by the presence of a hydroxyl group at position three of ring A (Figure 1b). They are found naturally in plants (phytosterols), animals (zoosterols), and fungi. One well known zoosterol that is vital to animal cell membrane structure is cholesterol, a precursor to numerous vitamins and steroid hormones. Steroid hormones, according to the particular receptor to which they bind, can be grouped into five distinct classes: estrogens, progestogens, androgens, glucocorticoids, and mineralocorticoids. Steroid hormones are signalling molecules that are found naturally in humans, assisting with metabolism, immune functions, and the development of sexual characteristics.[3]Estrogens and progestogens. Estrogens are a class of steroid hormones that are characterized by an estrane skeleton made up of 18 carbons. Although they are found naturally in both sexes, estrogens are significantly more prevalent in females than males; they are the major female sex hormones and are produced primarily in the ovaries. The three major estrogens that naturally occur in women are estrone, estradiol, and estriol (Figure 2a). Estriol is released during pregnancy and is the least potent of the three. Estrone is the least prevalent of the three, being released only during menopause. Because it is released throughout the reproductive years of women, estradiol is the most prevalent of the three. Furthermore, it is also the most potent, being around 80 times more potent than estriol.[4]
Progestogens, which are named after their pro-gestational functions, are a class of steroid hormones characterized by a pregnane skeleton made up of 21 carbon atoms. Because they precede other steroids in the first step of the steroidogenic pathway, certain progestogens are the precursors to all other steroids—thus, all steroid-producing tissues must be capable of producing progestogens. The major naturally occurring progestogen is progesterone (Figure 2b). Synthetic progestogens are called progestins.[5]
“The Patch.” Ethinyl estradiol (EE) and norelgestromin, an estrogen and progestogen, respectively, are the APIs in the transdermal combined contraceptive patch (matrix-type) Ortho Evra®, known simply as “the Patch”. Its primary mechanism of action is ovulation prevention, but the Patch also inhibits sperm penetration through the cervix by increasing the amount of and viscosity of cervical mucus. Covering an area of 20 cm2, the Patch is applied once-weekly for three consecutive weeks (21 days), followed by a patch-free week. Each Ortho Evra® patch contains 6 mg norelgestromin and 0.75 mg EE and delivers 150 µg norelgestromin and 20 µg EE daily to the systemic circulation for seven full days.[6]
Microbial steroid biotransformation
Because of the numerous widespread medicinal applications of steroids, pharmaceutical companies and scientists alike are continually looking for better, more efficient methods (chemical and biosynthetic) to mass-produce steroids. Microbial conversion (or transformation) is one biosynthetic method that has been used to industrially produce steroids for many decades. Steroid biotransformation is achieved through hemisynthesis that mainly starts with β-sitosterol (or diosgenin and other phytosterols) and involves a varying number of sophisticated chemical and microbial bioconversion steps.[2,7]
Advantages. One major advantage of microbial steroid conversion is that functionalization (namely hydroxylation) can be performed both regio- and stereospecifically—thus, conversions can be made at certain sites on the sterol that would otherwise be unavailable using chemical reactions. Additionally, even under relatively mild conditions, several reactions can be completed in one step which is not feasible in chemical-based approaches. Furthermore, metabolic pathways can be constructed in specific sequences in the newly generated strain. Lastly, biosynthetic pathways are, in general, more ecologically friendly than chemical syntheses.[8]
Needs and issues. The overarching need in this area of pharmaceutical research and development is cost-efficient, economical processes to produce steroids. Because many chemical reactions are economical, the number of steroid biotransformations that can compete with chemical reactions on a cost basis on an industrial scale is limited. The primary issue with microbial steroid conversion is the low aqueous solubility of steroids, resulting in poor availability of substrate to whole-cell biocatalysts. Biotransformation in organic media has been developed to help circumvent this issue; however, the high toxicity of organic solvents to cells is a major limiting factor of this approach. Other limiting factors that are of concern include: the formation of side products; yield variations due to biological variations; undesirable degradation of the steroid product by whole cells; and low selectivity of whole-cell biocatalysts due to the inhibition effect. The advantages and disadvantages of the use of biosynthesis for the production of two APIs, EE and norelgestromin, have been considered and are discussed below.[8]
Ethinyl Estradiol
Use as an API. Ethinyl estradiol belongs to the estrogen class of steroid hormones, and it is a commonly used API in various hormonal contraceptives. Coupled with a progestin, EE is an API in both COCPs (i.e.,Ortho-Cyc) and transdermal contraceptive patches (i.e., Ortho-Evra®), and its primary mechanism of action is to prevent ovulation; this is achieved by inhibiting follicle-stimulating hormone (FSH) from being released.[7] The main difference between COCPs and contraceptives patches is the varying pharmacokinetic properties of EE.
The range of daily EE dosage delivered by Ortho-Evra® (~20 µg) is comparable to that of the COCP Ortho-Cyclen® (~35 µg),[7] but drug delivery via the Patch is much more consistent over a given time period. The average steady state concentration of EE and the area under its time-vs.-concentration curve are approximately 60% higher in women using the Patch than in those using Ortho-Cyclen®; however, the peak concentration of EE is 25% lower in women who use the Patch (Figure 3[1]).[6] The adverse effects of these differences are not yet known, but increased estrogen exposure has been shown to increase the risk of certain health complications such as venous thromboembolisms. Additionally, when delivered via the Patch, EE can stay in the blood for up to ten full days, suggesting that a patient could wait up to two days to apply a new patch after removing one and still be protected.[7]
Synthesis of ethinyl estradiol. EE can be synthesized via both chemical and biosynthetic pathways. The chemical process, requires numerous reaction steps and toxic chemicals and is quite burdensome and inefficient. Conversely, a biosynthesis of EE can be performed in three relatively straightforward steps that are explained below. The first two steps of this synthesis are carried out using biological catalysts, and the final step is a straightforward chemical reaction.Bioconversion of phytosterols to androstenedione. The first step of the synthesis of EE is converting phytosterols to androstenedione (AD), an important precursor to many steroid-based drugs. Peréz et al. tested three different soybean oil samples containing various proportions of three different phytosterols: stigmasterol, β-sitosterol, and campesterol (Figure 4). Increased concentration of β-sitosterol appears to correlate with the higher yield of AD (Figure 5). The final optimized reaction is shown in Scheme 1.[9] After the discovery of Mycobacterium, this method became widely used in industry because of its low cost and ease of transformation into AD.[10]
Mycobacterium MB3683 was chosen because it gave the highest yield of AD and the lowest yield of ADD. These cells were grown in NB medium for 48 hours, at 30°C, and with shaking at 200 rpm. The cultures were grown to 10 % (v/v), and placed in either 50 mL NB or MS media containing the phytosterols. The media were prepared with a phytosterol concentration of 1 mg mL-1. After the 5 days of incubation at the same temperature and shaking, the cultures were autoclaved and the product concentrations were tested.[9]
As shown in Scheme 1, the yield of AD was 65%, while the yield of the side product was only 2%. Based on the data presented in Figure 5 for yield of AD from the various soybean oils tested, it appears that the β-sitosterol concentration is weakly correlated to higher AD yield. VN-3 had a lower concentration of β-sitosterol and resulted in a lower percent yield of AD as compared to VN-1. The paper suggests that this could be due to a better bioaccessibility to substrates of mycobacterial cells or that a stigmasterol regulating mechanism in the enzyme 1,2 steroid dehydrogenase could be at work.[9]
Malaviya et al.[10] present a mechanism for the biotransformation of β-sitosterol to AD by the Mycobacterium. The side chain cleavage, which is the main step, requires the regeneration of cofactors such as NAD+ and FAD. This process starts by hydroxylating the C27, which is then oxidized to a carbonyl group. Then the C28 is carboxylated.[10] The numbering convention can be seen in Figure 6. The presence of sitosterol helps to induce the two enzymatic reactions. The first of these is catalyzed by three enzymes, while the second is dependent on the dissolved CO2 concentration. In Mycobacterium, side chain cleavage of β-sitosterol can be induced by propionate or by propinyl-SCoA. Dissolved CO2 (1%) affects the yield of AD positively, possibly because excess aeration changes the way the cell metabolizes the substrate. Through the cleavage of one sitosterol, 3 molecules of propionyl-SCoA, 3 molecules of FADH2 , 3 molecules of NADH, and one molecule of acetic acid were created. These products can then be used for the production of ATP. If the sitosterol is broken down completely, 18 molecules of NADH and 7 molecules of FADH2 are created, which means 80 molecules of ATP can be produced from one molecule of β-sitosterol. This presents a challenge for the biological catalysis of this cleavage because it is energetically favorable for the cell to entirely break down the molecule.[10]
The chemical synthesis goes through multiple reaction steps making it unfavorable when compared to the biosynthetic step. The main challenges are cleaving the side chain of the C17 and dealing with the very sensitive steroid ring structure. This, combined with the harmful and toxic reagents like pyridine, means a very lengthy, costly, and low yield, making the biosynthetic route the more adequate method.[10] However, there are still problems with the biosynthetic pathway and areas for improvement, for example, the problems with the conversion of sitosterols to AD are the degradation of the steroid nucleus and inhibition of the side chain degradation by the reaction products.
Some potential solutions proposed by Malaviya et al.[10] include inhibiting the 9-α-hydroxylase and screening for the improvement of the microorganism to give a higher yield of AD. In order to maintain the steroid nucleus of AD, the action of 9-α-hydroxylase and 3-ketosteroid-1(2)-dehydrogenase must be blocked. The activation of 3-ketosteroid-1(2)-dehydrogenase triggers the formation of a double bond between the C1 and the C2, leading to the formation of ADD. The activation of 9-α-hydroxylase leads to the addition of a hydroxyl group on C9 forming 9-α-hydroxy-4-androstene-3,17-dione. Inhibition of these enzymes would help to boost yields, but can be lethal to the cells. 9-α-hydroxylase is a monooxygenase that is important to the electron transport chain and contains proteins that require Fe2+; therefore, a good way to inhibit this enzyme is to chelate Fe2+, using a chemical like 8-hydroxyquinolone. Also in the commercial arena, scientist are trying to screen and improve the bacteria used. This improvement can be achieved by developing strains which are less sensitive to phytosterol toxicity or by mutagenic treatment that increases the efficiency with which the bacteria cleaves the side chain.[10]
There is also the major problem of low solubility of the phytosterols in aqueous media, which is the factor that makes this the bottleneck of EE production. Malaviya et al. [10] propose five major ways to get around this problem. These included: biotransformation in two-phase systems, bio-transformation in cloud point systems, biotransformation using immobilized biocatalysts, and microemulsions and liposomes as an alternative biotransformation system. However, these methods have proven inadequate for application industry as of 2008.[10] While research is being conducted to find solutions to the issues with biosynthesis, this process is used in industry because it is a better way to make AD than the chemical method.
Androstenedione (AD) to estrone. The next step of the synthesis is the conversion of the AD formed in the previous section to estrone (Scheme 2). To accomplish this, the A-ring has to be aromatized, the C19 has to be removed, and the carbonyl group at position 3 has to be oxidized. This step currently is not done biosynthetically in industry, but we propose a biosynthetic method to form estrone biosynthetically using P450arom, a human aromatase.
Currently, the P450arom-expression plasmids have to be constructed. Kagawa et al. [11] describe a method to accomplish this. In brief, the GroES/GroEL expression plasmid pGro12 was obtained from an outside source and the gene of interest was spliced in using site directed mutagenesis. Kagawa et al. also describe a process for the preparation of P450arom, it is summarized below. In order to prepare the P450arom, E.Coli DH5α cells with the expression plasmids were incubated overnight in 5 ml TB with 100 µg mL-1 ampicillin at 37°C. These cultures (2 mL) were diluted into 250 mL TB media in 3 L culture flask and incubated for 4 hours at 37°C.[11] IPTG, gamma-aminolevulinic acid, and arabinose for induction of molecular chaperones GroES/GroEL were added to the culture. Then, the culture was incubated for 28 hours at 28°C.[11] Cells were harvested by centrifugation and lysed, then centrifuged in order to separate supernatant. This solution was purified to extract the P450arom. The yield for this reaction was 13.4 nmol P450 (mg protein)-1.[11]
This P450arom was then added directly to the AD solution. First, the P450arom would oxidize C19, and then activate the concerted elimination of C19. The C19 is eliminated as formic acid, and the 1-β and 2-β hydrogen atoms are eliminated from the A-ring.[11] The third step that oxidizes the carboxyl group is unclear, but it results in an estrone. We propose doing this step of the synthesis in a fermentation setup. While this change would introduce new issues, such as diffusion and solubility limitations, it could potentially decrease both capital and operating costs by eliminating the protein purification steps. However, both methods would need to be analyzed to ensure that the most economical process is used.
Estrone to ethinyl estradiol. The ethinylation process is a very old one that started in the 1930s.[12] This process includes adding ethyne, sodium, and sodium amide to the estrone (Scheme 3). The ethyne will attack the carbonyl carbon causing the oxidation of the carboxyl group forming a hydroxyl group. This leaves us with the final product, EEl. This process is necessary to change the bioavailability of the estradiol.[13]
Norelgestromin
The progestin norelgestromin (Figure 7) is the second API of Ortho Evra®. It is the progestational component of the patch and acts by preventing the release of luteinizing hormone, which is associated with the initiation of ovulation. The molecule is derived from norgestrel, another progestin, and is the active metabolite of norgestimate.[14] In a study by Abrams et al. [7], administration of norelgestromin through a patch was shown to maintain more consistent levels of drug in the body, which helps to ensure the drug is at a therapeutically effective concentration throughout the administration period. The patch was designed to deliver 150 µg day-1, compared to the pill’s 250 µg day-1. The average concentration achieved via oral delivery was found to be 0.75±0.23 ng mL-1, while the steady state concentration delivered by the patch was 0.83±0.21 ng mL-1. The area under the curve was comparable for the two delivery methods as well.[7]
Norelgestromin presents significant challenges for biosynthesis because of the variations between the API and naturally occurring progestogens. The presence of an ethyl group at C13 of the molecule makes this synthesis particularly difficult because no naturally occurring progestogen has this group,[5] and selective methylation of C18 (methyl group on C13) cannot be carried out biologically. Similarly, the addition of the oxime group at C3 and the ethinyl group at C17 must be be done chemically. As such, norelgestromin is not a good candidate for a microbial based synthesis; however, a synthesis involving a small biological component can be done.
Synthesis of norgestrel. The synthesis of norelgestromin begins with the synthesis of norgestrel, the key intermediate in the synthesis. The proposed synthesis was adapted from Gibbian et al. [14] and Kleeman et al. [13], and is shown in Scheme 4. Beginning with a two-ring structure, 6-methoxy-1-tetralone, the first step is the addition of a vinyl group via a Grignard reaction, which must be carried out in an organic solvent such as THF or diethyl ether. Following this, a Michael addition is carried out with Product I and 2-ethyl-1,3-cyclopentadione in the presence of a base. This step simultaneously performs a base-catalyzed hydrolysis resulting in Product II, which has three of the four rings of the steroid backbone formed.
The next step is the only biologically active step of the synthesis of norelgestromin. Product II is added to a culture of yeast, Saccharomyces uvarum, to stereospecifically reduce one of the ketone functional groups on what will become ring D (see Product III). The enzyme responsible is a keto reductase, and as such the media may need to be doped with NADP+.[15] The reaction has an overall yield of 44-52%.[14] One method of boosting this yield would be isolating the enzyme and performing the reaction independent of a fermentation process. This would help to rid the system of some of the mass transport limitations that exist in whole-cell catalyzed processes.[13,14]
Following the biosynthetic step, acetic anhydride and a strong acid, in this case toluenesulfonic acid, are used to protect the remaining hydroxyl group on ring D and to hydrolyze the ketone group, allowing ring C to form (see Product IV). Unnecessary double bonds are then saturated in the presence of hydrogen and a palladium-carbon catalyst, followed by potassium hydroxide, methanol, lithium, ammonia, and aniline. This step also unprotects the hydroxyl group on ring D and leaves two unsaturated bonds on ring A (see Product V). The addition of the ethinyl group and the oxidation of the ester to a ketone are then carried out to give norgestrel. This is done by addition of a hindered base (Al(O-iPr)3) in MEK, followed by reaction with lithium acetylide, and, finally, addition of a strong base.
Overall, this synthesis consists of fairly simple organic reactions. Again, this is because of the ethyl group at C13. This group is not naturally occurring, so the use of a natural precursor is not feasible for this synthesis. Therefore, a total synthesis beginning with commercially available reagents is necessary. The use of S. uvarum helps in providing in enantiomerically pure product, but does not shorten the overall synthesis as in the synthesis of EE. With this in mind, a biological synthesis of norgestrel does not provide large advantages over a traditional chemical route.
Synthesis of norelgestromin. The norgestrel from the previous synthesis can be transformed into norelgestromin via a three-step synthesis. The synthesis described was developed in the patent by Tuba et al. [16], and is shown in Scheme 5. The only modification that needs to be completed is the addition of the oxime group at C3. In order to ensure that the group is only added to the appropriate location on the molecule, the hydroxyl group must be protected.
This protection is again accomplished with acetic anhydride and a strong acid, in this case hydrochloric acid. Under nitrogen, a suspension of norgestrel with acetic acid (100 mL), acetic anhydride (6 mL), zinc chloride (2 g), and 6.7% hydrochloric acid in acetic acid (1.6 mL) is stirred for 20 min, then water (5 mL) is added to the mixture. After stirring for an additional 15 min, 18% aqueous hydrochloric acid (3 mL) is added, and the mixture is stirred for another 45 min. At this point, the reaction is complete, and a crude product is isolated by adding ice water (600 mL), filtering, washing with water, and drying. The total reaction time was about 80 minutes.[16]
Following this step, the intermediate, norgestrel acetate, must be purified. This is accomplished by dissolving the product in dichloromethane (100 mL) and stirring with silica gel (10 g). The mixture is stirred for 30 min, and the gel is then filtered off. The solvent is then boiled off. The remaining product is refluxed in isopropyl ether:acetonitrile (9:1 v/v ratio) mixture for 15 min. The mixture is then cooled in ice water to 0°C, precipitating the product, which is then filtered and dried. The mother liquor can be reprocessed using the same purification method to obtain more product. This step has a yield of 15.4 g (90.5%).[1]
The pure product can then undergo oximation to form norgestimate, a reaction that is again carried out under nitrogen. Hydroxylammonium chloride (76 g) is added to a stirred solution of norgestrel acetate (120 g), acetic acid (1200 mL), and anhydrous sodium acetate (90.2 g). The reaction takes 1 hr and must be maintained under 30°C. The resulting mixture is added to water (10 L) and stirred for 30 min, after which the crude precipitated product is filtered off and dried at 40°C.[16]
The crude product is then dissolved in boiling ethanol (2500 mL), clarified with charcoal (12 g), and filtered. The filtrate is concentrated under partial vacuum (below 40°C) to 400 mL, then cooled to 0°C for 3 hr to precipitate the product. The product is then filtered off and washed with two portions of ethanol (125 mL each), and dried under 40°C. The yield of norgestimate was 102 g (81.6%) with a purity of over 99.5%.[16]
The final step of the synthesis is unprotection of the hydroxyl group on ring D. Again under nitrogen, norgestimate (10 g), methanol (100 mL), and sodium hydroxide (3.25 g) are stirred together at 22°C. After approximately 10 min, the temperature rises by about 10°C and a homogeneous mixture is obtained. The mixture is then stirred for 3 hr at 25°C, and added to chilled water (1000 mL). The pH is adjusted with acetic acid (3mL) to 7-7.5, and the suspension is stirred for 20 min more. The product is then filtered, washed with water, and dried over phosphorous pentoxide at 40°C under partial vacuum. The yield of norgestimate was 8.4 g (94.8%) with a purity of 99.9%.[16]
The overall yield for this synthesis was 70.0%, which seems acceptable. However, the amount of solvent used for each step is quite large, especially when considering scale up of the process. Before a scale up is proposed, a pilot scale operation meant to optimize the solvent use would be prudent to minimize operating and capital cost for a commercial process. Some optimization can also be done when incorporating the two processes. For example, the hydroxyl group on ring D is protected and unprotected twice, so protecting the group early on and leaving it protected until the final step of the synthesis.
Challenges of norelgestromin biosynthesis. Norelgestromin is not a good candidate for microbial synthesis, unlike EE, because of the functional groups circled in Figure 8. The ethyl group is again the most challenging of these groups because it is not naturally occurring, and both selective methylation and ethylation are difficult to accomplish. Thus, using a common natural precursor such as AD for both of the syntheses is not practical. The other two problematic reactions are the oximation and the ethinylation, both of which must be done chemically. Thus, the majority of the functional groups on norelgestromin would have to be added chemically, and the backbone itself has to be made with a total synthesis to ensure the presence of the ethyl group.
Conclusions
Microbial synthesis provides a myriad of benefits in the manufacture of steroid molecules. While it has limitations caused by mass transport, solubility, etc., microbes are able to perform complicated chemistry, which would often require multiple chemical steps, in one process. A very good example of this is the microbial synthesis of AD from phytosterols, where the microbe removes all of the unnecessary sidechains and leaves the target intermediate product in a single fermentation process. The fact that the biosynthesis of estrone can be done in two steps and that forming EE only requires one further reactions, makes it an ideal candidate for biosynthesis.
Microbial synthesis does however have severe limitations when the target molecule does not resemble natural compounds well enough as shown with the synthesis of norelgestromin. The presence of the ethyl group at C13 severely limits the potential for biosynthesis of norelgestromin. Because of this group, the backbone need to be totally synthesized from raw materials as we have shown.
Future work in microbial synthesis includes strain improvement,[10] continued work with two-phase reactor systems, development of green solvent systems, and development of novel hydrophobic delivery methods.[8] These goals are driven by the need to continue improving the productivity of various microbial strains in order to make them more competitive compared to traditional chemical syntheses.